S9 E13 | Angular SEO 2025 |Jeff Whelpley - The Angular Plus Show

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
M hm.

Speaker 2 (00:08):
Welcome to the Angular Plus Show. We're app developers of
all kinds share their insights and experiences. Let's get started.

Speaker 3 (00:21):
Hello, good good morning, good afternoon, whatever it is for you,
and welcome back to another episode of the Angular Plus Show.
My name is Brian Love. I'm going to be your
host today and I am joined by two good friends, Jan.
How are you today?

Speaker 4 (00:35):
I'm doing very well, and I every time we're on
the site on the podcast together, I'm just always so
jealous of your perfect radio voice, and that's like this perfect.

Speaker 3 (00:45):
It's just yes, it's nice.

Speaker 4 (00:48):
If you ever happen to record an audiobook, just let
me know and I will just play that and repeat
to put me to sleep, and in a very positive
way and the nicest way.

Speaker 3 (00:54):
I mean this, that's wonderful thing to hear.

Speaker 5 (00:57):
Jean.

Speaker 3 (00:57):
I'm so glad that you're here with us. You know,
it's funny. I fell a quick story. I didn't realize
that the first time somebody had told me that I
was doing some courses for Thinkster was my friend Joe,
and he said the same thing. He was like, except Brent,
sometimes I just want to listen to your videos just
on background, because I like you go into this kind
of smooth, very comfortable voice, and he's like, I really

(01:19):
enjoy it. He's like, some good job, keep it, keep
it coming. Anyways, We're also joined by Chow. Hopefully Chow
is kind of dialing in here remotely. That's kind of
the world we live in in twenty twenty five, Chow,
Are you with us? I am with you?

Speaker 5 (01:36):
And I shared a sam sentiment with Nicholas here. I
love your boys, man, So I mean I'm in Yeah,
I mean is in the nicest way possible. Uh, but yeah,
I love when you when your late the podcast.

Speaker 3 (01:52):
Yeah, I'm glad to have you here. I also love that,
like I introed you, I felt like I was dialing
up to like the International Space Station. It's like, we
have you with us. Are you on the line. It's
like the eighties, you know, and like the broadcaster would
like reach out and just hope that like somebody's going
to pick up the other line. And then the way
it sounded over your phone was perfect. It sounded like

(02:13):
you're you're literally an outer space talking to us. But
I know that you're helping out on some family stuff
and we just want to honor that and thank you,
Thank you for being here with us. All right, let's
get right to it. So our guest today is Jeff Welpley.
Jeff Welpley is a Google developer, expert Angular, longtime contributor
to the Angular community, speaker, author, Like, I know you've

(02:33):
just done a million things, Jeff, personally for me, I
remember when I got started in the Angular community around
I think Angler JS, like twenty fourteen, twenty fifteen. I
remember when twenty sixteen I went to n g KOV.
You might have spoken to the conference. I don't remember,
but I think so. I just remember like getting a
chance to say hi to you and and just to

(02:53):
learn from you. So first of all, just thank you
for all that you've done for the Angular community over
the years. Yeah, you're a legend, man, and we're just
proud to have you, like on the podcast today so much.

Speaker 1 (03:04):
I appreciate that. Thank you of course.

Speaker 3 (03:07):
Man. Today we're gonna be talking about Angular and SEO
of Search Engine Optimization. So go ahead and introduce yourself.
I think we've had you on the podcast, I hope so,
but yeah, listener, go ahead and yeah, introduce yourself to us.
And then why don't you kick us off right into
the topic. Let's start talking about Angular and SEO and

(03:27):
twenty twenty five.

Speaker 1 (03:29):
Yeah, so Jeff Welpley, the co founder and CTO of
get Human also, like you mentioned, Google developer expert, long
time in the community, doing a lot of different things
that I'll some of which I'll talk about today, doing
a lot of SEO stuff for over a decade. And
it's kind of funny to talk about Angular and SEO
especially well less so now, but for sure in the

(03:52):
past because they were like oil and water at many
times in the past.

Speaker 3 (03:57):
Is mixed, Yeah, totally or you it was definitely try
to universal.

