November 6, 2024 • 58 mins
Dive into an enlightening conversation with Allen, Adi together and special guest Bruce Tate as he shares his passion for teaching underrepresented communities in tech. In this episode, Bruce discusses the transformative power of mentorship through Alexa Chat, a weekly volunteer-driven session aimed at fostering inclusivity and learning among aspiring coders. Discover how creating a safe space can lead to profound personal and professional growth.
Key takeaways from this episode:
- The importance of learning by doing: Encouraging new programmers to code actively.
- Building community support: Creating a vulnerable yet respectful environment where mistakes are welcomed.
- Inspiring success stories: Real-life examples of individuals thriving through mentorship.
Join them for an inspiring discussion that highlights the impact of inclusive education in technology.
Socials
Become a supporter of this podcast: https://www.spreaker.com/podcast/elixir-mix--6102049/support.
Key takeaways from this episode:
- The importance of learning by doing: Encouraging new programmers to code actively.
- Building community support: Creating a vulnerable yet respectful environment where mistakes are welcomed.
- Inspiring success stories: Real-life examples of individuals thriving through mentorship.
Join them for an inspiring discussion that highlights the impact of inclusive education in technology.
Socials
Become a supporter of this podcast: https://www.spreaker.com/podcast/elixir-mix--6102049/support.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:06):
Hey, everyone, and welcome to another episode of Alexa Mix.
Today on the panel we have Alan Wima, Hello, and
we have a really special guest today and Bruce Taye.
Speaker 2 (00:15):
Hey, Bruce, Hi, everyone from Jed Nuga, Tennessee. I don't
get to say that on this podcast pretty often.
Speaker 3 (00:22):
Yeah, it's great to have Bruce back.
Speaker 1 (00:24):
We've had Bruce a couple of times in this podcast,
and he was one of the hosts of the podcast
a few years ago. So uh yeah, Bruce, your name
will always be kind of tied.
Speaker 3 (00:33):
To Alexa Mix in somewhere somewhere or another. But yeah, I.
Speaker 1 (00:36):
Guess like today we had Bruce to talk about a
few different topics.
Speaker 3 (00:39):
Bruce has a new book coming out. We'll touch on that.
Speaker 1 (00:42):
But Bruce also mentioned before you start recording, and every
time we have Bruce, we love to touch on the.
Speaker 3 (00:47):
Teaching aspect of things.
Speaker 1 (00:49):
Bruce does a lot of courses on groxio and there's
also a volunteer Mentorship slash Mentee session that happens once
a week. Alexa Chad, I would love to actually talk
about that first, If that's okay, Bruce, I would love
to Awesome, Awesome, Well why don't you why don't you
tell our listeners what Alexa Chad's about. I'm a huge
(01:10):
fan of that, and listeners will understand why in a bit.
Speaker 3 (01:14):
But yeah, in your own words, what is Alexa chat.
Speaker 2 (01:16):
Yeah, Audi is just a little bit humble because he's
one of our mentors and he's been a tremendous influence
in our attendees' lives. But this is a program to
teach underrepresented people Elixir, and we don't try to be
to lean really heavily into the underrepresented part. The idea
(01:37):
is you can be underrepresented in technology and you get
to define what that is when you join, or you
get to bring someone who is. So we're not exclusive,
but we do want the Elixir world to look a
lot like the rest of the world. And we're about
three things. The first thing is we believe that programmers
learn to code by coding. The second thing is that
(01:59):
we believe making mistakes in front of other people. So
if you're going to learn by coding, and if you're
going to drive in language you don't understand making mistakes
is a vulnerable thing. So the third thing is that
you have to lead with love and respect, and so
we spend more time getting to know each other than
you would otherwise expect. So this is not unusual for
(02:21):
us to shut down like the technical aspects of an
Elixir Chat program and then talk about somebody that gets
laid off or has a big interview coming up and
kind of rally around them and support them. So that's
what we're about. Coders learned to code by coding, making
mistakes a vulnerable thing, so you have to lead with love.
Speaker 1 (02:40):
Yeah, that's really amazing and I think that really kind
of sums up my experience being a mentor and alex
a chat as well. There is like a different side
of people that comes out when you build that comfort level,
like focusing on like making sure we create like a
comfortable slash safe space for them to talk to thrings
knowing each other more, They're able to ask questions more openly.
(03:02):
It's something that people look forward to every week. I've
noticed that, Like I haven't been to Alexa Chat in
a few months because of my work schedule, but it's
something I'm very eager to get back to as a mentor,
and I know as mentees there are others as well.
I want to give like a quick short testimonial. My
wife used to be one of the mentees the Alexa
Chat and Thin two or three years ago, and she
(03:23):
still talks about like how much she learned from like
she was definite, like six months, Like she like half
a year, but that six months of learning once a
week just over the course of years, that like really
contributes to a significant part of how much she has learned,
just because of the variety of ideas that is there
in that session. But yet it's really amazing. And I
already know the answer to this a little bit, but
(03:44):
I still want to touch on those cruise like what
inspired you to start Alexa Chat.
Speaker 2 (03:49):
Yeah, so this is a little bit of a dark story.
I think I was at a programming conference somewhere else,
and I'm not not going to say where that is,
and I'm not going to say which conference, but there
were four hundred and three attendees, and four hundred of
them were white males, and not that gosh, I'm a
white male, and I don't think that there's a problem
(04:10):
with that, but it didn't look like the rest of
the world, and that troubled me because this was supposed
to be a headlights type conference, a conference that that
kind of set the tone for where programming was going,
and if this was where programming was going, I had
a little bit of a problem with it. But I
didn't want to do a program that was super that
was hyper aggressive and political. So we came up with
(04:32):
this thing of define underrepresentative whatever way you want to
and either be underrepresentat or bring someone who is. And yeah, yeah,
when we did that, we found that when we engaged
with people and when we engaged around their own projects
and their own desires and we just stayed with them
(04:53):
and loved them until they were able to get some
traction in the community, good things happened. So there's a
there are a couple of success stories. I just talked
to Flora who gave a talk at code Beam Berlin
and Flora if you're listening high. So Flora was had
a marketing career and came to an Elixer chat and
(05:15):
she had a newborn baby and we had One of
the things that we do is we put on a
conference and the conference is his Gig City Elixer and
it's something that we put on in Chattanooga. It's really
about this program and if people don't understand that, the
idea is to put some of the best programmers in
the world. Sit them at the same table as our
(05:36):
mentees and start buyers, right see what happens. And so
we invited her to come. We said, your ticket is
free and all you have to do is get here.
We'll find a place for you to stay and everything.
And we did, and and she came, and she brought
the baby, and she was loved so hard through that whole,
that whole experience, and she met her eventual employer HDA
(05:58):
and Brian Hunter and that fabulous team hired her and
that relationship is an absolutely fantastic one. She's giving conference talks,
she's growing ways that nobody expected, and she's reading documentation
and reading code and finding bugs and code just from
being in the code base right just a year and
(06:19):
a half into doing this, and it's it's really a
fabulous thing. And that's that's kind of what we do.
It's not a rapid progress, but it's an effective way
to kind of share connections and experience with the people
in our community. And it's showing up on the stages
and the conferences that that's nothing could be more exciting
to me.
Speaker 4 (06:38):
The Elextrid I tried doing some Google searching and I
couldn't find any information about it, and I spawned it wrong.
Speaker 2 (06:43):
It's kind of it's kind of stealthy. So it's normally
invitationally we're casting a waternet.
Speaker 4 (06:51):
Today's okay, I know where I'm not wanted, you know.
Speaker 2 (06:59):
Yeah, well, well it's so it's it's a little bit delicate,
right because if you have to have enough mentors to
love the people that come, and if you don't, then
the people notice and you know, so that there's there's
a balance, right. So but if you want to join,
you can reach out to roxio Learning on x or
you could send an email to info at rox dot
(07:22):
io or to Maggie at rox dot io and she'll
she'll get you added to the program.
Speaker 3 (07:28):
Yeah, high, highly recommend it.
Speaker 1 (07:30):
I think I think one of the best parts about
joining that group on a weekly basis. I joined as
a mentor directly, but I think I obviously when you teach,
you learn yourself.
Speaker 3 (07:41):
That's the thing as well.
Speaker 1 (07:42):
But I think one of the things I realized, specifically
from Bruce and some of the other mentors there who
had more experience, like how to go about mentoring people
who are newer to coding as a whole and don't
help me at my work too as a senior and
a staff engineer to mentor others, Like, there is a
specific way to approach how to teach someone coding, and
(08:04):
obviously like taking that empathetic slash. Las Bruce says love approach,
like making making them feel comfortable and first before teaching
them something. It's seeing the effects of that really changed
the way I approach mentorship at work as well. So yeah,
highly recommend it for anyone who's in that underrepresentered category.
Speaker 3 (08:21):
Obviously to reach out to groxio. Yeah, it's a great group.
Speaker 1 (08:26):
I've done just like very not smoothly search our topics
to just groxio as a whole. Yeah, I guess pring
on proxio. I'm sure everyone has some idea what groxios
it's a learning website. How would you pitch roxio as
a whole? What is groxio about? Maybe some some of
your most popular courses on croxio.
Speaker 2 (08:46):
Yeah, So we do three different things. The first is
that we do we do trainings. Right now, our public
trainings are sold out for a bit, but we are
taking slots for commercial commercial slots. So if you have
a team that you'd like to train, we'd love to
talk to you. We also do video courses, and the
video courses are on a wide variety of topics. Our
(09:08):
most popular three are Phoenix Live View and of course
we'll be updating the courses shortly when live v one
dot O comes out. There's the OTP course, and there's
also an Elixir course, and there's a new emphasis on
groxio with a live book platform, and we have some
projects that kind of walk you through projects projects there.
(09:29):
And so what's interesting is that we have a Black
Friday sale. We have one massive sale that happens every
year and it is a forty percent off sale, and
so that's a site subscription forty percent off and that's
all the way through the last day of October, and
that'll be passed by the time we publish here, but
all the way through November. So hear this. We don't
(09:51):
do this often, but you get your forty percent off
at groxio just just by hanging out and coming.
Speaker 3 (09:57):
Yeah, another lot of things. Highly recommend it.
Speaker 1 (10:00):
I have the yearly subscription through one of the hackathons
prices and I've been it's been on my list to
check out the Nerves course. I see that some of
them were updated a little bit ago too, so that's
pretty cool.
Speaker 3 (10:13):
Yeah, I'm seeing some live book courses like you mentioned.
It's it's quite exciting.
Speaker 1 (10:17):
To see that being used on your on your website.
What are some of your most popular courses? I see
the Elixir one O TP live view. There's other languages too, right,
this Pony prologue. What are the codes that people usually
tend to just like go go for usually.
Speaker 2 (10:32):
Yeah, that's a that's kind of a little bit of
an embarrassing question. So this was intended to be a
seven languages in seven weeks. So the idea was just
going to be a programming passport for the world, right,
and you learn to code and then hopefully bring people
back through the licture by exposure the other languages. And
it was just too much, right, So learning a language
(10:53):
a month, I couldn't keep up the pace and it
about wreck me trying. So we we boiled the content
down to just mostly the Elixir. The prologue stuff is
probably still pretty good. But all those other courses are free,
and they're not really full courses. They're just like four
four videos and then a mini book. The idea is
that different learning methods. Are you open learning channels by
(11:16):
different learning methods? But yeah, mostly we're in an Elixer site,
got it? So Paulo, So, I don't know if everybody knows,
but I guess people know by now. But Paulo Vallin,
who is Joseph's brother, is working with us, and I
don't know if people know this though. He came from
a YouTube channel where he taught chemistry in Brazil. And
take a guess. I want to get a guest from
(11:38):
both of you, Audi and Allen how many views his
channels had in Brazil.
Speaker 1 (11:43):
I think I remember you mentioning it sometime, and yeah,
I think might be last time he came out. I
would say half, not so much, maybe maybe one hundred million.
Speaker 2 (11:53):
Sixty million, Yes, sixty million views in Brazil. Very good? Yeah,
great memory. But but anyway, he we brought Paulo on
because he really has a good understanding of how to
organize learning and learning content. And so since he came on,
we've we've kind of taken this haphazard for videos and
(12:13):
a microbook, and we we kind of took that down
to like a focused agenda with objectives and and we
kind of blew out a full course for the Elixir
Live View OTP, and we're kind of pushing that into
into the other the other Elixir topics as well. But
the OTP, I think is the big one that we're
kind of famous for, both the Elixir and OTP.
Speaker 4 (12:36):
I'm still kind of curious about Sorry to kind of sidetrack,
but about the Live View book because that's been in
writing for.
Speaker 2 (12:43):
For part some times.
Speaker 4 (12:46):
I understand why because I say something very sensitive, it sounds.
Speaker 2 (12:50):
Like, no, it's just that it's that. So it is
a we knew it was a big undertaking when we
took it on, and it has been a massive undertaking.
Speaker 1 (12:59):
Right.
Speaker 2 (12:59):
We've written the book about six times, right, And it's.
Speaker 4 (13:04):
Also because there they changed the APIs and everything right
was and I almost feel like maybe they did it
just to play with you, because it was pretty stable
for quite some time, right and then all right.
Speaker 2 (13:13):
Right, right, And I know our timing has not been good.
Our timing has not been good. Right, So so anyway,
there's say there's a beta out now that covers uh,
that covers Live. But what's what's happening with Live now
is that we're going to the new generating code is
going to I believe remove core components. Is that the
(13:35):
way it works. I think that's right. So if you
go into the main branch of Phoenix, you don't see
these modiles and these other components like that. It's it's
going to the generator is going to work a little
bit differently, and that obviously isn't going to change most
applications very much, right because you generate one time and
(13:55):
then you're good. It's going to hit us really really
hard because you know, we lean on like walking through
this generated code and showing this as the idiomatic live
you right. So yeah, but once liveew one dot zho
is released, we'll take a take a fool rewrite. It'll
go pretty quickly because we have the system down. And
let me give a shout out to the Phoenix team.
