March 27, 2021 • 31 mins
Side projects can be daunting. You have a brilliant idea in your mind, but how do you get it into code? Kenton and Darren, both current uni students and side project veterans are here to give you guidance on how to start, finish and maintain your side projects.
Darren Fu: "Hi everyone, my name is Darren and I’m a very-late-year software engineering and maths student at UQ. Along the way I’ve worked on some cool things in web development, data analytics and machine learning, and also co-created the timetabling app UQ Planner. I go to uni on Mondays and Fridays, and you can find me on the UQCS Slack as @dr2n."
Kenton Lam: "Hi, I'm Kenton Lam (@k on Slack). I'm in my 4th year of computer science and maths at UQ. I've tutored some courses, done some professional web dev and other software things. Along the way, I've made a bunch of projects to solve problems big and small."
Links:
Kenton’s Blackboard enhancements: https://github.com/kentonlam/Blackboard-Enhancements
Kenton’s GitHub: https://github.com/kentonlam
Darren and Will’s UQPlanner: https://github.com/WillCS/uqplanner, https://www.uqplanner.app/
Darren’s GitHub: https://github.com/Dr2n
UQCSBot: https://github.com/UQComputingSociety/uqcsbot
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Matt (00:00):
Hi everyone.
And welcome to the router, theofficial podcast of the UQ
computing society, where weexplore the human side of tech.
I'm your host, Matt.
And I'm joined today by twomembers of the UQCS committee,
Darren and Kenton.
Today.
They're going to be taking usthrough their experiences,
working on side projects and howyou too can develop your own
little side projects with someguidance and lessons along the
(00:22):
way.
Hope you enjoy.
All right.
Hello, Kenton.
And Darren, how are you both?
Darren (00:35):
Hello?
I'm good.
I'm doing good.
Thanks.
Kenton (00:38):
I'm good.
Thanks as well.
Matt (00:39):
All right, wonderful.
Um, so before we begin talkingabout side projects and all that
, uh, that'd be good if you twocould, uh, I guess introduce
yourselves to, uh, to thelisteners for anyone who's not
already familiar.
Uh, Darren, if you want to gofirst.
Darren (00:55):
Okay.
Uh, hi everyone.
My name is Darren, uh, Darren Fuand I'm doing a degree in
software engineering and maths.
Um, you might have heard of me,uh, before, uh, for the UQ
Planner project that I did, uh,earlier last year.
Kenton (01:16):
Okay.
Hello everyone.
My name is Kenton, or you canfind me at K on Slack.
That's the single letter K I'min my fourth year, of a computer
science and maths dual degree.
Um, you might've seen me aroundin courses or I've also tutored
some courses, most interestinglyalgorithms and data structures.
(01:40):
Uh, and I've worked at someplaces doing web dev and other
software things as well.
Matt (01:45):
All right, wonderful.
Okay.
So I guess, so I guess you'veintroduced yourselves and
you're, you're both unistudents.
So I guess one of the keycharacteristics of a university
is procrastination.
Um, and I guess one of the wayswhere you can have productive
procrastination is throughside-project.
So, um, I guess, firstly, I wantto know what sort of side
(02:09):
projects you've done.
So I know Darren, you mentionedyou've done in timetable
planner, and I think Kentonyou've done something similar as
well.
Um, but throughout your whole,your whole, you need agree.
Uh, I guess from the verybeginning, if you can remember
like your first, your firstlittle project that you did, uh,
I'd love to know where, where itall started.
Kenton (02:31):
Yeah.
Um, I think I can talk aboutthis actually.
Um, my projects tend to be notvery creative.
Um, for me they're just aboutsolving problems or nits that I
have with things in myday-to-day life.
Um, when I first started uni, Ithink it was the end of first
(02:52):
year or something.
One of these nets in myuniversity life was Blackboard.
Blackboard is super kind of hardto navigate.
Uh, so the first thing I madewas actually a little search
engine for Blackboard.
Darren (03:07):
Yeah, yeah,
Kenton (03:08):
Yeah.
You might've seen it or evenused it yourself.
Darren (03:12):
Is this still up?
That's that's very interesting.
I'd like to see it now.
Kenton (03:17):
I'll be sure to, uh, let me know where you can find it.
Darren (03:20):
I'll link all this stuff in the description, but I do
remember it was like a sort oflike search box over there.
Yeah, exactly.
Yeah.
Kenton (03:29):
So this uses a script, which is kind of like a Chrome
extension, but more dodgy opensa search box and you can search
all your courses with a singlekeyboard shortcut on the
Blackboard and it was actuallysuper useful and a couple of my
friends used it.
Um, so I think that was thefirst one I did.
And it kind of set the tone forthe rest of my side projects
(03:50):
through my university career.
Um, yeah, I think Darren, whatabout you?
Yeah,
Darren (03:56):
Very nice.
Uh, can I ask you whether youstill use this, uh, Blackboard
extension today?
Uh, that sounds ultra useful,
Kenton (04:04):
Unfortunately, not talk about this later if needed, but
it was my first encounter withJavaScript and it was horrendous
actually terrible.
Darren (04:17):
Yeah, no.
Um, so from my end, I thinkgenerally it's a little bit like
what you said Kenton.
Like when I do a side project,it's generally if, uh, I want to
see something or there'ssomething that I'm kind of
annoyed at.
Um, often I think probably thesmallest version of that is
would probably just be, uh, allthe keyboard shortcuts that I've
set up for my computer.
(04:37):
Um, just to make it easier formyself to sort of, uh, go about
my day, um, doing thingsnavigating around it, all that.
Um, but yeah, I think, uh, whenI initially started, uh, I know
Matt, you talked about, uh, whenI first started university, but
I think it was side projects.
Uh, I first started, uh, when Iwas sort of just in high school.
Um, and I was really curious tolearn about, uh, programming,
(05:00):
coding, uh, electronics and allthat.
So I'd always try and find timeto, um, do some of my own things
to sort of get a better, abetter understanding of, uh, of
the different technology outthere.
Um, and I remember when I firstkind of learned to program, uh,
I made lots of little games, uh,things like, um, you know,
doodle jump.
Uh, I wrote that with JavaScriptand that was a lot of fun.
(05:22):
Uh, this is a little, this was awhile ago now often it's like,
uh, what do you call it?
Uh, you know, the game 2048?
Um, yeah, I made that one timeand that was also a lot of fun.
And I think when I first startedprogramming, it was these kinds
of small projects that reallygot me motivated in, uh,
learning how to, uh, how towrite code.
And yeah, it was, it was a lotof fun.
(05:43):
Um, but yeah, these days it'susually things that sort of
annoyed me a little bit, um,that I want to, uh, I, I want to
fix up and probably, I thinkthat the best example of that
would be, um, would be the UQtimetable planner that I've made
, um, which has just reallystarted as an annoyance, um, of
me going, like, why have theyremoved this?
Um, can I try and like replacethis with something else that I
(06:04):
can use?
Um, and that that's sort of howI started it started.
Yeah.
Matt (06:08):
Hmm.
And then did you make any, anygames?
Any,
Kenton (06:14):
No, unfortunately I was not creative enough to make any
games only boring things for me.
Matt (06:20):
Uh, no, that's all right.
I mean, yeah.
I mean, personally for me, Ithink I had a similar path to
Darren.
I did well, they weren't sideprojects, but like projects.
Like I, I played a lot of them.
Yeah.
So I played this, uh, this gamecalled step mania, uh, which is
like, uh, you can imagine it'slike Dance Dance Revolution, but
(06:40):
on a keyboard.