(14:16):
So all of these rewrites mean that the team has
been working incredibly hard for the people in the community,
and we were really lucky that there are so many
professionals that will dedicate so much of this their careers
to supporting this kind of new model of building well
I say new model, but a communal model of building software.
(14:38):
So I believe that open source software is like one
of these last bastions of hope right where people work
together for a common good. And so shout out to
the Phoenix team. Also a shout out to Sophie de Benedetto.
I had started taking on this project myself, and I
could not have done this myself. She's been I could
not have asked for a better writing partner, somebody that's
(15:01):
more intelligent, someone who's more gifted. And in fact, she
did such a good job with this. As I moved
out of my editorial role for the Pragmatic Bookshelf, she
was the one that I hand picked to fill that role.
Speaker 4 (15:13):
I haven't checked the recent edits to the book yet.
I started trying to look into it, but I haven't
gone back to it. But now it seems like we
have like basically a divergence. You have O dot twenty
and then you have one dot O r C. Is
your book doing one dot I'm guessing because and also
there's some differences but not like major. But at the
same time there is some differences that yes, yeah, it's.
Speaker 2 (15:33):
Kind of the one the one r C is there's
there's going to be a one more major divergence after
that that we haven't picked up on yet and that
hasn't been written yet, so.
Speaker 4 (15:45):
Kind of Also some breaking news and stuff that just
came to my mind right now. Did you see the
recent news that the sorrent I don't remember his name
now this is real name, but I think he comes
on my Soren too, and then that they believe is
products point one. Yeah, Parker exactly. Yeah, par he just
announced that he's doing open web open source. Now, yeah,
that's pretty cool.
Speaker 3 (16:06):
I would be.
Speaker 4 (16:06):
Interested to see if you can take a look with whenever.
I don't know when it's going to be released. It
was kind of talked about on the I think it's
alixir mentor YouTuber podcast video podcast. They mentioned it and
a lot of people were saying, oh, it's a really
really great live implementation. So I'm kind of curious if
you're going to hold off on the book and say,
take a look at that and see could be some
more interesting patterns or something that would be nice to
(16:27):
have in the book. Did you, well, now you know
about it, do you think that before right.
Speaker 2 (16:32):
I don't know, I don't know. I'll have to take
a look. Yeah, no commitment, sir.
Speaker 4 (16:37):
Yeah, I'm also kind of curious to see how he
implemented it, because I remember the first came out, I
was I got to like kind of beta trial it,
and I was pretty impressed with how everything kind of flowed.
The stuff you can do with live view is really
amazing for for everything and CSS however far that's advanced
since I mean you were all older programmers, you know.
I showed somebody, you know, how you turn an integer
(16:59):
and to a string with plus empty string, and they
were like, what are you talking kind of how weird
is that? I said that was JavaScript back then, what
are you talking about?
Speaker 3 (17:07):
Right?
Speaker 2 (17:08):
Right?
Speaker 4 (17:09):
CSS? Now it's like I barely ever really need to
write JavaScript, and it's been wonderful.
Speaker 2 (17:14):
Yeah, it's pretty amazing. And I mean that's that's been
one of the evolutions that the Phoenix teams. So when
you when you park so much infrastructure on top of
the web, which moves quickly anyway, and then you're sitting
on top of an emerging language like Elixer, it's a
pretty exciting thing, but it's it could be pretty demanding
for a team also, And one of the things I
(17:35):
wanted to talk about today was this idea that Elixer
is a language that you can't put a box around.
Speaker 1 (17:41):
Right.
Speaker 2 (17:42):
It's like when when I started working with Elixer, I
could point to two main gaping holes in it, right,
and I could say with certainty that those holes would
never be filled.
Speaker 3 (17:54):
Right.
Speaker 2 (17:54):
The first one was this this whole idea of immutability,
but particularly how it relates to machine learning and machine
learning models, right, Jose kind of he kind of said, well, well,
what if I swapped out the compiler and swapped in
this other piece, and you know, we did just in time,
(18:15):
and they kind of wrap this with the extensions of
the language, the protocols, and so I said, well, yeah,
you could do that, and he did that and was
stunningly successful. And the other thing was to go from
a strictly dynamically typed language to something that is gradually
typed now, and I thought, how do you do that?
(18:35):
How do you do that? And I wonder if what
we're seeing with the Elixir type system, I wonder where
types are going to go, right, because there's this tension
between what can you express with the type system and
what do you have to express with the type system?
Speaker 3 (18:51):
Right.
Speaker 2 (18:51):
One is is kind of how do you make developers
more productive? And one is how do you make the
how do you enabled the computer to make sure that
these these solutions are right?
Speaker 1 (19:04):
Right?
Speaker 2 (19:05):
So my head is exploding.
Speaker 1 (19:07):
Yeah, I think I think that's yeah, that's a good
way to put it, like the correctness versus productivity, Like
it's yeah, they're.
Speaker 3 (19:13):
At odds for sure.
Speaker 1 (19:14):
I think Alexa was a good like you kind of
like alluded to. It's a good middle ground right now,
it's going maybe going more towards a bit more on
the correctness side. But yeah, it's competing priorities for sure.
Speaker 2 (19:25):
So there's another element of this that I think is interesting,
and it's so I don't think that you could do that.
You could you could give any late leeway at all
in terms of the type system if you didn't have
OTP underneath you.
Speaker 3 (19:37):
Right.
Speaker 2 (19:37):
So I talk about this conversation a lot that I
overheard between Joe Armstrong and a man named David Turner
who was the creator of a language called Miranda that
was one of the first lazy languages. And so David
Turner so at this party and I think it was
code Beam in London. He said he said a word Joe,
and then Joe kind of walked over to him and
(19:59):
he said, I, I didn't think that you would be
able to achieve the reliability that you did with a
dynamic type system in Earlin, right, And it really is
pretty stunning that this whole idea of all this math
on one side and lamb at calculus and this tons
and tones of math making sure that things are correct,
(20:19):
and then on the other side this idea of did
you try turning it off and on again? Dude? Right,
But that's that's effectively what we had. And but that
system does work. The idea that you can reset to
a known, good starting state is extraordinarily powerful. And maybe
when you pair that with a type system that is
(20:41):
gradual but is enough but enables the programmer, maybe that
is a stunningly powerful combination. That's the way that we
want to build business applications. I wouldn't bet against.
Speaker 4 (20:52):
I also want to kind of lean in to hear it.
So I've been hosting a monthly Earlining Elixir focused meetup
and it's Asia based and I've been lucky enough because
of this podcast, and I also went to code being
Berlin and I got to meet a lot of people.
So I just been saying, Hey, can you do you
mind to come present somebody who had got to present
at the very beginning was Robert, one of the careers
of erline.
Speaker 2 (21:11):
Yeah, and it was.
Speaker 4 (21:12):
He gave a really great presentation and then at the
end there's a Q and a section and I think
I was the one who asked, hey, you know, why
is it that Erlang didn't have, you know, a strong
type system? And he said yeah, because actually in the
presentation he actually said people said we you know, we
couldn't do it, but actually we could.
Speaker 2 (21:27):
We just didn't.
Speaker 4 (21:28):
We just didn't want to do it. And he kind
of like said something like that and kind of skipped
over it. So then at the end, like I said,
I asked him, I could you expand that some more?
He said, well, basically, when you do hot code reload,
which we needed, things got a little bit weird with
with strong types. That's why we decided to remove it.
And okay, it makes sense. So now with this gradual
typing system, I'm curious what will happen now, Right, will
we have issues?
Speaker 3 (21:48):
Right?
Speaker 2 (21:49):
I think that probably not. I think that probably what
you'll see is that the gradual typing system will take
it as far as it can. It will basically push
the type system as far as math will go, right
and no further, And so that might take us out
some of these like highly theoretical like the monatic thing.
And I think it's really more than just type systems, right,
(22:12):
It's it's the idea that we in the computer science
field are kind of married to this idea of functional programming.
And there are very few fully functional functional languages, right,
and I think that maybe we need to get away
from that. And in fact, if you look at the
Erlang and Thelix or ecosystem, if you look at there's
a there's a state machine that you can build into
(22:34):
a process, right that a gen server can be implemented
as a state machine, right. And it's the message passing
that gives you the installation, the message passing to this
stateful entity that that gives you the protection and your
own design that's around that, right. So I said that
kind of poorly, right, So I guess what I mean
(22:57):
to say is that functional programming and immutability are all
just kind of there. There's shades of grade in there,
and there are some pretty hard and aggressive collisions between
between programmer productivity and correctness. And the secret to building
a good language is making the right compromises in the
(23:18):
right places, and we need to be less concerned about
about having about shoving all of the guarantees up to eleven,
which also shoves all the math and the complexity up
to eleven, and more concerned about about programmer productivity in
catching the correctness and the right places. Right like marrying
(23:40):
a type system that is that's gradually typed with the
digit try turning it off and on again, Kerosy, where
do you think, Gotti?
Speaker 3 (23:48):
It makes sense, makes a lot of sense.
Speaker 1 (23:49):
I think that's like a big productivity is a big
reason why people push back again this Taskell for example,
right and right example. But you do need some sort
of correctness, and I think there's an eighty twenty over
here where you know you could get eighty percent of
the correctness benefits with like only twenty percent or very
little drop in productivity.
Speaker 3 (24:08):
I think that totally makes sense. Honestly, haven't.
Speaker 1 (24:11):
I haven't the experienced or thought much about the direction
of the gradual typing yet, so this is the first
time I'm getting to think about it.
Speaker 3 (24:19):
So I'm being a little slow to respond. Sorry about that.
Speaker 2 (24:22):
No, that's okay, that's okay, And I think that one
of the things that you're seeing is an intentional slowness
and carefulness in Jose's approach to the problem. And that's
really marvelous, really marvelous. I mean I've said this before,
but a great blues player is great because of the
of the rapid notes that that they can play, but
(24:43):
also the notes that they don't play right. And and
Jose has this has this beautiful balance and he has
this this beautiful way of sitting out. I mean, he's
not afraid to throw away a bad approach to a problem, right,
He's not afraid to step back from an experiment that
doesn't work. I mean, this is this is what the
(25:03):
second or third attempt that we've heard it typing, and
this one looks like it's going to sit right. Thanks, cool.
And I think that by being careful, he's going to
allow more of his customers to come along and to
and to adopt it and to you know, see the benefits.
I mean, everybody has been protected by now if you've
been using the modern elixir the type system, I've seen
(25:26):
three or four errors that will really really doosies that
the that the type system caught. I said, WHOA, how
did he do that.
Speaker 3 (25:33):
Yeah, I kind of want to give an example, like
a couple of years ago.
Speaker 1 (25:36):
This is like a similar way of thinking, I would say,
and Bruce correct me if I'm wrong. Like I was
working on building this application for a startup and one
of the parts of their problems was like a kind
of like a bend packing problem. And I might have
mentioned on the podcast a couple of times like more
like an NP hard problem not ourtimnistic, for which like
type system would be great to capture a lot of
(25:56):
use cases through type and compile time. And that's where
like I shows to use gleam because easily executable through
Elixir as like a binary Berline module. Right, However, you
choose to do that and let it, let it compile
and run that problem and gleam and the rest of
the elixir can be Elixir.
Speaker 3 (26:12):
It wasn't.
Speaker 1 (26:13):
It didn't quite work one hundred percent because once once
the gleam is done, there was still a few use
case of for us to cash, but they were limited,
right that you don't necessarily need to cash them, and
compilation or build you can write for expert test cases
instead of one hundred for that the type is going
to catch for you. I don't know if this is
like a similar kind of isn't a similar vein because
I think gradual typing allows you to type like a
(26:33):
from what I've seen, like only a part of your code,
right like just add that darting on the Yeah.
Speaker 2 (26:38):
Yeah, So I mean the whole idea is that is
that the system will infer what it can from the types,
the type clues that you provide and the heuristics of
the system, and and it'll just do the work. That's yeah,
it's it's really fantastic. And I think that that's really
going to be what you're going to see with the
(26:58):
with Gleam and the lick. It's the approach is how
much can I express and how much do I have
to express? And there are massive sections of the system
that we shouldn't have to think about typing, and and
we'll be more productive if we don't.
Speaker 4 (27:15):
So I do have a concern, right because I think
also Earlane is working on another type system, and then
of course we have Glean, but Glean, I'm aware they
do things differently, So I almost feel like, well, maybe
we should kind of push more stuff to Earline and
then just downstream it because I feel like, you know,
things are going to be coming from Earling, why not
start with a language? But maybe I'm thinking like because
(27:36):
they're not the same, right, So maybe it's not it's
not really required, but I feel like we're doing double
the work. Can we somehow lean more into the compiler
and more into the system and get more benefits rather
than you do your thing and then I'm going to
do this one and I'm going to We're going to
have three different type languages and different you know, can
we do something more universal that you can?
Speaker 2 (27:55):
I don't want to talk out of turn because I
don't really understand this system enough.
Speaker 4 (27:59):
But I know they're working on it. Each of them
have have their own style. Can we do something upstream
and then leverage the compilers who can get a little
bit more out of it? Can we I'm not saying
we should, but can we? Is the question? Sorry? But
go on?
Speaker 2 (28:12):
Yeah, yeah, I don't know, but I do think that
you know, there are warning signs, right we started to
see a fragmenting of the list community and that that's
probably what killed would eventually kill the Lisp language. So yeah,
it's a concern. I don't know the answers.
Speaker 4 (28:28):
Now, I'm also well, as long as we don't go
down the typescript style where nobody knows the difference between
what is it types?
Speaker 2 (28:36):
In?
Speaker 1 (28:36):
Is it?
Speaker 4 (28:37):
Like I forgot? Was there something I always had to
look up? I haven't gotten a good clear answer. It's
like you have type and then you have something else,
and I've never figured out the difference between the two.
Do you know what I'm talking about?
Speaker 2 (28:45):
No, don't. I'm not the type script guy at all.
Speaker 4 (28:48):
I'm not a types of person either. That's why I
tried looking up and everybody has a different description and
I have no idea which one. Sorry, Yes, that's that's
why I think the one is types of interface. So
I'm like, I don't know which one to use. Like,
I built this conversion, so I read in a graphic
file and I split out typescript file. First I sped
out classes, which they didn't like that. Then they said
spell out types and spelled types and they said, okay,
(29:09):
actually there's a better way to spit out interface. I
don't know the difference anymore. I know the difference between
interface and in a class, but not types of interfaces.
And this is a hot topic in the Typescript community,
and it's like monads. Everybody talks about it, but nobody
can really clearly explain it to where everybody gets it right.
Speaker 2 (29:26):
Yeah, I can't help you here.
Speaker 1 (29:28):
Yeah, I haven't used much of types. I think I
think the interface was is just the contract. You can
have an interface without a type too, right, like a
type is I want to start passing the object to
those functions, which is what you want to use objects
for in Typescript generally. But you could have an interface
without a type from vaguely.
Speaker 4 (29:47):
Yeah, that sounds alien to me because it is quite confusing.
They're almost interchangeable to a sort of I know there's
a difference, but anyways, I do hope that what Elixir
and also Erlaining and even glean what would do is
other people can also learn from it and expand on
it and maybe make it better in other languages, because
I feel so. I used to do a lot of
(30:08):
rails and I felt like they were kind of trendsetters,
and then I moved to Elixer, and now now I
feel like Elixer started trend steadying. Maybe it's whatever I'm doing.
People started getting magic. I don't know, but because with
like live view, then live view, then everybody started doing
a bunch of live view like kind of things, and
now we're working with types. Now I'm wondering if maybe
Ruby or Python or Python has something that it has
some kind of typing system, but maybe Ruby were also
starting to have something too.
Speaker 2 (30:28):
I think, well, I think the Ruby is Crystal, right,
So there's that Argentinian team that built the Crystal language
and it is it is gradually typed. Also, I believe
it's pretty cool.
Speaker 1 (30:40):
I think stor Bay is another thing right with Ruby
that I know. I know some teams use, so.
Speaker 2 (30:45):
That's follow that one.
Speaker 3 (30:47):
Okay. I think that's a gem that you can add
which adds typing. I think that's not gradual from.
Speaker 1 (30:53):
Our Remember, if you introduce it, it has a bleeding
effect the code.
Speaker 2 (30:59):
Yeah, it's been a while since I've written Ruby and Anger.
Speaker 4 (31:02):
So what have you done in Anger recently? That just
came to my mind.
Speaker 2 (31:07):
So mostly it's it's a elixer we've been kind of
focused on. We're doing a lot of mentoring style projects.
And we have a customer here in Chattanooga that is
a healthcare based customer that we're working with, and so
we're doing. We're doing some enrollment based code for them,
and we're using a framework that I like a lot
(31:27):
called Flop and that's kind of a table how to
represent paginated tables with with sorting and filtering, and it's
a It allows me to extend models through interfaces and protocols.
And yeah, I like it, like it quite a lot.
Speaker 4 (31:44):
Actually, you said flop you talk about the library Flop.
Speaker 2 (31:46):
Yeah, Flop and Flop Phoenix. Yeah.
Speaker 4 (31:49):
Yeah, I use Flop quite a lot. I quite like Flop.
You can do a lot of really interesting things with it.
Speaker 3 (31:54):
Yeah.
Speaker 2 (31:54):
Yeah. But in terms of languages, I've been I haven't
been eating my own dog food, right the least. The
last other programming language that that I wrote was probably
a little bit of Lua a couple of years ago.
Speaker 4 (32:07):
Well, well, speaking of of the of the programming language,
there's another new book from Praig Product there. I don't
think you wrote that one, also related to languages, but
a similar tone is what you had before.
Speaker 2 (32:17):
Yes, I was not involved with that one. It's the
first seven series book that that I haven't been involved with.
But it's fun to see that series come alive again
kind of so that that time was interesting because in
you know, at that point, most books were about were
focused on a very narrow problem, right, and very few
books were about broad explorations of topics. And so Seven
(32:41):
Languages in seven Weeks caught Joe Armstrong's imagination because he
wanted to trigger people's love of programming languages again, and
that book was very much that for me. And when
we when we wrote, when when I wrote the first
one that bradt said, Hey, this is a this is
a theme that could be pretty interesting, and this method
(33:03):
of thinking about problem solving could be interesting across databases
and currency models and things like that. So, yes, the
Obscure Languages, The Seven Obscure Languages in seven Weeks came
out recently, but I don't know exactly when the publication
date was, but but I saw that and attracted it
with a lot of interest.
Speaker 1 (33:22):
Yeah, what you said about seven Languages seven weeks, I've
taught enough about that book.
Speaker 3 (33:25):
It's it's amazing.
Speaker 1 (33:26):
I got that book because of the first language in
it was Ruby, and I wanted to learn Rube like quickly,
like a book that can teach me really quickly, because
it was my first job and I had to.
Speaker 3 (33:35):
Do Ruby and not know ruby.
Speaker 1 (33:36):
I read through it and it was like very smooth,
and I was like, oh wow, I feel like I
know a good amount of it. And that was like
just one one day and I'm like, you know, let
me take a week's break, and then the next week
I read to the entire book. It just was amazing,
and I think that the language blew my mind.
Speaker 3 (33:50):
Was prologue.
Speaker 1 (33:51):
I was like fourth, that just like kind of built
a lot of momentum. I just couldn't stop after that.
And then you finished the passchool. It's still amazing. Yeah,
it did ignite programming love, like you were saying, right, like,
I have not heard you say that before, but it
did for me. It's like one of the inflection points
for me. It to increase my ability to code longer
periods of time because I just like enjoyed that so much,
(34:12):
right Like, whenever I want to go longer, I kind
of tap into that feeling and like, oh, I'm.
Speaker 3 (34:16):
Able to like extend my how long and code.
Speaker 1 (34:18):
It's pretty awesome for anyone trying to like reignite your
love for programming.
Speaker 3 (34:22):
I'd highly recommend that book.
Speaker 2 (34:24):
Yeah, thank you so much. And so that book, there's
a couple of key points in the evolution of that
book that were really important, and one was Joe Armstrong
took me under his wing when we sent him three chapters.
We sent him Ruby, prologue and Erline. He skipped Ruby,
it didn't make any notes, and rather the prologue in
(34:45):
erline so arline I had based on his examples. Right,
I did the highly politically incorrect Russian Roulette game, and
you know, and so I wrote that, and so Joe said,
I get the sense that Bruce Tate understands Arling very well,
which was decidedly not true. Right, And then he said,
(35:06):
I don't get the same sense for prologue. This chapter
is a wreck. You can't publish this book, right. And
so I saw my bride and I reached out to him,
and he kind of welcomed me with open arms and
walked through and helped me write a better prologue checker.
And I can remember being on a phone or chat
or something with Erling or with Joe, and so I
(35:28):
was saying, well, I don't know where to go with
this program. He says, what are you writing? I say,
I'm doing a Suku solver, And he said, well, what
are you trying? To do and I said, well, I've
established the rules of the game. He said, did you
try to run it? And I ran the program and
it started spinning out answers and then he started laughing
and he said, that's your prolog moment. That is your
prolug moment. Those are really cool. So that was very
(35:50):
pivotal in the life of the book. And the other
one was when Stu Hallaway of Who's a Closure person
helped me think through distill the example to give in Scala. Yeah,
he gave a technical review of that book that was
just absolutely excellent. He helped me rethink what to do
in Skala as well as what to do in closure.
(36:11):
And I feel like with with those contributions, I was
able to adopt these new languages and not completely embarrass myself.
And those people also gave me access to others, uh,
others in the industry where I could ask some questions
and have some interviews about these these languages. And this
(36:32):
is really a cool moment, a career defining moment for me.
Speaker 3 (36:35):
That's really awesome.
Speaker 4 (36:36):
So he skipped the earling part and went straight in
the prologue. Okay, it's right, no, it's interest. Well, I mean,
I I understand there's definitely a lot of similarities. And
I and I think the first earlaning compiler or VM
or what do you want to call it, was was
on prologue and.
Speaker 2 (36:53):
Joe loves Loves loves prologue or loved.
Speaker 4 (36:56):
I guess, yeah, I mean, I guess you stopped loving
and started love arline to a certain extent. But then
it's so similar my understanding that not much changed necessarily, right,
maybe some better ideas, right.
Speaker 2 (37:08):
Yeah, So I guess so we let go of the
unification and started embracing the message passing, but kind of
the pattern matching grew out of that. The syntax who
is makes a whole lot more sense. So erline made
a whole lot more sentence after I wrote a little
bit of prologue. You know, I can remember going to
one of the first eucs, which is the aerline user
(37:30):
community in Sweden and Stockholm, and so I was, I
was answering a question about about something I said in
some languages, and so somebody said something about the erline syntax,
and I said, well, well, I guess I said that
I repeated what I said in the book, right, I
said in this cafeteria where all the people, all this
(37:51):
this erline community is around me. I repeated this this comment,
so I guess erline has an unnatural syntax. And then
halfway across to the cafeteria, Joe Armstrong stood up and said,
what do you mean? Erlang has a beautiful syntax? And
then you had this expectant grund you like you you
wanted to kick off a debate, and so I just
(38:12):
kind of I turned bright red and and just crumpled
and caved in the moment. It was really really funny.
Speaker 4 (38:19):
It definitely has a if you're not used to it,
it's definitely an interesting syntax, to say the least.
Speaker 2 (38:24):
It's a different family of languages, right. It's so we're
used to languages that that have come from you know,
the sea or the pythonline, and this is a this
is a language that that has come down from the
you know, the prologue and then on up that tree
that way.
Speaker 4 (38:39):
If I remember, isn't something like us commas after each
line or something and then use a semicolon, I forget
the rules. It's definitely I heard like it follows English
grammar sort of with the I want to say, punctuation,
but it's I don't know what it does.
Speaker 2 (38:55):
It does your clauses are separated by commas, and there's
a period at the end of everything, and so you know,
you always that's that's the way that all of Erline
is punctuated. And when you get that, it's it's a
lot simpler. But I didn't get that for a long
long time until I rote a little prologue and I thought, oh.
Speaker 1 (39:15):
In your book, I think you had prologue before Erlang.
Maybe for that reason. It's I had not experienced Airline
before that. It clicked right away. But what to be
at think one of the hilarious things about prologue is
I think it just shows a whole logic oriented aspect
of it.
Speaker 3 (39:29):
I think the statements are called facts, right, like the facts.
Speaker 1 (39:33):
And then the what you're trying to solve the question
is called a query. It's just incredible that it's it
doesn't feel like the programming luge. It feels like it
feels like almost like an AI, right, I mean, like
like it's more directly an EI than just like something
are coding. It just feels like something that solves something
for you. It's yeah, it's pretty pretty crazy, but yeah,
you're right. The whole syntax of erlang like matches quite
(39:54):
similarly to prologue.
Speaker 3 (39:55):
Can see the influencer.
Speaker 2 (39:56):
It's crazy, it does, it does, and if you want
to try something interesting, there's a interesting marriage. More of
like a prologue like language where the hard part about
prologue is getting is integrating with the rest of the
world after you have these insights, right, and more of
a traditional multi multi mode language. But there's a lisp
(40:20):
version of prologue called mini Canon that we covered in
seven more languages in seven weeks, and that's a really
cool experimentation of what you can do when you can
tie these concepts together. So for the listeners that are
looking to break their brains beyond what you can do
(40:40):
with prologue, this is a great way to go.
Speaker 1 (40:43):
Yeah, that's that's the one that I honestly and I
think that was at last chapter in your book, the
seven More I vaguely remember it was.
Speaker 3 (40:50):
I think it was kind of going over my head
a little bit.
Speaker 2 (40:54):
Right, Yeah, that was a demanding chapter to write. I
forgot who wrote that one, was it, Frank, I don't remember.
Speaker 4 (41:01):
I had a feeling like you tend to attract hard
books is that what it is? You have recent ones
to live you when you were talking about this one
was a tough fee and you almost kind of bring
some controversy right when you make the comment. And then Joe,
you know, spoke up or were you're trying to have
him speak up? Is that why you said what you said?
Speaker 2 (41:19):
Oh? No, no, no. So I don't lean into controversy
most of the times in my career. The few times
I have, I've really put my foot in it. But
I think in terms of difficult books, my best books
have always been when I was afraid or searching and
brought my readers, like genuinely brought my readers along for
(41:41):
the exploration because I think that during those times I
was searching for the same things that my readers were. Right, Like,
seven languages in seven weeks wasn't really about learning seven languages.
It was about saying, oh, wait a minute, we're out
of gas in this particular line of programming. Programming is
going somewhere else where's it going? And so that was
(42:02):
a journey, and the languages were chosen to specifically take
us step by step in a journey towards functional programming
from dynamically typed to strongly typed, and yeah, that's the
way that they were presented, and that's that's how I
actually did the research. And so some of the harder books,
like Designing Elixer Systems with OTP, that was one that
(42:23):
was written with James Gray, and he struggled with that
book for a while before I got involved, and he
had some immaculate research. But you know, you talk about
a list, James Gray is a human list machine. He
couldn't write a sentence without like seventeen parentheticals because he
knows everything about everything. And I taught James to lie
(42:45):
a little bit, right. I taught him to lie a
little bit and simplify and then go back and touch
those things up, right, and so then you can build
this framework. And then the other thing that I did
is we came up with the phrase together do fun
things with big loud worker bees. That helped walk people
through the layers of an Elixir system. And then once
we had that, the book just kind of flowed and
(43:07):
you know, it went from a year's long project and
then we just kind of hammered it out quickly. But yeah,
I like that. I like the challenge of writing about
blux topics and distilling them down to something interesting. And
the only reason that the Live View book is hard
is that the technology is fluid, and that is a
good thing. That means that we have a wonderful team
(43:29):
supporting this effort. And I'm just catching you know. We
made a conscious decision to be involved early in that
process because it's very hard to write a book and
framework at the same time.
Speaker 4 (43:42):
For the Actually, I do want to kind of go
back to the Design System's book. Yeah, Design Systems. I
have a couple of questions, right, but I'm just going
to start with the first one. Give you time to
answer that one. Let me just a little bit more.