Um, and one of the very firstside projects I did or kind of
side projects, this is like, I,I, there was this theme, uh,
that I really liked using it wascalled simply love.
And it was like this like istheme where you can like
customize all the differentcolors and like the gameplay and
what it looks like andeverything.
(07:00):
And I made my own little versionof it for like, with like
scoring features and things likethat.
And it was all written in Lua,but I didn't actually write much
Lua.
I just looked at the existingcode and like guessed how to,
how to change it so that itwould work.
Um, but yeah, that was fine.
In fact, I didn't even reallythink about that, but
Darren (07:21):
So that kind of thing, uh, almost all started side
projects kind of start small.
And when you're starting at thevery beginning, there's no
intention of saying that allthis is going to end, and this
isn't even a requirement for aside project quite like this, no
requirement to say, Oh, this isgoing to like, uh, become a huge
thing or like it's going to beused by all these people.
Right.
Um, and then when it's in thatstate, like, there's nothing
wrong with saying, Oh, that's areally cool style that I want to
(07:43):
try and mimic.
Let's see how, how, how close Ican get to it.
Um, and often like, uh, justgetting to that point, right.
Can be really, really rewarding.
Matt (07:50):
Yeah.
And like, I guess, liketinkering with things that
already existed as a good way oflike getting comfy with like
tech and stuff.
So it doesn't even have to belike a, a brand new thing, but I
guess like, so, so like, I thinkboth of you, you've done things
that are like from scratchthere.
Right.
So like the UQ Planner I'mguessing, or some scratch and
Kenton your, I think a few ofyour projects are just like,
(08:12):
from the very beginning, right.
You had an idea and you're like,alright, I will do it.
Um, so I guess the first, one ofthe first decisions, like, okay,
so how do you, how do you evenlike start, like when you have
an idea, what do you do?
Like you just go something up orto sketch it out or,
Darren (08:34):
Um, I've got a list.
Uh, I've got a list of thingsthat I want to do when I get the
time to yes.
Kenton (08:40):
I think definitely the first thing to do when you have
an idea is to write it down
Darren (08:46):
Because these, uh, sorts of, I think, uh, as I can call
them sort of like nitpicks thatyou usually find right there,
probably the smallest of thingsyou just kind of come across, uh
, as you're doing something elsefor uni or for work or whatever.
Right.
Um, and then if you don't writethem down, I mean, easily just
forget them.
Um, and so it's good to keep alist.
Uh, and then when you've got thetime, you can come back and go,
(09:08):
hold, hold on.
Uh, that might be worth doing,uh, and then that's the start of
your side project there.
Matt (09:13):
Um, and so like another thing, I guess, so when you,
when you, when you, when youhave an idea, uh, and you're
like, I want to do this now, orwhat's your, what's your next
step?
Darren (09:27):
That's a hard one to answer Kenton.
You have you got anything to,uh,
Kenton (09:32):
Um, yeah, it is a hard one to answer, but there are a
few tips.
I think, I think a lot of thetime, if you're asking about
choosing, like what do you evendo?
Um, it's constrained by what theidea is.
So for example, if you want todo something, um, or just run
with my example of Blackboard,then it needs to work with
(09:55):
Blackboard.
So that pretty much means it hasto be JavaScript and it has to
run into browser.
Um, once you have like the basicconstraints set up, you can
start Googling.
Google is your best friend
Matt (10:10):
Or Bing or, duck, duck, or your search, internet choice,
internet indexer of choice.
Kenton (10:19):
But yeah, you can look up like, how do I run
JavaScript?
How can I run custom JavaScripton a webpage and pretty quickly
you'll end up at user scripts ormaybe even browser extensions.
And then from that, it's just,um, one step at a time you might
download a basic sampleextension and try to get it to
(10:42):
load, try to get it to dosomething small, like print its
existence to the console.
And then it's just baby steps,one Google query at a time.
How do I add a button?
How do I search for text on apage until you build it all
together into something useful?
Matt (11:02):
So, so I guess, so your, your experience continue.
It's not like you didn't go andlook up an entire tutorial on
how to write user scripts fromthe beginning and then, uh, do
it learn it and then implementedafterwards, right.
You just kind of did it at onestep at a time.
Kenton (11:19):
Yeah.
And our, um, this may just be mypersonal style, but I tend to do
poorly learning from longtutorials that try to teach you
everything the, the, or learnthings when they come up, uh,
because then I can immediatelyapply them.
And I just find that it's moreuseful for myself, but there are
lots of tutorials for all sortsof different things.
(11:41):
If you prefer learning that wayalso videos, I'm not a fan of
videos, but they exist.
Lots of them learn JavaScript
Matt (11:49):
In 16 hours.
Isn't it
Darren (11:51):
Exactly.
40 minutes.
I've seen some videos like thatReact in 40 minutes or something
like that.
Um,
Matt (11:59):
So for like a, for like a quick, for like a quick
refresher, I guess
Darren (12:04):
I've seen some where they even try to go from the
very beginning, all the way toadvanced concepts.
Um, and 40 minutes, I've, I'venever tried to use a tutorial
like that.
Um, but yeah, I, I agree withKenton in that if you're
starting a project and you'retrying to follow a tutorial, it
might make sense to do the rowtutorial, to learn about one
component of the project thatyou're doing to get a better
understanding of that tech orsomething like that.
(12:25):
Um, but if you're sort of doinga tutorial, often, I find that,
uh, it's a little bitconstraining that I guess that
the tutorial is the, uh, thething that's steering the
direction of your project ratherthan, um, what you're thinking
yourself.
Yeah.
Um, but yeah.
Anyway, hold on.
What was the original question?
Matt (12:46):
I mean, I guess, yeah, so I guess Kenton kind of answered
it, but in his own way about howto choose a technology and how
you're, you're constrained by,by the, um, by the constraints
Darren (13:03):
I was going to say.
Yeah, yeah.
Well, we can say, yeah, no.
Um, I, I remember originallyyou've asked, uh, once, you
know, the project, what are yougoing to do afterwards?
Right.
Uh, what the next step is, andbasically said, you start from
Googling.
Um, and then eventually you justhave the project unfold from
there.
Um, and I think that's basicallyexactly right.
And the, the, the, uh, the twothings that I thought that I'd
(13:24):
add is first, uh, you can usesort of knowledge that you
already have.
Uh, what do I know about, um,the system, this, uh, this thing
that I want to try and getworking.
Right.
Um, and how can I sort of get asclose as I can to the objective
first and then keep going.
Right.
Um, and then they'll one, andactually, this is part of the
story for UQ planner wasactually just tell people, uh,
if you don't know how you mightget it to work, go and tell
(13:46):
people, Hey, I think that, um, Ireally want this.
Right.
Um, and, uh, I, I don't know howhard of a project, but it might
be that it might be, um, but Ireally want to do it.
And I think this is somethingthat's really worth making.
Um, and that's actually whathappened to me.
Um, we just, uh, sort of had agroup chat, um, me and, uh, me
and will, and in that chat, Isaid, Oh, I can't believe
they've taken the planner down.
(14:08):
Right.
Um, and then Will happened tosee that message.
Anyone who said, Hey, hold on.
Uh, I was doing a course lastsemester and they made a really
basic version of a timetableplanner.
Uh, do you think we could maybeuse that, uh, and then sort of
get somewhere from it.
Um, and that's where it like,sort of eventually unfolded from
there.
Right.
And so the third one would be,uh, asking people and then
(14:28):
seeing what, what happens there.
Matt (14:32):
Mm.
Yeah.
And, and I guess so, so when,when you, when you've done well,
I guess, like, do you have a,let's say you start a project
from scratch.
Right.
So, so a lot of these projectsare kind of web-based right.
Darren (14:47):
Often.
Yeah, I guess often.
Yes.
Uh, um, do you have a stack ofchoice that you jumped to?
I guess, like, obviously if itwas a user script, then, you
know, you're limited, but ifit's your own application, uh,
do you, do you prefer a certainfront end framework or anything
like that?
Um, at least for me, um, when westarted off the planner Will's
(15:10):
project was already in Angular.
Um, and, uh, I've got, uh, uh,some experience using Angular.
And so let's, let's keep goingfrom there.
Um, but then I dunno, today I've, uh, if I were to start the
project again today, I wouldsay, Oh, I've got more
experience in React.
So I might start with that.
Um, but really, I think, uh, itmight depend on sort of just
(15:31):
what you want to learn, what youmight feel most comfortable in.
Um, and then if you were tryingto try to think about the
project longterm, right.
Uh, if, if you want to sort ofkeep this up for a while, uh,
what's the technology that I canuse to make this the most
maintainable for me, uh, and,and the cheapest to, uh, to sort
of run.
Matt (15:50):
Yeah.
Kenton (15:51):
Yeah.
I think, um, yeah, definitely,uh, a lot of personal projects
of mine have been web deployed,so that deployed, uh, onto the
website and definitely React andTypeScript is probably the
easiest stack to work with there.
(16:11):
And Darren has a really goodpoint, which is you should try
and make it maintainable becausetoo often I feel like personal
projects or side projects end upbeing temporary, or just like
you create it once and youforget about it.
Um, which is a shame becausethese are useful things and it
would be nice if they stayedaround so more people can use
(16:32):
them.
Uh, so I think on that note, itwould be good to avoid things
like having to run your ownserver.
Yeah.
And there are lots of toolsavailable, which let you do just
that, like serverless is prettybig right now.
And it works very well withReact and TypeScript.
(16:53):
And there are things likeGoogle's Firebase, which will
give you cloud functions to runcode on the cloud.
They give you a database, theygive you authentication and a
whole bunch of things, whichwere pretty generous free tier
and it just does all of thatwork for you.
And the best part is you canjust leave it there and it'll
(17:13):
keep chugging along whenever itneeds to.
And you don't need to worryabout your server crashing.
You don't need to worry aboutmaintaining your own database.
And it's just pretty cool.
I think it can help to create aside-project which lives a bit
longer than the average one.
Matt (17:31):
Yeah.
I I've used, I've used Firebasemyself and I love it because
it's just, so yeah, you can, youcan kind of just leave it, leave
it alone.
Um, and it will just do itsthing.
Um, have any of your projectslike needed or have you, have
you, have you ever gone back toyour projects and been like, I
need to fix something?
(17:52):
Or is it just, yeah,
Darren (17:54):
I think on that one, actually one, one good thing to
add is that if you're writingcode, um, make sure to, I guess
you kind of know, uh, or you,you might have the feeling at
the beginning that you don'tknow whether this is going to go
anywhere.
And so you don't, you aren'treally that inclined to write
really clean code or really, uh,maintainable code.
Um, but I think one lesson thatI've learned is that if you
(18:16):
don't, uh, it will probably comeback to bite you.
Um, and so do try to write nicecode, um, because you don't know
whether you'll have to come backto this in a semester or two.
Um, and, uh, yeah, like, uh, thebest way to make them
maintainable is to make surethat everything you write is
sort of, uh, nice and clean.
You don't have things allscattered across different files
in a way that's completely, uh,or like really hard to manage or
(18:38):
anything like that.
Um, yeah.
Um, always good to sort of keepyour code maintainable.
Matt (18:45):
I guess everything is temporary until it's not exactly
everything is temporary untilit's not.
That's good advice, um, on thatnote as well.
Oh, sorry.
What were you saying?
Darren (18:55):
Oh, uh, I think I also had some things to add about
what Kenton was saying aboutsort of making sure that, uh,
your project is scalable.
Um, at least if you can plan, orI can say that, uh, I've used a
service called Netlify.
Um, and because when you'rebuilding a react application, uh
, because it's JavaScript, it'sentirely run from, uh, from
inside the browser, right.
Um, you don't need a backend atall, which means that, um, there
(19:18):
are services like Netlify, whichpurely serve the assets of your
app statically.
Right.
Um, but don't actually sort ofdo any calculations in the
backend.
Um, they serve, they purelyserve the files of, of your app.
Um, you've got lots of serviceslike that.
Um, like Netlify, that can serveyour app for free.
Um, and then you can use them.
(19:38):
Uh, so that's not very clear,uh, know you've got, you've got
services, like Netlify that, uh,serve your app for free.
Um, and then you can basicallyjust run them without worrying"I
have to go running into fees".
Um, and then the second thingwas, uh, cloud functions.
And then with, uh, the UQplanner, uh, what I actually did
was I actually looked at all ofthe cloud service providers, all
the, uh, all the providers forLambda functions.
(19:59):
And I found that CloudFlare, uh,was a service provider that gave
the maximum number of requests,but, uh, for free per day.
And that was the reason why Iuse them.
Uh,
Matt (20:11):
I think actually I think that part of the question, I
guess, have you ever gottenclose to hitting the free
requests limit on CloudFlare forUQ planner?
It's pretty popular.
Darren (20:23):
Um, it actually has.
Um, but Oh, that was actuallybecause of a bug.
Uh, it was, it was quite funny.
I don't remember the exactdetails of the bug.
Um, but ah, yeah, I remember,um, so I think in the, in the
second semester that weintroduced, uh, the planner, uh,
(20:44):
we had this idea that, um,because the times are often
changing even after you'veinitially loaded them, um, that
would basically sort ofautomatically make requests to
try and see if changes werebeing made and then sort of
update them in the front end,uh, automatically.
Um, and then when we first madethat change, I can't remember
exactly what it was all we'redoing in the front end.
(21:05):
Um, but what ended up happeningwas that, uh, all of our clients
were basically making a ton ofrequests to a Lambda function,
uh, requesting to see where thechanges had been made to any of
the course times.
Uh, and then, because we hadtons of computers running this,
this front end, uh, roundaboutwhen we had course registrations
and things like that.
(21:25):
Uh sign-ons um, we hit that, uh,free tier limit pretty quickly.
Um, and then I think we ended uppaying something like$5 or
something around that, uh, justto, to open up the limit a
little bit.
But I think that was the onlytime when we ran into, uh, into
that problem.
There were a few times when Ithink we got close, uh, under
(21:46):
normal usage conditions, but,um, generally I think, I mean,
if you've got a university with,uh, 20, 30, 40,000 people and
that doesn't happen too often,it's not something that you need
to worry about too much.
It's pretty cool that you canhave a, have essentially the
service that so many people useand it costs nearly nothing to
run.
(22:07):
Honestly, it's kinda crazy.
I, I think, um, I think it's,it's a really good deal, uh, and
yeah, kinda kind of one reasonwhy you can say that, um, if
you're sort of launching aproject, uh, you don't have to
worry about the cost too muchlater on.
Uh, if you sort of think aboutthe way that you architect a
little bit at the beginning, um,you, you likely won't even have
(22:28):
to pay anything at all,especially if you can get all
the work being done, uh, in theclient and in the browser
instead of, uh, in some backend.
Matt (22:40):
Um, I guess one last thing I, I wanted to ask, um, is about
when, when you have done aproject, uh, do you, do you have
any, I guess, ways of testing itor, or getting the word out
about your project, anypreferred methods of, uh,
shilling, I guess,
Kenton (23:00):
Yeah, I'll go first on this one because my shilling is
much less effective thanDarren's.