When I heard about Alick early, when it kind of
first started coming out and becoming more popular, I believe
the talk was, oh, if you want to arline, you
have to go work at an Earlin company, because basically
(44:02):
there's no material, there's no patterns kind of out there
in the open, et cetera. Now we have your book,
and I think your book is a great example. In fact,
after I started, I didn't finish it still, but I
read enough to start to understand it. And what I
have to say is, now you've changed the way I
program stuff, and I'm really happy what that pattern has
made it more testable. I think with each book I
pick up from prag Prague in the Mixture, I just
start to change the way I write code differently. And
(44:23):
it's been all positives, right. Maybe I don't adopt everything
that's been given to me, but I adopt what I
feel makes sense. And so what I wanted to say was,
for your book, how did you learn how to design
ELIXTURER systems? I guess I could say, you know where,
where did you get your inspiration from? Was it just
asking people reading code?
Speaker 2 (44:40):
That's James, That's all James really, So James, he's so
little thing about him is he asked the mess and
so that means that he has very little of the
function in his hands, and so he types very very
slowly and has like some dictation that it works with.
But he's also one of the fastest programmers I've ever known, right,
(45:02):
because he's mensa, very smart he reads everything, and he
has to automate everything, and he's a wonderful human being
with tremendous empathy. So he writes for the people reading
his code, and you put all those things together, and
you get just this fountain of information. And so what
(45:23):
he did was put together the research. What I did
is helped to diskill the research into this layering system
that was easy to understand. And so what I did
is took all of James's immaculate research, we organized that,
we found this framework to do fund things framework. And
then after we found that, then we were able to
(45:43):
compartmentalize that research and throw out the things that didn't
fit right. And so then you got this marvelous exploration
of the data structure's gen elixir, which shapes the way
that I teach, shapes the way that I programmed. Then
you had this marvelous exploration of what a function is
and how you deal with errors and functions. And then
(46:03):
and then you have this this really cool treatment of
tests where we we build these tests that look like
nice pipelines, right, and then you know, later upon, later
upon layer the collaboration built something that neither one of
us could could have built alone.
Speaker 4 (46:19):
Well, my follow up question, what's funny that you start
mentioning James is what is James doing? Because I kind
of feel like he came in the community and he
kind of I don't know, I wouldn't even say left.
It just kind of disappeared. And and uh, because I'd
say you're definitely still out there, still doing things. A
lot of people are still in the community or sometimes
popping it up. But James seems to he helped with
this book, obviously saying that he did a lot of
(46:40):
the work on it. But then I don't really hear
too much of him since then. Is he still involved
with Elixir? Has he moved on or if you have, I'm.
Speaker 2 (46:48):
Gonna plug one of my competitors here. So he has
a so if you google Scrappy Programmer, James has a
has a course, has a live book as lips of
course really cool and it's especially good for training new
Elixir developers. And that's yeah. I'm a big fan of
his work, and I think there's room for what both
(47:10):
of us do. Yeah, But in terms of why you
don't see as much of them, he's a great dad.
He's got a wonderful daughter named Summer, and he's got
a great family. And yeah, so kudos to you, James
for being a great dat if you're listening.
Speaker 1 (47:25):
Yeah, I do not know about scrappy Programmer. That's very
very cool. Yeah, Yeah, we can link the Elixir forum
through it.
Speaker 3 (47:32):
We can link in the show notes for others to
grab that.
Speaker 2 (47:35):
Yeah, right under the Grazio forty percent off coupon of course. No. Yeah,
I think that readers will be really really happy to
get both. You'd be glad you did, and it'll jump
start your Elixir career.
Speaker 3 (47:50):
Awesome.
Speaker 1 (47:50):
I guess I wanted to talk about another book now.
I think this is like a bit of a different
book you mentioned earlier, Bruce currently away than something you've
been working on. You want to talk a little bit
about that.
Speaker 2 (48:02):
Yeah, So this is one that's been out for a while.
It's been out almost a year now. So in twenty
twenty two, my wife and I were battling depression, you know,
like a lot of us were in the electric community,
and she came up with this idea of doing the
COVID quarantine on a boat encircling the eastern United States.
So this book is one from the Pragmatic Bookshelf. You know, Alan,
(48:25):
you mentioned the Pragmatic Bookshelf and that it's a it's
a great resource for us. Well, it is. But one
of the things that the whole technical publishing industry is
facing is the competition with chat, GPT and all these
other resources that are online. In combination with the amount
of time that it takes to write a technical book,
(48:45):
by the time that it's out, it has a very
short shelf life, right.
Speaker 3 (48:48):
Well.
Speaker 2 (48:49):
One of the one of the prag responses to that
was to get some stories from people that are well
known in the in the industry, like myself, and so
we took this trip around the eastern United States and
told that story, right And so the story is about
so any travel memoirs, a story within a story. It's
about ninety thousand words, which works out to you know,
(49:11):
three to four hundred pages, and it talks about the
kind of underlying truths as we as we took on
this trip, you know, amidst the COVID pandemic and the
Russia's invasion of the Ukraine and that impact on fuel prices,
and people were buying boats and you know, sucking up
(49:31):
all the space and marinas and weather changing. All these
things were happening as we were doing this trip and
we were finding ourselves again. And so if people want
to find that, you can go to the Pragmatic bookshelf,
or you can go to Amazon and just look for
Currently Way, and we'd love to hear what you have
to say about that book.
Speaker 4 (49:50):
It's such a different book compared to what Pregmaty program
usually ships, Right, So I was kind of curious about
the discussions with that. Was that something that they how
do I say this pitched? And how was the reception
initially because usually there's there could be some changes in between.
Speaker 2 (50:05):
Right yeah, yeah, yeah. So the idea is that there's
a there is a need for publishers to have books
with a long tail, with a long shelf life, right,
And so the pragmatic bookshelf needs books like currently Away
and other stories that have some type of the technical interest.
(50:26):
And in this case, the technical interest is is kind
of doing a is a professional doing the job on
the road around the waterways, if you will. And there
are some other stories about about emerging technologies or or security,
but they cover more of the human side. And I
think that this is that it's a great way to
(50:48):
support the bookshelf. And it's also a great way to
kind of to kind of break away from to take
a small step away from your your technical education regimen
and kind of kind of fold in something else, if
you will. If that makes sense.
Speaker 4 (51:04):
Yeah, it does make sense. Just if somebody told me, oh,
we have this kind of book and it's from Prep,
I would say that sounds more like a I'm just
gonna throw Manning out there, because Manning does a lot
of different books that are not only tech related in
terms of like how to code something, but also how
to do your career or how to do whatever. Right,
this was a prec part when I think a PREP
packet's like more hands on, you're gonna learn a skill,
(51:25):
They're going to take you on a journey. It's just
such a different book in my mind.
Speaker 2 (51:29):
Yeah, it is. It is, and I think that's intentional
and I think that expanding casting the water net is
is pretty important.
Speaker 4 (51:35):
Now it has this triggered more potential other books in
this series. So you're gonna have this. I don't know
what what is the turn that you called this kind
of book long tail or or how would you long tail?
Speaker 2 (51:45):
Yeah, So basically this line is called pragmatic stories, and
it's basically non fiction with a with a technical bent,
and and so it's it's basically a story with the
human side that might be of Ventorius to the rothmatic community.
Speaker 3 (52:01):
Nice, it's really cool.
Speaker 1 (52:02):
I actually, until you did this in Trent twenty two, Bruce,
I didn't realize you could do a loop.
Speaker 3 (52:07):
I didn't realize you could do from one.
Speaker 1 (52:09):
Of the lakes all the way to the south of
the continent.
Speaker 3 (52:13):
That I didn't realize that's the thing. But it's really cool.
Speaker 2 (52:17):
Yeah, yeah, yeah. So well, we came down the Mississippi
for just a very short we were just on there
for three days. There's also this long canal called the
Tin Tom Waterway that goes between Alabama and Mississippi. I
mean roughly, it just kind of creeps away. But the
weird branch if you look at the map on the book,
is the cut the slash across Canada. There's this wild
(52:38):
system of canals and waterways, right. So some of the trip,
like about two days of that trip, we were on
these massively commercial waterways and there might have been a
container ship that was many, many tons right next to us, right,
And we have some picture taken right out of the
window of these big container boats and we actually used
(53:00):
the same locks that they did, and they would kind
of throw these these ropes down from way high on
the wall, and you know, we were kind of getting
blown off the wall and kind of some embarrassing stories
we're told in that book, because you know, we're not big,
big boating people. But then there are also these hand
operated locks right that that are a lot like the
(53:21):
Eery Canal, where the Eery Canal is operated mostly by engineers.
Well in Canada, these are operated by parks and wreck
people that are trained in hospitality and then they're taught
to do the canal work, right. And so we we
had a tremendous time cutting across the Canadian system with
these these canals. And then there's this place called Georgian
(53:44):
Bay and what's called the Small Ship Channel, where you
have some of the most rugged and beautiful terrain in
all the world. Right, so if you were to google
for the most beautiful cruising waters, one of the things
that would come up would be Georgian Bay. And we
spent a good like two or three weeks there.
Speaker 4 (54:02):
Are you only on the boat? Do you ever fly
a plane or no? Because I feel like most tech
guys when they start to get towards, you know, towards
some time in the career there they tend to be flying.
Speaker 2 (54:12):
Do you ever feel never have? And there's there's a
reason for that. You know, I'd consider getting my pilot's license,
but I am not a checklist kind of person. And
if you want to fly, you really really have to be.
And you know that's that's the same reason I don't
climb right, is my my brain doesn't doesn't click that way.
So I do a little bit of gem climbing, but
(54:34):
I don't do like outdoor high rock climbing because I
I forget little things right, and you can't. You can't
be a pilot and do that. And actually in Miami,
there was a man that said, Bruce, get a mission.
You got to get a boat like currently here. If
you change the mission, you got to change the boat.
And so we heard that and we thought about that,
(54:56):
and we sold our boat after the loop, and then
someday we're kind of looking around and someday we'll get
another boat that is more tuned for just cruising on
the on the Tennessee River. Right, we don't need something
that's made to do a great loop anymore. We need
something that's uh, that's just for.
Speaker 1 (55:14):
Yeah, it's pretty cool. I haven't checked out this book yet,
but I have bought it. It's on my list of
books to read it's it's really cool.
Speaker 3 (55:22):
Alan. Do you have anything else you want to come?
Speaker 4 (55:24):
Oh, I mean there's always tons of stuff I can
ask Bruce, but there does there does have to be
an end to this one. I mean, nothing's really sticking
out at the moment. Just I've been pretty happy with
the material that he's been putting out.
Speaker 3 (55:35):
Uh.
Speaker 4 (55:36):
And and you do kind of do these books with
other people, which is which is great. Where you don't
have a typical writing partner, which is good. But you
do have your name a lot of things. So maybe
that's your strategy. It's kind of tatching your name into
many things to show up get higher than that James
in the search when you're competing with him.
Speaker 3 (55:51):
Is that right?
Speaker 2 (55:52):
Oh, I'm not. I'm not trying to to compete with anybody.
I just like, uh, this human thing, right, it's a
it's all about people.
Speaker 4 (55:59):
I guess I can kind of most And my last
question with this is you know you just I think
you said you just went to code Beam. Where's your
next kind of public outing? I know what you call that?
The next place you're coming out in public, next kind
of conference or something like that you've planned.
Speaker 2 (56:12):
I am not traveling for the rest of the year.
I'll be going to a small conference in Portugal next
year next April, and then we have gig City Lakser
in May.
Speaker 4 (56:20):
That's it, all right, that was pretty easy. I think
you have a much much more. That's your schedule, but yeah,
that's my.
Speaker 1 (56:27):
Said Bruce is did we cover everything they want to cover?
Is there anything else would you like to kind of
pitch before you go to the picks?
Speaker 2 (56:34):
That's it for me, so thank you so much for
inviting me on. I've enjoyed this tremendously. Can't believe it's
been so long.
Speaker 1 (56:40):
Yeah, I know, that's crazy. So I guess I'm going
to transition as two picks. Alan, do you want to
go first? Do you have any picks today? Wow?
Speaker 4 (56:47):
I didn't even think about it, pick, it's been so long.
I guess the other thing I can pick is I've
been still playing around with this Ash framework and I
still think it's a great thing. You know, Bruce mentioned
about flop. We used to use a lot of flop,
but Ash actually does add a flop like thing to it,
So I find myself adding less and less plugins nowadays.
So if people haven't tried it out, I highly recommend
just play around with it.
Speaker 3 (57:07):
Nice.
Speaker 1 (57:08):
Yeah, I hadn't heard a flop until now, like really cool.
I feel like there's been a lot of frameworks that
do things in this direction, filtering, sorting, pagination, and yeah,
ash is a good call out. I'm yet to convince
a company to use it, but I have used it
on Alan since the recommended It's really cool.
Speaker 3 (57:27):
I guess a pick from me.
Speaker 1 (57:29):
I forgot who goes first the host, so I guess
I'll just go before I guess. I have these headphones
that I'm wearing right now. They're pretty interesting to the
company's shocks, and the model is open run, so they're
like not they don't go in second ears right, which
I don't. I'm not a fan of them. But they're
also not like headphones, like they're blocka ears. You just
(57:49):
put them on the side of your ear and it
transmits down through bone transmission. It's sae if I Google
research this, but it's pretty interesting. I did not know
this existed. It's really cool because you can listen to
you know what you were listening, but still be aware
of your surroundings. So I was always like a little
nervous about that. So yeah, to check check these out
(58:10):
shocks open run. I'll link it in the notes of
the show. Yeah, that's that's it for my pick today. Bruce,
do you have any picks for us?
Speaker 2 (58:18):
Yeah, I'm going to go with our Roxio for off.
It's a Black Friday sale we do. We do this
once a year and it's going to be all all
November and so come on, come on. Yeah, we have
an interesting way of teaching.
Speaker 3 (58:32):
Great.
Speaker 1 (58:32):
Well, I think that's it for today. Then, thanks so
much Bruce for coming. It was great chatting with you.