I think, I don't know.
Darren (23:10):
I just want to say that I think at least with the
planner, uh, we just very muchhappened to be in the right
place at the right time.
Um, I don't really think thatthere was anything in particular
that I did that sort of, Iguess, like made the
differences, especially in termsof shilling.
Uh, yeah.
Uh, but yeah, continue.
Sorry.
Kenton (23:28):
Yeah, I was just saying that, um, I think if you want to
test, it's a good idea to reachout to some friends who might
find benefit from what you'vemade and they're always able to
give feedback and they won't betoo mad at you if something goes
wrong.
Um, beyond that#projects on theUQCS Slack is super cool.
(23:53):
There are lots of people talkingabout their own projects, uh, so
you can share it there and youcan get some feedback.
And also, um, if you have anyquestions, that's also a good
place to go and get help fromexperienced people in the UQCS
community.
Uh, but apart from that, I don'treally share my projects out
that much.
Matt (24:14):
They don't have to be shared.
Right.
Like, I guess it, as long as itsolves a problem that, that you
wanted to solve, I guess.
Kenton (24:22):
So that's my thinking.
Like I make a lot of these forme and if it works good enough,
it's good enough for me.
Matt (24:31):
Um, Darren, I guess, do you want to say anything?
Darren (24:35):
Uh, yeah.
Yeah, I guess from me, uh, uh,once you're sort of at a point
where with a project where youthink, Oh, I really want to get
this out there.
Let's see what happens.
Let's give it to some people.
Um, you can put it on, uh,places like Facebook, uh, Reddit
, um, groups like that, coursespace, stalker space, um, things
like that.
Uh, I think these days, uh,maybe even on a discord, you can
(24:57):
see a Slack and it's just areally good place to, to, uh,
get word out there on yourproject.
Um, but then other than that, I,I really agree with what Kenton
said, um, about sort of askingyour friends, um, it's a really
good way to sort of, if you wantto eventually get your project
really out there.
Right.
Um, or sort of, uh, have itbecome something that, uh,
people would actually go anduse.
(25:18):
Uh, it's really good to just goand ask people, Hey, uh, I've,
I've gone and done this.
What do you think about it?
Um, how do you think I couldimprove it?
Um, and just ask questions likethat.
Um, because I've done thatbefore with a lot of the
projects that I've done and Ithink that, um, it's definitely
something that's been reallyhelpful and usually people are
more than, more than willing tohelp out.
Kenton (25:38):
Yeah.
I think that, especially ifyou're going to go and try and
get lots of people to use it,you want to make sure that it
works pretty well before you didthat because you only have one
chance to make a firstimpression.
Darren (25:51):
Yeah, I, a hundred percent.
Um, I think, uh, yeah, with theplanner just before we launched,
uh, we were so stressed out thatthere'd be issues of it or
things like that.
Um, and, uh, it turned out thatwe'd done our homework, we
tested it well, and, uh, not toomuch happened, uh, not, not too
much, uh, sort of disastershappened, uh, on, uh, on the, on
(26:13):
the day that we started to poston Facebook and everything like
that.
Matt (26:17):
Yeah.
I guess you can never expectwhat happens when you get
bombarded with traffic and newusers and all that kind of, it's
just, whatever happens, happens,I guess.
Exactly.
Yeah.
You can try your best to preparefor it, but yeah.
Um, and I, I dunno, I don'tthink I have really anything
else, um, that I, I wanted toask in particular, but as a, as
(26:41):
a final thing, do you have anyadvice for the first years or
maybe second years or even Xyears?
You just want to try out someside projects for the first
time, any
Kenton (26:52):
Hmm.
I dunno for me, for sideprojects with me.
The most difficult thing hasbeen motivation, well ideas.
And then once you have the ideaof motivation, um, so I think,
yeah, just find some motivation,whether that's, um, friends
telling you that, Hey, this willbe super useful or whether this
(27:13):
is something that you reallywant to solve some problem.
Um, yeah, some kind ofmotivation will go a long way to
making sure that you a finisheda side project.
First of all, because there aretoo many unfinished side
projects.
I have a few, I'm sure everyonehas a hundred something that
they can talk about, but yeah.
(27:35):
And B, uh, that you do it well,because if you care about
something, uh, you'll naturallywant to do a good job and you
want to improve user experience.
So everyone, um, enjoys usingwhat you've made, I think, yeah.
Matt (27:53):
Darren?
Darren (27:56):
Um, yeah, completely agree Kenton.
I think, uh, you should reallytake an attitude where you're
just kind of curious to try andfix things, uh, learn things and
try and help other people.
Um, and then like, you don'tknow when you might end up
creating something that, uh, uh,could be really, really useful
to a lot of other people.
Um, and then on what Kenton wassaying about motivation, uh, one
sort of hack that I've done witha few times is honestly just try
(28:18):
and link it to your courses alittle bit, um, or links to
things where you do have to putin an effort, a certain amount
of effort, uh, every once in awhile.
Um, so that, I guess you, youkind of have a, uh, sort of a
deadline pushing you a littlebit.
Um, I don't know if you've sortof linked it to a course that
it's, it can still be considereda side project.
Um, but, uh, one thing that I'vefound is like, say, for example,
(28:41):
um, I've, I've, I've done theside project where I scraped the
New York times with like, Ithink six gigs of data or
something, and then visualizethe entire thing.
Um, then I linked that to thecourse COSC3000, uh, I think
what was the name of the course?
I think it was
Matt (28:55):
The visualization course?
Visualization
Darren (28:58):
Data visualization.
I can't a hundred percentremember, um,
Matt (29:04):
Visualization, computer graphics and data analysis, I
think.
Darren (29:07):
Yeah.
That's the one, something likethat.
Um, but yeah, um, having, so Iwanted to do that project, um,
or I wanted to do some kind of adata analysis project, um, and
then enrolling myself into thatcourse, uh, really sort of
pushed me to, to keep going withit.
Um, and it sort of gave me theinitial push to get the project
to a point where I was quitehappy with it.
Matt (29:26):
Hmm.
All right.
Well, thanks so much, uh, bothof you and, um, yeah, it's been
really good to hear about yourexperiences as a side project,
and I hope that encourages, uh,our listeners to try of their,
um, and yeah, uh, just one lastword from me, uh, there is a
side project that you shouldtotally try out that's related
(29:48):
to the UQ computing society.
Um, and it goes by the name ofUQCSBot, uh, which I will link,
um, in the, you know, in thedescription.
Uh, and I don't know if itcounts really as a side project,
but, you know, it might besomething worth trying out in
your time that there's a bunchof things that you can, a bunch
(30:09):
of features that you can add toUQCSBot.
You know, if you've ever had agrievance on the UQCS Slack
about something that you wishwas, uh, functionality, uh, I
guess it's worth somethingtrying out on that repo.
So I'll leave that
Kenton (30:23):
UQCSBot is great.
Um, there's always features todo as well as bugs to fix.
Uh, of course I think inparticular, it's great because
it's made by the members of UQCSfor the members of UQCS.
Um, so it's extremely easy toget help, um, on the UQCS Slack,
(30:44):
if you're stuck or don'tunderstand something with the
bot, we're all happy to helpout.
And it's a pretty good place tostart if like it's your first
project as well.
Because as I said, lots ofsupport along the way.
Matt (31:00):
Sounds like a, sounds like a good deal.
Right?
Sounds pretty cool.
I'd like to have a go at it.
All right.
Thanks so much.
Good luck to everyone with yourside projects.
Kenton (31:08):
Yeah, they're great fun.
Darren (31:10):
Happy hacking.
Matt (31:12):
All right.
That's all we have time fortoday.
Hope you enjoyed.
The next episode of the Routerwill be out in a fortnight.
Apologies for the delay on thisone, but we hope it was worth
the wait.
Until then don't forget to joinour Slack community at
slack.uqcs.org.
My name is Matthew Low and thispodcast was created by the UQ
computing society with gracioussupport from our industry
sponsors.
Hi everyone.
And welcome to the router, theofficial podcast of the UQ
computing society, where weexplore the human side of tech.
I'm your host, Matt.
And I'm joined today by twomembers of the UQCS committee,
Darren and Kenton.
Today.
They're going to be taking usthrough their experiences,
working on side projects and howyou too can develop your own
little side projects with someguidance and lessons along the
(00:22):
way.
Hope you enjoy.
All right.
Hello, Kenton.
And Darren, how are you both?
Darren (00:35):
Hello?
I'm good.
I'm doing good.
Thanks.
Kenton (00:38):
I'm good.
Thanks as well.
Matt (00:39):
All right, wonderful.
Um, so before we begin talkingabout side projects and all that
, uh, that'd be good if you twocould, uh, I guess introduce
yourselves to, uh, to thelisteners for anyone who's not
already familiar.
Uh, Darren, if you want to gofirst.
Darren (00:55):
Okay.
Uh, hi everyone.
My name is Darren, uh, Darren Fuand I'm doing a degree in
software engineering and maths.
Um, you might have heard of me,uh, before, uh, for the UQ
Planner project that I did, uh,earlier last year.
Kenton (01:16):
Okay.
Hello everyone.
My name is Kenton, or you canfind me at K on Slack.
That's the single letter K I'min my fourth year, of a computer
science and maths dual degree.
Um, you might've seen me aroundin courses or I've also tutored
some courses, most interestinglyalgorithms and data structures.
(01:40):
Uh, and I've worked at someplaces doing web dev and other
software things as well.
Matt (01:45):
All right, wonderful.
Okay.
So I guess, so I guess you'veintroduced yourselves and
you're, you're both unistudents.
So I guess one of the keycharacteristics of a university
is procrastination.
Um, and I guess one of the wayswhere you can have productive
procrastination is throughside-project.
So, um, I guess, firstly, I wantto know what sort of side
(02:09):
projects you've done.
So I know Darren, you mentionedyou've done in timetable
planner, and I think Kentonyou've done something similar as
well.
Um, but throughout your whole,your whole, you need agree.
Uh, I guess from the verybeginning, if you can remember
like your first, your firstlittle project that you did, uh,
I'd love to know where, where itall started.
Kenton (02:31):
Yeah.
Um, I think I can talk aboutthis actually.
Um, my projects tend to be notvery creative.
Um, for me they're just aboutsolving problems or nits that I
have with things in myday-to-day life.
Um, when I first started uni, Ithink it was the end of first
(02:52):
year or something.
One of these nets in myuniversity life was Blackboard.
Blackboard is super kind of hardto navigate.
Uh, so the first thing I madewas actually a little search
engine for Blackboard.
Darren (03:07):
Yeah, yeah,
Kenton (03:08):
Yeah.
You might've seen it or evenused it yourself.
Darren (03:12):
Is this still up?
That's that's very interesting.
I'd like to see it now.
Kenton (03:17):
I'll be sure to, uh, let me know where you can find it.
Darren (03:20):
I'll link all this stuff in the description, but I do
remember it was like a sort oflike search box over there.
Yeah, exactly.
Yeah.
Kenton (03:29):
So this uses a script, which is kind of like a Chrome
extension, but more dodgy opensa search box and you can search
all your courses with a singlekeyboard shortcut on the
Blackboard and it was actuallysuper useful and a couple of my
friends used it.
Um, so I think that was thefirst one I did.
And it kind of set the tone forthe rest of my side projects
(03:50):
through my university career.
Um, yeah, I think Darren, whatabout you?
Yeah,
Darren (03:56):
Very nice.
Uh, can I ask you whether youstill use this, uh, Blackboard
extension today?
Uh, that sounds ultra useful,
Kenton (04:04):
Unfortunately, not talk about this later if needed, but
it was my first encounter withJavaScript and it was horrendous
actually terrible.
Darren (04:17):
Yeah, no.
Um, so from my end, I thinkgenerally it's a little bit like
what you said Kenton.
Like when I do a side project,it's generally if, uh, I want to
see something or there'ssomething that I'm kind of
annoyed at.
Um, often I think probably thesmallest version of that is
would probably just be, uh, allthe keyboard shortcuts that I've
set up for my computer.
(04:37):
Um, just to make it easier formyself to sort of, uh, go about
my day, um, doing thingsnavigating around it, all that.
Um, but yeah, I think, uh, whenI initially started, uh, I know
Matt, you talked about, uh, whenI first started university, but
I think it was side projects.
Uh, I first started, uh, when Iwas sort of just in high school.
Um, and I was really curious tolearn about, uh, programming,
(05:00):
coding, uh, electronics and allthat.
So I'd always try and find timeto, um, do some of my own things
to sort of get a better, abetter understanding of, uh, of
the different technology outthere.
Um, and I remember when I firstkind of learned to program, uh,
I made lots of little games, uh,things like, um, you know,
doodle jump.
Uh, I wrote that with JavaScriptand that was a lot of fun.
(05:22):
Uh, this is a little, this was awhile ago now often it's like,
uh, what do you call it?
Uh, you know, the game 2048?
Um, yeah, I made that one timeand that was also a lot of fun.
And I think when I first startedprogramming, it was these kinds
of small projects that reallygot me motivated in, uh,
learning how to, uh, how towrite code.
And yeah, it was, it was a lotof fun.
(05:43):
Um, but yeah, these days it'susually things that sort of
annoyed me a little bit, um,that I want to, uh, I, I want to
fix up and probably, I thinkthat the best example of that
would be, um, would be the UQtimetable planner that I've made
, um, which has just reallystarted as an annoyance, um, of
me going, like, why have theyremoved this?
Um, can I try and like replacethis with something else that I
(06:04):
can use?
Um, and that that's sort of howI started it started.
Yeah.
Matt (06:08):
Hmm.
And then did you make any, anygames?
Any,
Kenton (06:14):
No, unfortunately I was not creative enough to make any
games only boring things for me.
Matt (06:20):
Uh, no, that's all right.
I mean, yeah.
I mean, personally for me, Ithink I had a similar path to
Darren.
I did well, they weren't sideprojects, but like projects.
Like I, I played a lot of them.
Yeah.
So I played this, uh, this gamecalled step mania, uh, which is
like, uh, you can imagine it'slike Dance Dance Revolution, but
(06:40):
on a keyboard.
Um, and one of the very firstside projects I did or kind of
side projects, this is like, I,I, there was this theme, uh,
that I really liked using it wascalled simply love.
And it was like this like istheme where you can like
customize all the differentcolors and like the gameplay and
what it looks like andeverything.
(07:00):
And I made my own little versionof it for like, with like
scoring features and things likethat.
And it was all written in Lua,but I didn't actually write much
Lua.
I just looked at the existingcode and like guessed how to,
how to change it so that itwould work.
Um, but yeah, that was fine.
In fact, I didn't even reallythink about that, but
Darren (07:21):
So that kind of thing, uh, almost all started side
projects kind of start small.
And when you're starting at thevery beginning, there's no
intention of saying that allthis is going to end, and this
isn't even a requirement for aside project quite like this, no
requirement to say, Oh, this isgoing to like, uh, become a huge
thing or like it's going to beused by all these people.