Speaker 3 (58:37):
After a while. And yeah, we'll see all your listeners
probably next week hopefully. I don't know.
Speaker 1 (58:43):
We don't have any guests scheduled, but we haven't been
recording as often as we used to.
Speaker 3 (58:47):
But yeah, we'll probably see you guys. So thanks everyone.
Speaker 2 (58:50):
Take care,
Hey, everyone, and welcome to another episode of Alexa Mix.
Today on the panel we have Alan Wima, Hello, and
we have a really special guest today and Bruce Taye.
Speaker 2 (00:15):
Hey, Bruce, Hi, everyone from Jed Nuga, Tennessee. I don't
get to say that on this podcast pretty often.
Speaker 3 (00:22):
Yeah, it's great to have Bruce back.
Speaker 1 (00:24):
We've had Bruce a couple of times in this podcast,
and he was one of the hosts of the podcast
a few years ago. So uh yeah, Bruce, your name
will always be kind of tied.
Speaker 3 (00:33):
To Alexa Mix in somewhere somewhere or another. But yeah, I.
Speaker 1 (00:36):
Guess like today we had Bruce to talk about a
few different topics.
Speaker 3 (00:39):
Bruce has a new book coming out. We'll touch on that.
Speaker 1 (00:42):
But Bruce also mentioned before you start recording, and every
time we have Bruce, we love to touch on the.
Speaker 3 (00:47):
Teaching aspect of things.
Speaker 1 (00:49):
Bruce does a lot of courses on groxio and there's
also a volunteer Mentorship slash Mentee session that happens once
a week. Alexa Chad, I would love to actually talk
about that first, If that's okay, Bruce, I would love
to Awesome, Awesome, Well why don't you why don't you
tell our listeners what Alexa Chad's about. I'm a huge
(01:10):
fan of that, and listeners will understand why in a bit.
Speaker 3 (01:14):
But yeah, in your own words, what is Alexa chat.
Speaker 2 (01:16):
Yeah, Audi is just a little bit humble because he's
one of our mentors and he's been a tremendous influence
in our attendees' lives. But this is a program to
teach underrepresented people Elixir, and we don't try to be
to lean really heavily into the underrepresented part. The idea
(01:37):
is you can be underrepresented in technology and you get
to define what that is when you join, or you
get to bring someone who is. So we're not exclusive,
but we do want the Elixir world to look a
lot like the rest of the world. And we're about
three things. The first thing is we believe that programmers
learn to code by coding. The second thing is that
(01:59):
we believe making mistakes in front of other people. So
if you're going to learn by coding, and if you're
going to drive in language you don't understand making mistakes
is a vulnerable thing. So the third thing is that
you have to lead with love and respect, and so
we spend more time getting to know each other than
you would otherwise expect. So this is not unusual for
(02:21):
us to shut down like the technical aspects of an
Elixir Chat program and then talk about somebody that gets
laid off or has a big interview coming up and
kind of rally around them and support them. So that's
what we're about. Coders learned to code by coding, making
mistakes a vulnerable thing, so you have to lead with love.
Speaker 1 (02:40):
Yeah, that's really amazing and I think that really kind
of sums up my experience being a mentor and alex
a chat as well. There is like a different side
of people that comes out when you build that comfort level,
like focusing on like making sure we create like a
comfortable slash safe space for them to talk to thrings
knowing each other more, They're able to ask questions more openly.
(03:02):
It's something that people look forward to every week. I've
noticed that, Like I haven't been to Alexa Chat in
a few months because of my work schedule, but it's
something I'm very eager to get back to as a mentor,
and I know as mentees there are others as well.
I want to give like a quick short testimonial. My
wife used to be one of the mentees the Alexa
Chat and Thin two or three years ago, and she
(03:23):
still talks about like how much she learned from like
she was definite, like six months, Like she like half
a year, but that six months of learning once a
week just over the course of years, that like really
contributes to a significant part of how much she has learned,
just because of the variety of ideas that is there
in that session. But yet it's really amazing. And I
already know the answer to this a little bit, but
(03:44):
I still want to touch on those cruise like what
inspired you to start Alexa Chat.
Speaker 2 (03:49):
Yeah, so this is a little bit of a dark story.
I think I was at a programming conference somewhere else,
and I'm not not going to say where that is,
and I'm not going to say which conference, but there
were four hundred and three attendees, and four hundred of
them were white males, and not that gosh, I'm a
white male, and I don't think that there's a problem
(04:10):
with that, but it didn't look like the rest of
the world, and that troubled me because this was supposed
to be a headlights type conference, a conference that that
kind of set the tone for where programming was going,
and if this was where programming was going, I had
a little bit of a problem with it. But I
didn't want to do a program that was super that
was hyper aggressive and political. So we came up with
(04:32):
this thing of define underrepresentative whatever way you want to
and either be underrepresentat or bring someone who is. And yeah, yeah,
when we did that, we found that when we engaged
with people and when we engaged around their own projects
and their own desires and we just stayed with them
(04:53):
and loved them until they were able to get some
traction in the community, good things happened. So there's a
there are a couple of success stories. I just talked
to Flora who gave a talk at code Beam Berlin
and Flora if you're listening high. So Flora was had
a marketing career and came to an Elixer chat and
(05:15):
she had a newborn baby and we had One of
the things that we do is we put on a
conference and the conference is his Gig City Elixer and
it's something that we put on in Chattanooga. It's really
about this program and if people don't understand that, the
idea is to put some of the best programmers in
the world. Sit them at the same table as our
(05:36):
mentees and start buyers, right see what happens. And so
we invited her to come. We said, your ticket is
free and all you have to do is get here.
We'll find a place for you to stay and everything.
And we did, and and she came, and she brought
the baby, and she was loved so hard through that whole,
that whole experience, and she met her eventual employer HDA
(05:58):
and Brian Hunter and that fabulous team hired her and
that relationship is an absolutely fantastic one. She's giving conference talks,
she's growing ways that nobody expected, and she's reading documentation
and reading code and finding bugs and code just from
being in the code base right just a year and
(06:19):
a half into doing this, and it's it's really a
fabulous thing. And that's that's kind of what we do.
It's not a rapid progress, but it's an effective way
to kind of share connections and experience with the people
in our community. And it's showing up on the stages
and the conferences that that's nothing could be more exciting
to me.
Speaker 4 (06:38):
The Elextrid I tried doing some Google searching and I
couldn't find any information about it, and I spawned it wrong.
Speaker 2 (06:43):
It's kind of it's kind of stealthy. So it's normally
invitationally we're casting a waternet.
Speaker 4 (06:51):
Today's okay, I know where I'm not wanted, you know.
Speaker 2 (06:59):
Yeah, well, well it's so it's it's a little bit delicate,
right because if you have to have enough mentors to
love the people that come, and if you don't, then
the people notice and you know, so that there's there's
a balance, right. So but if you want to join,
you can reach out to roxio Learning on x or
you could send an email to info at rox dot
(07:22):
io or to Maggie at rox dot io and she'll
she'll get you added to the program.
Speaker 3 (07:28):
Yeah, high, highly recommend it.
Speaker 1 (07:30):
I think I think one of the best parts about
joining that group on a weekly basis. I joined as
a mentor directly, but I think I obviously when you teach,
you learn yourself.
Speaker 3 (07:41):
That's the thing as well.
Speaker 1 (07:42):
But I think one of the things I realized, specifically
from Bruce and some of the other mentors there who
had more experience, like how to go about mentoring people
who are newer to coding as a whole and don't
help me at my work too as a senior and
a staff engineer to mentor others, Like, there is a
specific way to approach how to teach someone coding, and
(08:04):
obviously like taking that empathetic slash. Las Bruce says love approach,
like making making them feel comfortable and first before teaching
them something. It's seeing the effects of that really changed
the way I approach mentorship at work as well. So yeah,
highly recommend it for anyone who's in that underrepresentered category.
Speaker 3 (08:21):
Obviously to reach out to groxio. Yeah, it's a great group.
Speaker 1 (08:26):
I've done just like very not smoothly search our topics
to just groxio as a whole. Yeah, I guess pring
on proxio. I'm sure everyone has some idea what groxios
it's a learning website. How would you pitch roxio as
a whole? What is groxio about? Maybe some some of
your most popular courses on croxio.
Speaker 2 (08:46):
Yeah, So we do three different things. The first is
that we do we do trainings. Right now, our public
trainings are sold out for a bit, but we are
taking slots for commercial commercial slots. So if you have
a team that you'd like to train, we'd love to
talk to you. We also do video courses, and the
video courses are on a wide variety of topics. Our
(09:08):
most popular three are Phoenix Live View and of course
we'll be updating the courses shortly when live v one
dot O comes out. There's the OTP course, and there's
also an Elixir course, and there's a new emphasis on
groxio with a live book platform, and we have some
projects that kind of walk you through projects projects there.
(09:29):
And so what's interesting is that we have a Black
Friday sale. We have one massive sale that happens every
year and it is a forty percent off sale, and
so that's a site subscription forty percent off and that's
all the way through the last day of October, and
that'll be passed by the time we publish here, but
all the way through November. So hear this. We don't
(09:51):
do this often, but you get your forty percent off
at groxio just just by hanging out and coming.
Speaker 3 (09:57):
Yeah, another lot of things. Highly recommend it.
Speaker 1 (10:00):
I have the yearly subscription through one of the hackathons
prices and I've been it's been on my list to
check out the Nerves course. I see that some of
them were updated a little bit ago too, so that's
pretty cool.
Speaker 3 (10:13):
Yeah, I'm seeing some live book courses like you mentioned.
It's it's quite exciting.
Speaker 1 (10:17):
To see that being used on your on your website.
What are some of your most popular courses? I see
the Elixir one O TP live view. There's other languages too, right,
this Pony prologue. What are the codes that people usually
tend to just like go go for usually.
Speaker 2 (10:32):
Yeah, that's a that's kind of a little bit of
an embarrassing question. So this was intended to be a
seven languages in seven weeks. So the idea was just
going to be a programming passport for the world, right,
and you learn to code and then hopefully bring people
back through the licture by exposure the other languages. And
it was just too much, right, So learning a language
(10:53):
a month, I couldn't keep up the pace and it
about wreck me trying. So we we boiled the content
down to just mostly the Elixir. The prologue stuff is
probably still pretty good. But all those other courses are free,
and they're not really full courses. They're just like four
four videos and then a mini book. The idea is
that different learning methods. Are you open learning channels by
(11:16):
different learning methods? But yeah, mostly we're in an Elixer site,
got it? So Paulo, So, I don't know if everybody knows,
but I guess people know by now. But Paulo Vallin,
who is Joseph's brother, is working with us, and I
don't know if people know this though. He came from
a YouTube channel where he taught chemistry in Brazil. And
take a guess. I want to get a guest from
(11:38):
both of you, Audi and Allen how many views his
channels had in Brazil.
Speaker 1 (11:43):
I think I remember you mentioning it sometime, and yeah,
I think might be last time he came out. I
would say half, not so much, maybe maybe one hundred million.
Speaker 2 (11:53):
Sixty million, Yes, sixty million views in Brazil. Very good? Yeah,
great memory. But but anyway, he we brought Paulo on
because he really has a good understanding of how to
organize learning and learning content. And so since he came on,
we've we've kind of taken this haphazard for videos and
(12:13):
a microbook, and we we kind of took that down
to like a focused agenda with objectives and and we
kind of blew out a full course for the Elixir
Live View OTP, and we're kind of pushing that into
into the other the other Elixir topics as well. But
the OTP, I think is the big one that we're
kind of famous for, both the Elixir and OTP.
Speaker 4 (12:36):
I'm still kind of curious about Sorry to kind of sidetrack,
but about the Live View book because that's been in
writing for.
Speaker 2 (12:43):
For part some times.
Speaker 4 (12:46):
I understand why because I say something very sensitive, it sounds.
Speaker 2 (12:50):
Like, no, it's just that it's that. So it is
a we knew it was a big undertaking when we
took it on, and it has been a massive undertaking.
Speaker 1 (12:59):
Right.
Speaker 2 (12:59):
We've written the book about six times, right, And it's.
Speaker 4 (13:04):
Also because there they changed the APIs and everything right
was and I almost feel like maybe they did it
just to play with you, because it was pretty stable
for quite some time, right and then all right.
Speaker 2 (13:13):
Right, right, And I know our timing has not been good.
Our timing has not been good. Right, So so anyway,
there's say there's a beta out now that covers uh,
that covers Live. But what's what's happening with Live now
is that we're going to the new generating code is
going to I believe remove core components. Is that the
(13:35):
way it works. I think that's right. So if you
go into the main branch of Phoenix, you don't see
these modiles and these other components like that. It's it's
going to the generator is going to work a little
bit differently, and that obviously isn't going to change most
applications very much, right because you generate one time and
(13:55):
then you're good. It's going to hit us really really
hard because you know, we lean on like walking through
this generated code and showing this as the idiomatic live
you right. So yeah, but once liveew one dot zho
is released, we'll take a take a fool rewrite. It'll
go pretty quickly because we have the system down. And
let me give a shout out to the Phoenix team.
(14:16):
So all of these rewrites mean that the team has
been working incredibly hard for the people in the community,
and we were really lucky that there are so many
professionals that will dedicate so much of this their careers
to supporting this kind of new model of building well
I say new model, but a communal model of building software.
(14:38):
So I believe that open source software is like one
of these last bastions of hope right where people work
together for a common good. And so shout out to
the Phoenix team. Also a shout out to Sophie de Benedetto.
I had started taking on this project myself, and I
could not have done this myself. She's been I could
not have asked for a better writing partner, somebody that's
(15:01):
more intelligent, someone who's more gifted. And in fact, she
did such a good job with this. As I moved
out of my editorial role for the Pragmatic Bookshelf, she
was the one that I hand picked to fill that role.
Speaker 4 (15:13):
I haven't checked the recent edits to the book yet.
I started trying to look into it, but I haven't
gone back to it. But now it seems like we
have like basically a divergence. You have O dot twenty
and then you have one dot O r C. Is
your book doing one dot I'm guessing because and also
there's some differences but not like major. But at the
same time there is some differences that yes, yeah, it's.
Speaker 2 (15:33):
Kind of the one the one r C is there's
there's going to be a one more major divergence after
that that we haven't picked up on yet and that
hasn't been written yet, so.
Speaker 4 (15:45):
Kind of Also some breaking news and stuff that just
came to my mind right now. Did you see the
recent news that the sorrent I don't remember his name
now this is real name, but I think he comes
on my Soren too, and then that they believe is
products point one. Yeah, Parker exactly. Yeah, par he just
announced that he's doing open web open source. Now, yeah,
that's pretty cool.
Speaker 3 (16:06):
I would be.
Speaker 4 (16:06):
Interested to see if you can take a look with whenever.
I don't know when it's going to be released. It
was kind of talked about on the I think it's
alixir mentor YouTuber podcast video podcast. They mentioned it and
a lot of people were saying, oh, it's a really
really great live implementation. So I'm kind of curious if
you're going to hold off on the book and say,
take a look at that and see could be some
more interesting patterns or something that would be nice to
(16:27):
have in the book. Did you, well, now you know
about it, do you think that before right.
Speaker 2 (16:32):
I don't know, I don't know. I'll have to take
a look. Yeah, no commitment, sir.
Speaker 4 (16:37):
Yeah, I'm also kind of curious to see how he
implemented it, because I remember the first came out, I
was I got to like kind of beta trial it,
and I was pretty impressed with how everything kind of flowed.
The stuff you can do with live view is really
amazing for for everything and CSS however far that's advanced
since I mean you were all older programmers, you know.
I showed somebody, you know, how you turn an integer
(16:59):
and to a string with plus empty string, and they
were like, what are you talking kind of how weird
is that? I said that was JavaScript back then, what
are you talking about?
Speaker 3 (17:07):
Right?
Speaker 2 (17:08):
Right?
Speaker 4 (17:09):
CSS? Now it's like I barely ever really need to
write JavaScript, and it's been wonderful.
Speaker 2 (17:14):
Yeah, it's pretty amazing. And I mean that's that's been
one of the evolutions that the Phoenix teams. So when
you when you park so much infrastructure on top of
the web, which moves quickly anyway, and then you're sitting
on top of an emerging language like Elixer, it's a
pretty exciting thing, but it's it could be pretty demanding
for a team also, And one of the things I
(17:35):
wanted to talk about today was this idea that Elixer
is a language that you can't put a box around.
Speaker 1 (17:41):
Right.
Speaker 2 (17:42):
It's like when when I started working with Elixer, I
could point to two main gaping holes in it, right,
and I could say with certainty that those holes would
never be filled.
Speaker 3 (17:54):
Right.
Speaker 2 (17:54):
The first one was this this whole idea of immutability,
but particularly how it relates to machine learning and machine
learning models, right, Jose kind of he kind of said, well, well,
what if I swapped out the compiler and swapped in
this other piece, and you know, we did just in time,
(18:15):
and they kind of wrap this with the extensions of
the language, the protocols, and so I said, well, yeah,
you could do that, and he did that and was
stunningly successful. And the other thing was to go from
a strictly dynamically typed language to something that is gradually
typed now, and I thought, how do you do that?
(18:35):
How do you do that? And I wonder if what
we're seeing with the Elixir type system, I wonder where
types are going to go, right, because there's this tension
between what can you express with the type system and
what do you have to express with the type system?
Speaker 3 (18:51):
Right.
Speaker 2 (18:51):
One is is kind of how do you make developers
more productive? And one is how do you make the
how do you enabled the computer to make sure that
these these solutions are right?
Speaker 1 (19:04):
Right?
Speaker 2 (19:05):
So my head is exploding.
Speaker 1 (19:07):
Yeah, I think I think that's yeah, that's a good
way to put it, like the correctness versus productivity, Like
it's yeah, they're.
Speaker 3 (19:13):
At odds for sure.
Speaker 1 (19:14):
I think Alexa was a good like you kind of
like alluded to. It's a good middle ground right now,
it's going maybe going more towards a bit more on
the correctness side. But yeah, it's competing priorities for sure.
Speaker 2 (19:25):
So there's another element of this that I think is interesting,
and it's so I don't think that you could do that.
You could you could give any late leeway at all
in terms of the type system if you didn't have
OTP underneath you.
Speaker 3 (19:37):
Right.
Speaker 2 (19:37):
So I talk about this conversation a lot that I
overheard between Joe Armstrong and a man named David Turner
who was the creator of a language called Miranda that
was one of the first lazy languages. And so David
Turner so at this party and I think it was
code Beam in London. He said he said a word Joe,
and then Joe kind of walked over to him and
(19:59):
he said, I, I didn't think that you would be
able to achieve the reliability that you did with a
dynamic type system in Earlin, right, And it really is
pretty stunning that this whole idea of all this math
on one side and lamb at calculus and this tons
and tones of math making sure that things are correct,
(20:19):
and then on the other side this idea of did
you try turning it off and on again? Dude? Right,
But that's that's effectively what we had. And but that
system does work. The idea that you can reset to
a known, good starting state is extraordinarily powerful. And maybe
when you pair that with a type system that is
(20:41):
gradual but is enough but enables the programmer, maybe that
is a stunningly powerful combination. That's the way that we
want to build business applications. I wouldn't bet against.
Speaker 4 (20:52):
I also want to kind of lean in to hear it.
So I've been hosting a monthly Earlining Elixir focused meetup
and it's Asia based and I've been lucky enough because
of this podcast, and I also went to code being
Berlin and I got to meet a lot of people.
So I just been saying, Hey, can you do you
mind to come present somebody who had got to present
at the very beginning was Robert, one of the careers
of erline.
Speaker 2 (21:11):
Yeah, and it was.
Speaker 4 (21:12):
He gave a really great presentation and then at the
end there's a Q and a section and I think
I was the one who asked, hey, you know, why
is it that Erlang didn't have, you know, a strong
type system? And he said yeah, because actually in the
presentation he actually said people said we you know, we
couldn't do it, but actually we could.
Speaker 2 (21:27):
We just didn't.
Speaker 4 (21:28):
We just didn't want to do it. And he kind
of like said something like that and kind of skipped
over it. So then at the end, like I said,
I asked him, I could you expand that some more?
He said, well, basically, when you do hot code reload,
which we needed, things got a little bit weird with
with strong types. That's why we decided to remove it.
And okay, it makes sense. So now with this gradual
typing system, I'm curious what will happen now, Right, will
we have issues?
Speaker 3 (21:48):
Right?
Speaker 2 (21:49):
I think that probably not. I think that probably what
you'll see is that the gradual typing system will take
it as far as it can. It will basically push
the type system as far as math will go, right
and no further, And so that might take us out
some of these like highly theoretical like the monatic thing.
And I think it's really more than just type systems, right,
(22:12):
It's it's the idea that we in the computer science
field are kind of married to this idea of functional programming.
And there are very few fully functional functional languages, right,
and I think that maybe we need to get away
from that. And in fact, if you look at the
Erlang and Thelix or ecosystem, if you look at there's
a there's a state machine that you can build into
(22:34):
a process, right that a gen server can be implemented
as a state machine, right. And it's the message passing
that gives you the installation, the message passing to this
stateful entity that that gives you the protection and your
own design that's around that, right. So I said that
kind of poorly, right, So I guess what I mean
(22:57):
to say is that functional programming and immutability are all
just kind of there. There's shades of grade in there,
and there are some pretty hard and aggressive collisions between
between programmer productivity and correctness. And the secret to building
a good language is making the right compromises in the
(23:18):
right places, and we need to be less concerned about
about having about shoving all of the guarantees up to eleven,
which also shoves all the math and the complexity up
to eleven, and more concerned about about programmer productivity in
catching the correctness and the right places. Right like marrying
(23:40):
a type system that is that's gradually typed with the
digit try turning it off and on again, Kerosy, where
do you think, Gotti?
Speaker 3 (23:48):
It makes sense, makes a lot of sense.
Speaker 1 (23:49):
I think that's like a big productivity is a big
reason why people push back again this Taskell for example,
right and right example. But you do need some sort
of correctness, and I think there's an eighty twenty over
here where you know you could get eighty percent of
the correctness benefits with like only twenty percent or very
little drop in productivity.
Speaker 3 (24:08):
I think that totally makes sense. Honestly, haven't.
Speaker 1 (24:11):
I haven't the experienced or thought much about the direction
of the gradual typing yet, so this is the first
time I'm getting to think about it.
Speaker 3 (24:19):
So I'm being a little slow to respond. Sorry about that.
Speaker 2 (24:22):
No, that's okay, that's okay, And I think that one
of the things that you're seeing is an intentional slowness
and carefulness in Jose's approach to the problem. And that's
really marvelous, really marvelous. I mean I've said this before,
but a great blues player is great because of the
of the rapid notes that that they can play, but
(24:43):
also the notes that they don't play right. And and
Jose has this has this beautiful balance and he has
this this beautiful way of sitting out. I mean, he's
not afraid to throw away a bad approach to a problem, right,
He's not afraid to step back from an experiment that
doesn't work. I mean, this is this is what the
(25:03):
second or third attempt that we've heard it typing, and
this one looks like it's going to sit right. Thanks, cool.
And I think that by being careful, he's going to
allow more of his customers to come along and to
and to adopt it and to you know, see the benefits.
I mean, everybody has been protected by now if you've
been using the modern elixir the type system, I've seen
(25:26):
three or four errors that will really really doosies that
the that the type system caught. I said, WHOA, how
did he do that.
Speaker 3 (25:33):
Yeah, I kind of want to give an example, like
a couple of years ago.
Speaker 1 (25:36):
This is like a similar way of thinking, I would say,
and Bruce correct me if I'm wrong. Like I was
working on building this application for a startup and one
of the parts of their problems was like a kind
of like a bend packing problem. And I might have
mentioned on the podcast a couple of times like more
like an NP hard problem not ourtimnistic, for which like
type system would be great to capture a lot of
(25:56):
use cases through type and compile time. And that's where
like I shows to use gleam because easily executable through
Elixir as like a binary Berline module. Right, However, you
choose to do that and let it, let it compile
and run that problem and gleam and the rest of
the elixir can be Elixir.
Speaker 3 (26:12):
It wasn't.
Speaker 1 (26:13):
It didn't quite work one hundred percent because once once
the gleam is done, there was still a few use
case of for us to cash, but they were limited,
right that you don't necessarily need to cash them, and
compilation or build you can write for expert test cases
instead of one hundred for that the type is going
to catch for you. I don't know if this is
like a similar kind of isn't a similar vein because
I think gradual typing allows you to type like a
(26:33):
from what I've seen, like only a part of your code,
right like just add that darting on the Yeah.
Speaker 2 (26:38):
Yeah, So I mean the whole idea is that is
that the system will infer what it can from the types,
the type clues that you provide and the heuristics of
the system, and and it'll just do the work. That's yeah,
it's it's really fantastic. And I think that that's really
going to be what you're going to see with the
(26:58):
with Gleam and the lick. It's the approach is how
much can I express and how much do I have
to express? And there are massive sections of the system
that we shouldn't have to think about typing, and and
we'll be more productive if we don't.
Speaker 4 (27:15):
So I do have a concern, right because I think
also Earlane is working on another type system, and then
of course we have Glean, but Glean, I'm aware they
do things differently, So I almost feel like, well, maybe
we should kind of push more stuff to Earline and
then just downstream it because I feel like, you know,
things are going to be coming from Earling, why not
start with a language? But maybe I'm thinking like because
(27:36):
they're not the same, right, So maybe it's not it's
not really required, but I feel like we're doing double
the work. Can we somehow lean more into the compiler
and more into the system and get more benefits rather
than you do your thing and then I'm going to
do this one and I'm going to We're going to
have three different type languages and different you know, can
we do something more universal that you can?
Speaker 2 (27:55):
I don't want to talk out of turn because I
don't really understand this system enough.
Speaker 4 (27:59):
But I know they're working on it. Each of them
have have their own style. Can we do something upstream
and then leverage the compilers who can get a little
bit more out of it? Can we I'm not saying
we should, but can we? Is the question? Sorry? But
go on?
Speaker 2 (28:12):
Yeah, yeah, I don't know, but I do think that
you know, there are warning signs, right we started to
see a fragmenting of the list community and that that's
probably what killed would eventually kill the Lisp language. So yeah,
it's a concern. I don't know the answers.
Speaker 4 (28:28):
Now, I'm also well, as long as we don't go
down the typescript style where nobody knows the difference between
what is it types?
Speaker 2 (28:36):
In?
Speaker 1 (28:36):
Is it?
Speaker 4 (28:37):
Like I forgot? Was there something I always had to
look up? I haven't gotten a good clear answer. It's
like you have type and then you have something else,
and I've never figured out the difference between the two.
Do you know what I'm talking about?
Speaker 2 (28:45):
No, don't. I'm not the type script guy at all.
Speaker 4 (28:48):
I'm not a types of person either. That's why I
tried looking up and everybody has a different description and
I have no idea which one. Sorry, Yes, that's that's
why I think the one is types of interface. So
I'm like, I don't know which one to use. Like,
I built this conversion, so I read in a graphic
file and I split out typescript file. First I sped
out classes, which they didn't like that. Then they said
spell out types and spelled types and they said, okay,
(29:09):
actually there's a better way to spit out interface. I
don't know the difference anymore. I know the difference between
interface and in a class, but not types of interfaces.
And this is a hot topic in the Typescript community,
and it's like monads. Everybody talks about it, but nobody
can really clearly explain it to where everybody gets it right.
Speaker 2 (29:26):
Yeah, I can't help you here.
Speaker 1 (29:28):
Yeah, I haven't used much of types. I think I
think the interface was is just the contract. You can
have an interface without a type too, right, like a
type is I want to start passing the object to
those functions, which is what you want to use objects
for in Typescript generally. But you could have an interface
without a type from vaguely.
Speaker 4 (29:47):
Yeah, that sounds alien to me because it is quite confusing.