Right.
Um, and then when it's in thatstate, like, there's nothing
wrong with saying, Oh, that's areally cool style that I want to
(07:43):
try and mimic.
Let's see how, how, how close Ican get to it.
Um, and often like, uh, justgetting to that point, right.
Can be really, really rewarding.
Matt (07:50):
Yeah.
And like, I guess, liketinkering with things that
already existed as a good way oflike getting comfy with like
tech and stuff.
So it doesn't even have to belike a, a brand new thing, but I
guess like, so, so like, I thinkboth of you, you've done things
that are like from scratchthere.
Right.
So like the UQ Planner I'mguessing, or some scratch and
Kenton your, I think a few ofyour projects are just like,
(08:12):
from the very beginning, right.
You had an idea and you're like,alright, I will do it.
Um, so I guess the first, one ofthe first decisions, like, okay,
so how do you, how do you evenlike start, like when you have
an idea, what do you do?
Like you just go something up orto sketch it out or,
Darren (08:34):
Um, I've got a list.
Uh, I've got a list of thingsthat I want to do when I get the
time to yes.
Kenton (08:40):
I think definitely the first thing to do when you have
an idea is to write it down
Darren (08:46):
Because these, uh, sorts of, I think, uh, as I can call
them sort of like nitpicks thatyou usually find right there,
probably the smallest of thingsyou just kind of come across, uh
, as you're doing something elsefor uni or for work or whatever.
Right.
Um, and then if you don't writethem down, I mean, easily just
forget them.
Um, and so it's good to keep alist.
Uh, and then when you've got thetime, you can come back and go,
(09:08):
hold, hold on.
Uh, that might be worth doing,uh, and then that's the start of
your side project there.
Matt (09:13):
Um, and so like another thing, I guess, so when you,
when you, when you, when youhave an idea, uh, and you're
like, I want to do this now, orwhat's your, what's your next
step?
Darren (09:27):
That's a hard one to answer Kenton.
You have you got anything to,uh,
Kenton (09:32):
Um, yeah, it is a hard one to answer, but there are a
few tips.
I think, I think a lot of thetime, if you're asking about
choosing, like what do you evendo?
Um, it's constrained by what theidea is.
So for example, if you want todo something, um, or just run
with my example of Blackboard,then it needs to work with
(09:55):
Blackboard.
So that pretty much means it hasto be JavaScript and it has to
run into browser.
Um, once you have like the basicconstraints set up, you can
start Googling.
Google is your best friend
Matt (10:10):
Or Bing or, duck, duck, or your search, internet choice,
internet indexer of choice.
Kenton (10:19):
But yeah, you can look up like, how do I run
JavaScript?
How can I run custom JavaScripton a webpage and pretty quickly
you'll end up at user scripts ormaybe even browser extensions.
And then from that, it's just,um, one step at a time you might
download a basic sampleextension and try to get it to
(10:42):
load, try to get it to dosomething small, like print its
existence to the console.
And then it's just baby steps,one Google query at a time.
How do I add a button?
How do I search for text on apage until you build it all
together into something useful?
Matt (11:02):
So, so I guess, so your, your experience continue.
It's not like you didn't go andlook up an entire tutorial on
how to write user scripts fromthe beginning and then, uh, do
it learn it and then implementedafterwards, right.
You just kind of did it at onestep at a time.
Kenton (11:19):
Yeah.
And our, um, this may just be mypersonal style, but I tend to do
poorly learning from longtutorials that try to teach you
everything the, the, or learnthings when they come up, uh,
because then I can immediatelyapply them.
And I just find that it's moreuseful for myself, but there are
lots of tutorials for all sortsof different things.
(11:41):
If you prefer learning that wayalso videos, I'm not a fan of
videos, but they exist.
Lots of them learn JavaScript
Matt (11:49):
In 16 hours.
Isn't it
Darren (11:51):
Exactly.
40 minutes.
I've seen some videos like thatReact in 40 minutes or something
like that.
Um,
Matt (11:59):
So for like a, for like a quick, for like a quick
refresher, I guess
Darren (12:04):
I've seen some where they even try to go from the
very beginning, all the way toadvanced concepts.
Um, and 40 minutes, I've, I'venever tried to use a tutorial
like that.
Um, but yeah, I, I agree withKenton in that if you're
starting a project and you'retrying to follow a tutorial, it
might make sense to do the rowtutorial, to learn about one
component of the project thatyou're doing to get a better
understanding of that tech orsomething like that.
(12:25):
Um, but if you're sort of doinga tutorial, often, I find that,
uh, it's a little bitconstraining that I guess that
the tutorial is the, uh, thething that's steering the
direction of your project ratherthan, um, what you're thinking
yourself.
Yeah.
Um, but yeah.
Anyway, hold on.
What was the original question?
Matt (12:46):
I mean, I guess, yeah, so I guess Kenton kind of answered
it, but in his own way about howto choose a technology and how
you're, you're constrained by,by the, um, by the constraints
Darren (13:03):
I was going to say.
Yeah, yeah.
Well, we can say, yeah, no.
Um, I, I remember originallyyou've asked, uh, once, you
know, the project, what are yougoing to do afterwards?
Right.
Uh, what the next step is, andbasically said, you start from
Googling.
Um, and then eventually you justhave the project unfold from
there.
Um, and I think that's basicallyexactly right.
And the, the, the, uh, the twothings that I thought that I'd
(13:24):
add is first, uh, you can usesort of knowledge that you
already have.
Uh, what do I know about, um,the system, this, uh, this thing
that I want to try and getworking.
Right.
Um, and how can I sort of get asclose as I can to the objective
first and then keep going.
Right.
Um, and then they'll one, andactually, this is part of the
story for UQ planner wasactually just tell people, uh,
if you don't know how you mightget it to work, go and tell
(13:46):
people, Hey, I think that, um, Ireally want this.
Right.
Um, and, uh, I, I don't know howhard of a project, but it might
be that it might be, um, but Ireally want to do it.
And I think this is somethingthat's really worth making.
Um, and that's actually whathappened to me.
Um, we just, uh, sort of had agroup chat, um, me and, uh, me
and will, and in that chat, Isaid, Oh, I can't believe
they've taken the planner down.
(14:08):
Right.
Um, and then Will happened tosee that message.
Anyone who said, Hey, hold on.
Uh, I was doing a course lastsemester and they made a really
basic version of a timetableplanner.
Uh, do you think we could maybeuse that, uh, and then sort of
get somewhere from it.
Um, and that's where it like,sort of eventually unfolded from
there.
Right.
And so the third one would be,uh, asking people and then
(14:28):
seeing what, what happens there.
Matt (14:32):
Mm.
Yeah.
And, and I guess so, so when,when you, when you've done well,
I guess, like, do you have a,let's say you start a project
from scratch.
Right.
So, so a lot of these projectsare kind of web-based right.
Darren (14:47):
Often.
Yeah, I guess often.
Yes.
Uh, um, do you have a stack ofchoice that you jumped to?
I guess, like, obviously if itwas a user script, then, you
know, you're limited, but ifit's your own application, uh,
do you, do you prefer a certainfront end framework or anything
like that?
Um, at least for me, um, when westarted off the planner Will's
(15:10):
project was already in Angular.
Um, and, uh, I've got, uh, uh,some experience using Angular.
And so let's, let's keep goingfrom there.
Um, but then I dunno, today I've, uh, if I were to start the
project again today, I wouldsay, Oh, I've got more
experience in React.
So I might start with that.
Um, but really, I think, uh, itmight depend on sort of just
(15:31):
what you want to learn, what youmight feel most comfortable in.
Um, and then if you were tryingto try to think about the
project longterm, right.
Uh, if, if you want to sort ofkeep this up for a while, uh,
what's the technology that I canuse to make this the most
maintainable for me, uh, and,and the cheapest to, uh, to sort
of run.
Matt (15:50):
Yeah.
Kenton (15:51):
Yeah.
I think, um, yeah, definitely,uh, a lot of personal projects
of mine have been web deployed,so that deployed, uh, onto the
website and definitely React andTypeScript is probably the
easiest stack to work with there.
(16:11):
And Darren has a really goodpoint, which is you should try
and make it maintainable becausetoo often I feel like personal
projects or side projects end upbeing temporary, or just like
you create it once and youforget about it.
Um, which is a shame becausethese are useful things and it
would be nice if they stayedaround so more people can use
(16:32):
them.
Uh, so I think on that note, itwould be good to avoid things
like having to run your ownserver.
Yeah.
And there are lots of toolsavailable, which let you do just
that, like serverless is prettybig right now.
And it works very well withReact and TypeScript.
(16:53):
And there are things likeGoogle's Firebase, which will
give you cloud functions to runcode on the cloud.
They give you a database, theygive you authentication and a
whole bunch of things, whichwere pretty generous free tier
and it just does all of thatwork for you.
And the best part is you canjust leave it there and it'll
(17:13):
keep chugging along whenever itneeds to.
And you don't need to worryabout your server crashing.
You don't need to worry aboutmaintaining your own database.
And it's just pretty cool.
I think it can help to create aside-project which lives a bit
longer than the average one.
Matt (17:31):
Yeah.
I I've used, I've used Firebasemyself and I love it because
it's just, so yeah, you can, youcan kind of just leave it, leave
it alone.
Um, and it will just do itsthing.
Um, have any of your projectslike needed or have you, have
you, have you ever gone back toyour projects and been like, I
need to fix something?
(17:52):
Or is it just, yeah,
Darren (17:54):
I think on that one, actually one, one good thing to
add is that if you're writingcode, um, make sure to, I guess
you kind of know, uh, or you,you might have the feeling at
the beginning that you don'tknow whether this is going to go
anywhere.
And so you don't, you aren'treally that inclined to write
really clean code or really, uh,maintainable code.
Um, but I think one lesson thatI've learned is that if you
(18:16):
don't, uh, it will probably comeback to bite you.
Um, and so do try to write nicecode, um, because you don't know
whether you'll have to come backto this in a semester or two.
Um, and, uh, yeah, like, uh, thebest way to make them
maintainable is to make surethat everything you write is
sort of, uh, nice and clean.
You don't have things allscattered across different files
in a way that's completely, uh,or like really hard to manage or
(18:38):
anything like that.
Um, yeah.
Um, always good to sort of keepyour code maintainable.
Matt (18:45):
I guess everything is temporary until it's not exactly
everything is temporary untilit's not.
That's good advice, um, on thatnote as well.
Oh, sorry.
What were you saying?
Darren (18:55):
Oh, uh, I think I also had some things to add about
what Kenton was saying aboutsort of making sure that, uh,
your project is scalable.
Um, at least if you can plan, orI can say that, uh, I've used a
service called Netlify.
Um, and because when you'rebuilding a react application, uh
, because it's JavaScript, it'sentirely run from, uh, from
inside the browser, right.
Um, you don't need a backend atall, which means that, um, there
(19:18):
are services like Netlify, whichpurely serve the assets of your
app statically.
Right.
Um, but don't actually sort ofdo any calculations in the
backend.
Um, they serve, they purelyserve the files of, of your app.
Um, you've got lots of serviceslike that.
Um, like Netlify, that can serveyour app for free.
Um, and then you can use them.
(19:38):
Uh, so that's not very clear,uh, know you've got, you've got
services, like Netlify that, uh,serve your app for free.
Um, and then you can basicallyjust run them without worrying"I
have to go running into fees".
Um, and then the second thingwas, uh, cloud functions.
And then with, uh, the UQplanner, uh, what I actually did
was I actually looked at all ofthe cloud service providers, all
the, uh, all the providers forLambda functions.
(19:59):
And I found that CloudFlare, uh,was a service provider that gave
the maximum number of requests,but, uh, for free per day.
And that was the reason why Iuse them.
Uh,
Matt (20:11):
I think actually I think that part of the question, I
guess, have you ever gottenclose to hitting the free
requests limit on CloudFlare forUQ planner?
It's pretty popular.
Darren (20:23):
Um, it actually has.
Um, but Oh, that was actuallybecause of a bug.
Uh, it was, it was quite funny.
I don't remember the exactdetails of the bug.
Um, but ah, yeah, I remember,um, so I think in the, in the
second semester that weintroduced, uh, the planner, uh,
(20:44):
we had this idea that, um,because the times are often
changing even after you'veinitially loaded them, um, that
would basically sort ofautomatically make requests to
try and see if changes werebeing made and then sort of
update them in the front end,uh, automatically.
Um, and then when we first madethat change, I can't remember
exactly what it was all we'redoing in the front end.
(21:05):
Um, but what ended up happeningwas that, uh, all of our clients
were basically making a ton ofrequests to a Lambda function,
uh, requesting to see where thechanges had been made to any of
the course times.
Uh, and then, because we hadtons of computers running this,
this front end, uh, roundaboutwhen we had course registrations
and things like that.
(21:25):
Uh sign-ons um, we hit that, uh,free tier limit pretty quickly.
Um, and then I think we ended uppaying something like$5 or
something around that, uh, justto, to open up the limit a
little bit.
But I think that was the onlytime when we ran into, uh, into
that problem.
There were a few times when Ithink we got close, uh, under
(21:46):
normal usage conditions, but,um, generally I think, I mean,
if you've got a university with,uh, 20, 30, 40,000 people and
that doesn't happen too often,it's not something that you need
to worry about too much.
It's pretty cool that you canhave a, have essentially the
service that so many people useand it costs nearly nothing to
run.
(22:07):
Honestly, it's kinda crazy.
I, I think, um, I think it's,it's a really good deal, uh, and
yeah, kinda kind of one reasonwhy you can say that, um, if
you're sort of launching aproject, uh, you don't have to
worry about the cost too muchlater on.
Uh, if you sort of think aboutthe way that you architect a
little bit at the beginning, um,you, you likely won't even have
(22:28):
to pay anything at all,especially if you can get all
the work being done, uh, in theclient and in the browser
instead of, uh, in some backend.
Matt (22:40):
Um, I guess one last thing I, I wanted to ask, um, is about
when, when you have done aproject, uh, do you, do you have
any, I guess, ways of testing itor, or getting the word out
about your project, anypreferred methods of, uh,
shilling, I guess,
Kenton (23:00):
Yeah, I'll go first on this one because my shilling is
much less effective thanDarren's.
I think, I don't know.
Darren (23:10):
I just want to say that I think at least with the
planner, uh, we just very muchhappened to be in the right
place at the right time.
Um, I don't really think thatthere was anything in particular
that I did that sort of, Iguess, like made the
differences, especially in termsof shilling.
Uh, yeah.
Uh, but yeah, continue.
Sorry.
Kenton (23:28):
Yeah, I was just saying that, um, I think if you want to
test, it's a good idea to reachout to some friends who might
find benefit from what you'vemade and they're always able to
give feedback and they won't betoo mad at you if something goes
wrong.
Um, beyond that#projects on theUQCS Slack is super cool.
(23:53):
There are lots of people talkingabout their own projects, uh, so
you can share it there and youcan get some feedback.