They're almost interchangeable to a sort of I know there's
a difference, but anyways, I do hope that what Elixir
and also Erlaining and even glean what would do is
other people can also learn from it and expand on
it and maybe make it better in other languages, because
I feel so. I used to do a lot of
(30:08):
rails and I felt like they were kind of trendsetters,
and then I moved to Elixer, and now now I
feel like Elixer started trend steadying. Maybe it's whatever I'm doing.
People started getting magic. I don't know, but because with
like live view, then live view, then everybody started doing
a bunch of live view like kind of things, and
now we're working with types. Now I'm wondering if maybe
Ruby or Python or Python has something that it has
some kind of typing system, but maybe Ruby were also
starting to have something too.
Speaker 2 (30:28):
I think, well, I think the Ruby is Crystal, right,
So there's that Argentinian team that built the Crystal language
and it is it is gradually typed. Also, I believe
it's pretty cool.
Speaker 1 (30:40):
I think stor Bay is another thing right with Ruby
that I know. I know some teams use, so.
Speaker 2 (30:45):
That's follow that one.
Speaker 3 (30:47):
Okay. I think that's a gem that you can add
which adds typing. I think that's not gradual from.
Speaker 1 (30:53):
Our Remember, if you introduce it, it has a bleeding
effect the code.
Speaker 2 (30:59):
Yeah, it's been a while since I've written Ruby and Anger.
Speaker 4 (31:02):
So what have you done in Anger recently? That just
came to my mind.
Speaker 2 (31:07):
So mostly it's it's a elixer we've been kind of
focused on. We're doing a lot of mentoring style projects.
And we have a customer here in Chattanooga that is
a healthcare based customer that we're working with, and so
we're doing. We're doing some enrollment based code for them,
and we're using a framework that I like a lot
(31:27):
called Flop and that's kind of a table how to
represent paginated tables with with sorting and filtering, and it's
a It allows me to extend models through interfaces and protocols.
And yeah, I like it, like it quite a lot.
Speaker 4 (31:44):
Actually, you said flop you talk about the library Flop.
Speaker 2 (31:46):
Yeah, Flop and Flop Phoenix. Yeah.
Speaker 4 (31:49):
Yeah, I use Flop quite a lot. I quite like Flop.
You can do a lot of really interesting things with it.
Speaker 3 (31:54):
Yeah.
Speaker 2 (31:54):
Yeah. But in terms of languages, I've been I haven't
been eating my own dog food, right the least. The
last other programming language that that I wrote was probably
a little bit of Lua a couple of years ago.
Speaker 4 (32:07):
Well, well, speaking of of the of the programming language,
there's another new book from Praig Product there. I don't
think you wrote that one, also related to languages, but
a similar tone is what you had before.
Speaker 2 (32:17):
Yes, I was not involved with that one. It's the
first seven series book that that I haven't been involved with.
But it's fun to see that series come alive again
kind of so that that time was interesting because in
you know, at that point, most books were about were
focused on a very narrow problem, right, and very few
books were about broad explorations of topics. And so Seven
(32:41):
Languages in seven Weeks caught Joe Armstrong's imagination because he
wanted to trigger people's love of programming languages again, and
that book was very much that for me. And when
we when we wrote, when when I wrote the first
one that bradt said, Hey, this is a this is
a theme that could be pretty interesting, and this method
(33:03):
of thinking about problem solving could be interesting across databases
and currency models and things like that. So, yes, the
Obscure Languages, The Seven Obscure Languages in seven Weeks came
out recently, but I don't know exactly when the publication
date was, but but I saw that and attracted it
with a lot of interest.
Speaker 1 (33:22):
Yeah, what you said about seven Languages seven weeks, I've
taught enough about that book.
Speaker 3 (33:25):
It's it's amazing.
Speaker 1 (33:26):
I got that book because of the first language in
it was Ruby, and I wanted to learn Rube like quickly,
like a book that can teach me really quickly, because
it was my first job and I had to.
Speaker 3 (33:35):
Do Ruby and not know ruby.
Speaker 1 (33:36):
I read through it and it was like very smooth,
and I was like, oh wow, I feel like I
know a good amount of it. And that was like
just one one day and I'm like, you know, let
me take a week's break, and then the next week
I read to the entire book. It just was amazing,
and I think that the language blew my mind.
Speaker 3 (33:50):
Was prologue.
Speaker 1 (33:51):
I was like fourth, that just like kind of built
a lot of momentum. I just couldn't stop after that.
And then you finished the passchool. It's still amazing. Yeah,
it did ignite programming love, like you were saying, right, like,
I have not heard you say that before, but it
did for me. It's like one of the inflection points
for me. It to increase my ability to code longer
periods of time because I just like enjoyed that so much,
(34:12):
right Like, whenever I want to go longer, I kind
of tap into that feeling and like, oh, I'm.
Speaker 3 (34:16):
Able to like extend my how long and code.
Speaker 1 (34:18):
It's pretty awesome for anyone trying to like reignite your
love for programming.
Speaker 3 (34:22):
I'd highly recommend that book.
Speaker 2 (34:24):
Yeah, thank you so much. And so that book, there's
a couple of key points in the evolution of that
book that were really important, and one was Joe Armstrong
took me under his wing when we sent him three chapters.
We sent him Ruby, prologue and Erline. He skipped Ruby,
it didn't make any notes, and rather the prologue in
(34:45):
erline so arline I had based on his examples. Right,
I did the highly politically incorrect Russian Roulette game, and
you know, and so I wrote that, and so Joe said,
I get the sense that Bruce Tate understands Arling very well,
which was decidedly not true. Right, And then he said,
(35:06):
I don't get the same sense for prologue. This chapter
is a wreck. You can't publish this book, right. And
so I saw my bride and I reached out to him,
and he kind of welcomed me with open arms and
walked through and helped me write a better prologue checker.
And I can remember being on a phone or chat
or something with Erling or with Joe, and so I
(35:28):
was saying, well, I don't know where to go with
this program. He says, what are you writing? I say,
I'm doing a Suku solver, And he said, well, what
are you trying? To do and I said, well, I've
established the rules of the game. He said, did you
try to run it? And I ran the program and
it started spinning out answers and then he started laughing
and he said, that's your prolog moment. That is your
prolug moment. Those are really cool. So that was very
(35:50):
pivotal in the life of the book. And the other
one was when Stu Hallaway of Who's a Closure person
helped me think through distill the example to give in Scala. Yeah,
he gave a technical review of that book that was
just absolutely excellent. He helped me rethink what to do
in Skala as well as what to do in closure.
(36:11):
And I feel like with with those contributions, I was
able to adopt these new languages and not completely embarrass myself.
And those people also gave me access to others, uh,
others in the industry where I could ask some questions
and have some interviews about these these languages. And this
(36:32):
is really a cool moment, a career defining moment for me.
Speaker 3 (36:35):
That's really awesome.
Speaker 4 (36:36):
So he skipped the earling part and went straight in
the prologue. Okay, it's right, no, it's interest. Well, I mean,
I I understand there's definitely a lot of similarities. And
I and I think the first earlaning compiler or VM
or what do you want to call it, was was
on prologue and.
Speaker 2 (36:53):
Joe loves Loves loves prologue or loved.
Speaker 4 (36:56):
I guess, yeah, I mean, I guess you stopped loving
and started love arline to a certain extent. But then
it's so similar my understanding that not much changed necessarily, right,
maybe some better ideas, right.
Speaker 2 (37:08):
Yeah, So I guess so we let go of the
unification and started embracing the message passing, but kind of
the pattern matching grew out of that. The syntax who
is makes a whole lot more sense. So erline made
a whole lot more sentence after I wrote a little
bit of prologue. You know, I can remember going to
one of the first eucs, which is the aerline user
(37:30):
community in Sweden and Stockholm, and so I was, I
was answering a question about about something I said in
some languages, and so somebody said something about the erline syntax,
and I said, well, well, I guess I said that
I repeated what I said in the book, right, I
said in this cafeteria where all the people, all this
(37:51):
this erline community is around me. I repeated this this comment,
so I guess erline has an unnatural syntax. And then
halfway across to the cafeteria, Joe Armstrong stood up and said,
what do you mean? Erlang has a beautiful syntax? And
then you had this expectant grund you like you you
wanted to kick off a debate, and so I just
(38:12):
kind of I turned bright red and and just crumpled
and caved in the moment. It was really really funny.
Speaker 4 (38:19):
It definitely has a if you're not used to it,
it's definitely an interesting syntax, to say the least.
Speaker 2 (38:24):
It's a different family of languages, right. It's so we're
used to languages that that have come from you know,
the sea or the pythonline, and this is a this
is a language that that has come down from the
you know, the prologue and then on up that tree
that way.
Speaker 4 (38:39):
If I remember, isn't something like us commas after each
line or something and then use a semicolon, I forget
the rules. It's definitely I heard like it follows English
grammar sort of with the I want to say, punctuation,
but it's I don't know what it does.
Speaker 2 (38:55):
It does your clauses are separated by commas, and there's
a period at the end of everything, and so you know,
you always that's that's the way that all of Erline
is punctuated. And when you get that, it's it's a
lot simpler. But I didn't get that for a long
long time until I rote a little prologue and I thought, oh.
Speaker 1 (39:15):
In your book, I think you had prologue before Erlang.
Maybe for that reason. It's I had not experienced Airline
before that. It clicked right away. But what to be
at think one of the hilarious things about prologue is
I think it just shows a whole logic oriented aspect
of it.
Speaker 3 (39:29):
I think the statements are called facts, right, like the facts.
Speaker 1 (39:33):
And then the what you're trying to solve the question
is called a query. It's just incredible that it's it
doesn't feel like the programming luge. It feels like it
feels like almost like an AI, right, I mean, like
like it's more directly an EI than just like something
are coding. It just feels like something that solves something
for you. It's yeah, it's pretty pretty crazy, but yeah,
you're right. The whole syntax of erlang like matches quite
(39:54):
similarly to prologue.
Speaker 3 (39:55):
Can see the influencer.
Speaker 2 (39:56):
It's crazy, it does, it does, and if you want
to try something interesting, there's a interesting marriage. More of
like a prologue like language where the hard part about
prologue is getting is integrating with the rest of the
world after you have these insights, right, and more of
a traditional multi multi mode language. But there's a lisp
(40:20):
version of prologue called mini Canon that we covered in
seven more languages in seven weeks, and that's a really
cool experimentation of what you can do when you can
tie these concepts together. So for the listeners that are
looking to break their brains beyond what you can do
(40:40):
with prologue, this is a great way to go.
Speaker 1 (40:43):
Yeah, that's that's the one that I honestly and I
think that was at last chapter in your book, the
seven More I vaguely remember it was.
Speaker 3 (40:50):
I think it was kind of going over my head
a little bit.
Speaker 2 (40:54):
Right, Yeah, that was a demanding chapter to write. I
forgot who wrote that one, was it, Frank, I don't remember.
Speaker 4 (41:01):
I had a feeling like you tend to attract hard
books is that what it is? You have recent ones
to live you when you were talking about this one
was a tough fee and you almost kind of bring
some controversy right when you make the comment. And then Joe,
you know, spoke up or were you're trying to have
him speak up? Is that why you said what you said?
Speaker 2 (41:19):
Oh? No, no, no. So I don't lean into controversy
most of the times in my career. The few times
I have, I've really put my foot in it. But
I think in terms of difficult books, my best books
have always been when I was afraid or searching and
brought my readers, like genuinely brought my readers along for
(41:41):
the exploration because I think that during those times I
was searching for the same things that my readers were. Right, Like,
seven languages in seven weeks wasn't really about learning seven languages.
It was about saying, oh, wait a minute, we're out
of gas in this particular line of programming. Programming is
going somewhere else where's it going? And so that was
(42:02):
a journey, and the languages were chosen to specifically take
us step by step in a journey towards functional programming
from dynamically typed to strongly typed, and yeah, that's the
way that they were presented, and that's that's how I
actually did the research. And so some of the harder books,
like Designing Elixer Systems with OTP, that was one that
(42:23):
was written with James Gray, and he struggled with that
book for a while before I got involved, and he
had some immaculate research. But you know, you talk about
a list, James Gray is a human list machine. He
couldn't write a sentence without like seventeen parentheticals because he
knows everything about everything. And I taught James to lie
(42:45):
a little bit, right. I taught him to lie a
little bit and simplify and then go back and touch
those things up, right, and so then you can build
this framework. And then the other thing that I did
is we came up with the phrase together do fun
things with big loud worker bees. That helped walk people
through the layers of an Elixir system. And then once
we had that, the book just kind of flowed and
(43:07):
you know, it went from a year's long project and
then we just kind of hammered it out quickly. But yeah,
I like that. I like the challenge of writing about
blux topics and distilling them down to something interesting. And
the only reason that the Live View book is hard
is that the technology is fluid, and that is a
good thing. That means that we have a wonderful team
(43:29):
supporting this effort. And I'm just catching you know. We
made a conscious decision to be involved early in that
process because it's very hard to write a book and
framework at the same time.
Speaker 4 (43:42):
For the Actually, I do want to kind of go
back to the Design System's book. Yeah, Design Systems. I
have a couple of questions, right, but I'm just going
to start with the first one. Give you time to
answer that one. Let me just a little bit more.
When I heard about Alick early, when it kind of
first started coming out and becoming more popular, I believe
the talk was, oh, if you want to arline, you
have to go work at an Earlin company, because basically
(44:02):
there's no material, there's no patterns kind of out there
in the open, et cetera. Now we have your book,
and I think your book is a great example. In fact,
after I started, I didn't finish it still, but I
read enough to start to understand it. And what I
have to say is, now you've changed the way I
program stuff, and I'm really happy what that pattern has
made it more testable. I think with each book I
pick up from prag Prague in the Mixture, I just
start to change the way I write code differently. And
(44:23):
it's been all positives, right. Maybe I don't adopt everything
that's been given to me, but I adopt what I
feel makes sense. And so what I wanted to say was,
for your book, how did you learn how to design
ELIXTURER systems? I guess I could say, you know where,
where did you get your inspiration from? Was it just
asking people reading code?
Speaker 2 (44:40):
That's James, That's all James really, So James, he's so
little thing about him is he asked the mess and
so that means that he has very little of the
function in his hands, and so he types very very
slowly and has like some dictation that it works with.
But he's also one of the fastest programmers I've ever known, right,
(45:02):
because he's mensa, very smart he reads everything, and he
has to automate everything, and he's a wonderful human being
with tremendous empathy. So he writes for the people reading
his code, and you put all those things together, and
you get just this fountain of information. And so what
(45:23):
he did was put together the research. What I did
is helped to diskill the research into this layering system
that was easy to understand. And so what I did
is took all of James's immaculate research, we organized that,
we found this framework to do fund things framework. And
then after we found that, then we were able to
(45:43):
compartmentalize that research and throw out the things that didn't
fit right. And so then you got this marvelous exploration
of the data structure's gen elixir, which shapes the way
that I teach, shapes the way that I programmed. Then
you had this marvelous exploration of what a function is
and how you deal with errors and functions. And then
(46:03):
and then you have this this really cool treatment of
tests where we we build these tests that look like
nice pipelines, right, and then you know, later upon, later
upon layer the collaboration built something that neither one of
us could could have built alone.
Speaker 4 (46:19):
Well, my follow up question, what's funny that you start
mentioning James is what is James doing? Because I kind
of feel like he came in the community and he
kind of I don't know, I wouldn't even say left.
It just kind of disappeared. And and uh, because I'd
say you're definitely still out there, still doing things. A
lot of people are still in the community or sometimes
popping it up. But James seems to he helped with
this book, obviously saying that he did a lot of
(46:40):
the work on it. But then I don't really hear
too much of him since then. Is he still involved
with Elixir? Has he moved on or if you have, I'm.
Speaker 2 (46:48):
Gonna plug one of my competitors here. So he has
a so if you google Scrappy Programmer, James has a
has a course, has a live book as lips of
course really cool and it's especially good for training new
Elixir developers. And that's yeah. I'm a big fan of
his work, and I think there's room for what both
(47:10):
of us do. Yeah, But in terms of why you
don't see as much of them, he's a great dad.
He's got a wonderful daughter named Summer, and he's got
a great family. And yeah, so kudos to you, James
for being a great dat if you're listening.
Speaker 1 (47:25):
Yeah, I do not know about scrappy Programmer. That's very
very cool. Yeah, Yeah, we can link the Elixir forum
through it.
Speaker 3 (47:32):
We can link in the show notes for others to
grab that.
Speaker 2 (47:35):
Yeah, right under the Grazio forty percent off coupon of course. No. Yeah,
I think that readers will be really really happy to
get both. You'd be glad you did, and it'll jump
start your Elixir career.
Speaker 3 (47:50):
Awesome.
Speaker 1 (47:50):
I guess I wanted to talk about another book now.
I think this is like a bit of a different
book you mentioned earlier, Bruce currently away than something you've
been working on. You want to talk a little bit
about that.
Speaker 2 (48:02):
Yeah, So this is one that's been out for a while.
It's been out almost a year now. So in twenty
twenty two, my wife and I were battling depression, you know,
like a lot of us were in the electric community,
and she came up with this idea of doing the
COVID quarantine on a boat encircling the eastern United States.
So this book is one from the Pragmatic Bookshelf. You know, Alan,
(48:25):
you mentioned the Pragmatic Bookshelf and that it's a it's
a great resource for us. Well, it is. But one
of the things that the whole technical publishing industry is
facing is the competition with chat, GPT and all these
other resources that are online. In combination with the amount
of time that it takes to write a technical book,
(48:45):
by the time that it's out, it has a very
short shelf life, right.
Speaker 3 (48:48):
Well.
Speaker 2 (48:49):
One of the one of the prag responses to that
was to get some stories from people that are well
known in the in the industry, like myself, and so
we took this trip around the eastern United States and
told that story, right And so the story is about
so any travel memoirs, a story within a story. It's
about ninety thousand words, which works out to you know,
(49:11):
three to four hundred pages, and it talks about the
kind of underlying truths as we as we took on
this trip, you know, amidst the COVID pandemic and the
Russia's invasion of the Ukraine and that impact on fuel prices,
and people were buying boats and you know, sucking up
(49:31):
all the space and marinas and weather changing. All these
things were happening as we were doing this trip and
we were finding ourselves again. And so if people want
to find that, you can go to the Pragmatic bookshelf,
or you can go to Amazon and just look for
Currently Way, and we'd love to hear what you have
to say about that book.
Speaker 4 (49:50):
It's such a different book compared to what Pregmaty program
usually ships, Right, So I was kind of curious about
the discussions with that. Was that something that they how
do I say this pitched? And how was the reception
initially because usually there's there could be some changes in between.
Speaker 2 (50:05):
Right yeah, yeah, yeah. So the idea is that there's
a there is a need for publishers to have books
with a long tail, with a long shelf life, right,
And so the pragmatic bookshelf needs books like currently Away
and other stories that have some type of the technical interest.
(50:26):
And in this case, the technical interest is is kind
of doing a is a professional doing the job on
the road around the waterways, if you will. And there
are some other stories about about emerging technologies or or security,
but they cover more of the human side. And I
think that this is that it's a great way to
(50:48):
support the bookshelf. And it's also a great way to
kind of to kind of break away from to take
a small step away from your your technical education regimen
and kind of kind of fold in something else, if
you will. If that makes sense.
Speaker 4 (51:04):
Yeah, it does make sense. Just if somebody told me, oh,
we have this kind of book and it's from Prep,
I would say that sounds more like a I'm just
gonna throw Manning out there, because Manning does a lot
of different books that are not only tech related in
terms of like how to code something, but also how
to do your career or how to do whatever. Right,
this was a prec part when I think a PREP
packet's like more hands on, you're gonna learn a skill,
(51:25):
They're going to take you on a journey. It's just
such a different book in my mind.
Speaker 2 (51:29):
Yeah, it is. It is, and I think that's intentional
and I think that expanding casting the water net is
is pretty important.
Speaker 4 (51:35):
Now it has this triggered more potential other books in
this series. So you're gonna have this. I don't know
what what is the turn that you called this kind
of book long tail or or how would you long tail?
Speaker 2 (51:45):
Yeah, So basically this line is called pragmatic stories, and
it's basically non fiction with a with a technical bent,
and and so it's it's basically a story with the
human side that might be of Ventorius to the rothmatic community.
Speaker 3 (52:01):
Nice, it's really cool.
Speaker 1 (52:02):
I actually, until you did this in Trent twenty two, Bruce,
I didn't realize you could do a loop.
Speaker 3 (52:07):
I didn't realize you could do from one.
Speaker 1 (52:09):
Of the lakes all the way to the south of
the continent.
Speaker 3 (52:13):
That I didn't realize that's the thing. But it's really cool.
Speaker 2 (52:17):
Yeah, yeah, yeah. So well, we came down the Mississippi
for just a very short we were just on there
for three days. There's also this long canal called the
Tin Tom Waterway that goes between Alabama and Mississippi. I
mean roughly, it just kind of creeps away. But the
weird branch if you look at the map on the book,
is the cut the slash across Canada. There's this wild
(52:38):
system of canals and waterways, right. So some of the trip,
like about two days of that trip, we were on
these massively commercial waterways and there might have been a
container ship that was many, many tons right next to us, right,
And we have some picture taken right out of the
window of these big container boats and we actually used
(53:00):
the same locks that they did, and they would kind
of throw these these ropes down from way high on
the wall, and you know, we were kind of getting
blown off the wall and kind of some embarrassing stories
we're told in that book, because you know, we're not big,
big boating people. But then there are also these hand
operated locks right that that are a lot like the
(53:21):
Eery Canal, where the Eery Canal is operated mostly by engineers.
Well in Canada, these are operated by parks and wreck
people that are trained in hospitality and then they're taught
to do the canal work, right. And so we we
had a tremendous time cutting across the Canadian system with
these these canals. And then there's this place called Georgian
(53:44):
Bay and what's called the Small Ship Channel, where you
have some of the most rugged and beautiful terrain in
all the world. Right, so if you were to google
for the most beautiful cruising waters, one of the things
that would come up would be Georgian Bay. And we
spent a good like two or three weeks there.
Speaker 4 (54:02):
Are you only on the boat? Do you ever fly
a plane or no? Because I feel like most tech
guys when they start to get towards, you know, towards
some time in the career there they tend to be flying.
Speaker 2 (54:12):
Do you ever feel never have? And there's there's a
reason for that. You know, I'd consider getting my pilot's license,
but I am not a checklist kind of person. And
if you want to fly, you really really have to be.
And you know that's that's the same reason I don't
climb right, is my my brain doesn't doesn't click that way.
So I do a little bit of gem climbing, but
(54:34):
I don't do like outdoor high rock climbing because I
I forget little things right, and you can't. You can't
be a pilot and do that. And actually in Miami,
there was a man that said, Bruce, get a mission.
You got to get a boat like currently here. If
you change the mission, you got to change the boat.
And so we heard that and we thought about that,
(54:56):
and we sold our boat after the loop, and then
someday we're kind of looking around and someday we'll get
another boat that is more tuned for just cruising on
the on the Tennessee River. Right, we don't need something
that's made to do a great loop anymore. We need
something that's uh, that's just for.
Speaker 1 (55:14):
Yeah, it's pretty cool. I haven't checked out this book yet,
but I have bought it. It's on my list of
books to read it's it's really cool.
Speaker 3 (55:22):
Alan. Do you have anything else you want to come?
Speaker 4 (55:24):
Oh, I mean there's always tons of stuff I can
ask Bruce, but there does there does have to be
an end to this one. I mean, nothing's really sticking
out at the moment. Just I've been pretty happy with
the material that he's been putting out.
Speaker 3 (55:35):
Uh.
Speaker 4 (55:36):
And and you do kind of do these books with
other people, which is which is great. Where you don't
have a typical writing partner, which is good. But you
do have your name a lot of things. So maybe
that's your strategy. It's kind of tatching your name into
many things to show up get higher than that James
in the search when you're competing with him.
Speaker 3 (55:51):
Is that right?
Speaker 2 (55:52):
Oh, I'm not. I'm not trying to to compete with anybody.
I just like, uh, this human thing, right, it's a
it's all about people.
Speaker 4 (55:59):
I guess I can kind of most And my last
question with this is you know you just I think
you said you just went to code Beam. Where's your
next kind of public outing? I know what you call that?
The next place you're coming out in public, next kind
of conference or something like that you've planned.
Speaker 2 (56:12):
I am not traveling for the rest of the year.
I'll be going to a small conference in Portugal next
year next April, and then we have gig City Lakser
in May.
Speaker 4 (56:20):
That's it, all right, that was pretty easy. I think
you have a much much more. That's your schedule, but yeah,
that's my.
Speaker 1 (56:27):
Said Bruce is did we cover everything they want to cover?
Is there anything else would you like to kind of
pitch before you go to the picks?
Speaker 2 (56:34):
That's it for me, so thank you so much for
inviting me on. I've enjoyed this tremendously. Can't believe it's
been so long.
Speaker 1 (56:40):
Yeah, I know, that's crazy. So I guess I'm going
to transition as two picks. Alan, do you want to
go first? Do you have any picks today? Wow?
Speaker 4 (56:47):
I didn't even think about it, pick, it's been so long.
I guess the other thing I can pick is I've
been still playing around with this Ash framework and I
still think it's a great thing. You know, Bruce mentioned
about flop. We used to use a lot of flop,
but Ash actually does add a flop like thing to it,
So I find myself adding less and less plugins nowadays.
So if people haven't tried it out, I highly recommend
just play around with it.
Speaker 3 (57:07):
Nice.
Speaker 1 (57:08):
Yeah, I hadn't heard a flop until now, like really cool.
I feel like there's been a lot of frameworks that
do things in this direction, filtering, sorting, pagination, and yeah,
ash is a good call out. I'm yet to convince
a company to use it, but I have used it
on Alan since the recommended It's really cool.
Speaker 3 (57:27):
I guess a pick from me.
Speaker 1 (57:29):
I forgot who goes first the host, so I guess
I'll just go before I guess. I have these headphones
that I'm wearing right now. They're pretty interesting to the
company's shocks, and the model is open run, so they're
like not they don't go in second ears right, which
I don't. I'm not a fan of them. But they're
also not like headphones, like they're blocka ears. You just
(57:49):
put them on the side of your ear and it
transmits down through bone transmission. It's sae if I Google
research this, but it's pretty interesting. I did not know
this existed. It's really cool because you can listen to
you know what you were listening, but still be aware
of your surroundings. So I was always like a little
nervous about that. So yeah, to check check these out
(58:10):
shocks open run. I'll link it in the notes of
the show. Yeah, that's that's it for my pick today. Bruce,
do you have any picks for us?
Speaker 2 (58:18):
Yeah, I'm going to go with our Roxio for off.
It's a Black Friday sale we do. We do this
once a year and it's going to be all all
November and so come on, come on. Yeah, we have
an interesting way of teaching.
Speaker 3 (58:32):
Great.
Speaker 1 (58:32):
Well, I think that's it for today. Then, thanks so
much Bruce for coming. It was great chatting with you.
Speaker 3 (58:37):
After a while. And yeah, we'll see all your listeners
probably next week hopefully. I don't know.
Speaker 1 (58:43):
We don't have any guests scheduled, but we haven't been
recording as often as we used to.
Speaker 3 (58:47):
But yeah, we'll probably see you guys. So thanks everyone.
Speaker 2 (58:50):
Take care,
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
CrimeLess: Hillbilly Heist
It’s 1996 in rural North Carolina, and an oddball crew makes history when they pull off America’s third largest cash heist. But it’s all downhill from there. Join host Johnny Knoxville as he unspools a wild and woolly tale about a group of regular ‘ol folks who risked it all for a chance at a better life. CrimeLess: Hillbilly Heist answers the question: what would you do with 17.3 million dollars? The answer includes diamond rings, mansions, velvet Elvis paintings, plus a run for the border, murder-for-hire-plots, and FBI busts.