And also, um, if you have anyquestions, that's also a good
place to go and get help fromexperienced people in the UQCS
community.
Uh, but apart from that, I don'treally share my projects out
that much.
Matt (24:14):
They don't have to be shared.
Right.
Like, I guess it, as long as itsolves a problem that, that you
wanted to solve, I guess.
Kenton (24:22):
So that's my thinking.
Like I make a lot of these forme and if it works good enough,
it's good enough for me.
Matt (24:31):
Um, Darren, I guess, do you want to say anything?
Darren (24:35):
Uh, yeah.
Yeah, I guess from me, uh, uh,once you're sort of at a point
where with a project where youthink, Oh, I really want to get
this out there.
Let's see what happens.
Let's give it to some people.
Um, you can put it on, uh,places like Facebook, uh, Reddit
, um, groups like that, coursespace, stalker space, um, things
like that.
Uh, I think these days, uh,maybe even on a discord, you can
(24:57):
see a Slack and it's just areally good place to, to, uh,
get word out there on yourproject.
Um, but then other than that, I,I really agree with what Kenton
said, um, about sort of askingyour friends, um, it's a really
good way to sort of, if you wantto eventually get your project
really out there.
Right.
Um, or sort of, uh, have itbecome something that, uh,
people would actually go anduse.
(25:18):
Uh, it's really good to just goand ask people, Hey, uh, I've,
I've gone and done this.
What do you think about it?
Um, how do you think I couldimprove it?
Um, and just ask questions likethat.
Um, because I've done thatbefore with a lot of the
projects that I've done and Ithink that, um, it's definitely
something that's been reallyhelpful and usually people are
more than, more than willing tohelp out.
Kenton (25:38):
Yeah.
I think that, especially ifyou're going to go and try and
get lots of people to use it,you want to make sure that it
works pretty well before you didthat because you only have one
chance to make a firstimpression.
Darren (25:51):
Yeah, I, a hundred percent.
Um, I think, uh, yeah, with theplanner just before we launched,
uh, we were so stressed out thatthere'd be issues of it or
things like that.
Um, and, uh, it turned out thatwe'd done our homework, we
tested it well, and, uh, not toomuch happened, uh, not, not too
much, uh, sort of disastershappened, uh, on, uh, on the, on
(26:13):
the day that we started to poston Facebook and everything like
that.
Matt (26:17):
Yeah.
I guess you can never expectwhat happens when you get
bombarded with traffic and newusers and all that kind of, it's
just, whatever happens, happens,I guess.
Exactly.
Yeah.
You can try your best to preparefor it, but yeah.
Um, and I, I dunno, I don'tthink I have really anything
else, um, that I, I wanted toask in particular, but as a, as
(26:41):
a final thing, do you have anyadvice for the first years or
maybe second years or even Xyears?
You just want to try out someside projects for the first
time, any
Kenton (26:52):
Hmm.
I dunno for me, for sideprojects with me.
The most difficult thing hasbeen motivation, well ideas.
And then once you have the ideaof motivation, um, so I think,
yeah, just find some motivation,whether that's, um, friends
telling you that, Hey, this willbe super useful or whether this
(27:13):
is something that you reallywant to solve some problem.
Um, yeah, some kind ofmotivation will go a long way to
making sure that you a finisheda side project.
First of all, because there aretoo many unfinished side
projects.
I have a few, I'm sure everyonehas a hundred something that
they can talk about, but yeah.
(27:35):
And B, uh, that you do it well,because if you care about
something, uh, you'll naturallywant to do a good job and you
want to improve user experience.
So everyone, um, enjoys usingwhat you've made, I think, yeah.
Matt (27:53):
Darren?
Darren (27:56):
Um, yeah, completely agree Kenton.
I think, uh, you should reallytake an attitude where you're
just kind of curious to try andfix things, uh, learn things and
try and help other people.
Um, and then like, you don'tknow when you might end up
creating something that, uh, uh,could be really, really useful
to a lot of other people.
Um, and then on what Kenton wassaying about motivation, uh, one
sort of hack that I've done witha few times is honestly just try
(28:18):
and link it to your courses alittle bit, um, or links to
things where you do have to putin an effort, a certain amount
of effort, uh, every once in awhile.
Um, so that, I guess you, youkind of have a, uh, sort of a
deadline pushing you a littlebit.
Um, I don't know if you've sortof linked it to a course that
it's, it can still be considereda side project.
Um, but, uh, one thing that I'vefound is like, say, for example,
(28:41):
um, I've, I've, I've done theside project where I scraped the
New York times with like, Ithink six gigs of data or
something, and then visualizethe entire thing.
Um, then I linked that to thecourse COSC3000, uh, I think
what was the name of the course?
I think it was
Matt (28:55):
The visualization course?
Visualization
Darren (28:58):
Data visualization.
I can't a hundred percentremember, um,
Matt (29:04):
Visualization, computer graphics and data analysis, I
think.
Darren (29:07):
Yeah.
That's the one, something likethat.
Um, but yeah, um, having, so Iwanted to do that project, um,
or I wanted to do some kind of adata analysis project, um, and
then enrolling myself into thatcourse, uh, really sort of
pushed me to, to keep going withit.
Um, and it sort of gave me theinitial push to get the project
to a point where I was quitehappy with it.
Matt (29:26):
Hmm.
All right.
Well, thanks so much, uh, bothof you and, um, yeah, it's been
really good to hear about yourexperiences as a side project,
and I hope that encourages, uh,our listeners to try of their,
um, and yeah, uh, just one lastword from me, uh, there is a
side project that you shouldtotally try out that's related
(29:48):
to the UQ computing society.
Um, and it goes by the name ofUQCSBot, uh, which I will link,
um, in the, you know, in thedescription.
Uh, and I don't know if itcounts really as a side project,
but, you know, it might besomething worth trying out in
your time that there's a bunchof things that you can, a bunch
(30:09):
of features that you can add toUQCSBot.
You know, if you've ever had agrievance on the UQCS Slack
about something that you wishwas, uh, functionality, uh, I
guess it's worth somethingtrying out on that repo.
So I'll leave that
Kenton (30:23):
UQCSBot is great.
Um, there's always features todo as well as bugs to fix.
Uh, of course I think inparticular, it's great because
it's made by the members of UQCSfor the members of UQCS.
Um, so it's extremely easy toget help, um, on the UQCS Slack,
(30:44):
if you're stuck or don'tunderstand something with the
bot, we're all happy to helpout.
And it's a pretty good place tostart if like it's your first
project as well.
Because as I said, lots ofsupport along the way.
Matt (31:00):
Sounds like a, sounds like a good deal.
Right?
Sounds pretty cool.
I'd like to have a go at it.
All right.
Thanks so much.
Good luck to everyone with yourside projects.
Kenton (31:08):
Yeah, they're great fun.
Darren (31:10):
Happy hacking.
Matt (31:12):
All right.
That's all we have time fortoday.
Hope you enjoyed.
The next episode of the Routerwill be out in a fortnight.
Apologies for the delay on thisone, but we hope it was worth
the wait.
Until then don't forget to joinour Slack community at
slack.uqcs.org.
My name is Matthew Low and thispodcast was created by the UQ
computing society with gracioussupport from our industry
sponsors.
Popular Podcasts
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Therapy Gecko
An unlicensed lizard psychologist travels the universe talking to strangers about absolutely nothing. TO CALL THE GECKO: follow me on https://www.twitch.tv/lyleforever to get a notification for when I am taking calls. I am usually live Mondays, Wednesdays, and Fridays but lately a lot of other times too. I am a gecko.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.