Speaker 1 (04:02):
Yeah, well, even going back before that funny story, my
first Angie comp actually it was the first Enngie comf
that the very first Angie COMF I was at. I
went to talk to Misco and you know, even at
that time, I was running a company that was heavily
dependent on SEO. Knew a lot about it back then,

(04:22):
even and I asked him because I loved Angular, but
it was not seal friendly at all. It was just
a single page it was all client side, just single
page app everything, And so I asked him, like, well,
is there some way we can modify it react with
starting to work on their server side rendered part, and

(04:44):
his response at that time was along the lines I
don't remember exactly, but it was. It was something along
the lines of like, oh, you know, you don't need that,
Like it's that's just like a short term thing that
is sort of still needed. But you know already the
search engine are starting to crawl client side JavaScript that. Yes, yeah,
everybody's don't worry about it. Like in another couple of years,

(05:08):
it's just going to be all weapon problems and everything. Yeah,
it won't be a problem.

Speaker 3 (05:11):
Yeah.

Speaker 1 (05:12):
And I always continue to joke with Misco about that
when I see him now, because, as you know, he left,
I mean, after all the stuff with Angular and whatever else,
he joined a company building a service side render framework Quick,
yes quick, which is a great framework or whatever. But

(05:32):
it's just so funny that I went to complete one eight.
But yeah, for a long time, Angular and SEO just
not built, not two things that worked well together.

Speaker 3 (05:45):
Let's start there, I think. So like you mentioned that,
and I agree with you. I remember when I first
got started with Angular and I was like, okay, well,
like what does this mean for SEO? What did my
dom look like right, and yeah, once it bootstraps and
the you know, everything's hydrated, like you have this nice,
beautiful you have don with all these elements and you've
got interactivity and binding and closures and everything's happening, and

(06:08):
it's like, oh, this is great. But when you just
like view source, you're just like, oh, like div approu interesting,
Like where's all my content? So tell me a little
bit about like you mentioned that you know, angular and
sea have often not worked well together. Tell me a
little bit more about why that is. And from the listener,
maybe that has is not familiar with that story. And

(06:30):
then let's bridge that into kind of where are we
today and where are things moving?

Speaker 1 (06:35):
Yeah, so I guess there's two main core parts to that.
One is that with ways that search engines have worked
for a long time, until we'll get to how they've
changed drastically in the past year because of AI related stuff.
But for since you know, Google first came out in

(06:56):
the early two thousands or late nineteen nineties or whatever,
or until a couple of years ago, the main focus
was on trying to have the gather signals that the
search engine could determine the user had a good experience
that that basically your website answered whatever question that they had.

(07:21):
And there's a number of ways that they were able
to signals that they used to determine that. A big
one was your formance related stuff. So it was you know,
how quickly content was rendered, how quickly the user saw
the page and was able to interact with it, you know,

(07:42):
a bunch of things around related around that was a
good indication. Search engines that the user probably at least
for the initial load like, had a good experience, and
then there's a whole bunch of other signals like later
on after that one. But just even on that very
first set of signals, anger like failed completely right because

(08:04):
like you were saying, there was nothing. There's usually a
delay the very first time, and then a lot of
jank that would go on even if you tried to
you know for a while. If you remember, there was
like a thought with web components to just put like
a skeleton page and then have it like fill in.

(08:25):
But that was sort of a bad experience too, because
it's like the user thought that they could maybe interact
with it, and then they just got frustrated, like trying
to click around and whatever. So it wasn't until we
got to the point that truly angler was able to
serve a render that those at least those set of

(08:46):
things improved. And the other part of it, just to
talk about the second part of it, is that the
search engines in general, Like a lot of people don't
think about this as much, but the crawler for Google
and for every other search engine and now for every
other AI company, it crawls the web. But it takes time.

(09:08):
It takes time to and resources and ultimately money for
them to crawl the web. And it's cheaper and easier
and faster for them if your site renders faster. So
like they wouldn't explicitly tell you this all the time,
but a lot of the people who are SEO professionals
and like really deep into it would know that your

(09:30):
site would benefit because if you made it easier for
the crawler to crawl your website. And so one of
those things is performance, but there's a whole bunch of
other things to do. It's just like remove those various
impediments because like Misco is trying to say way back then,
like oh they could just client render your app or whatever.
It's like, okay, sure, they technically they do have that,

(09:51):
but yeah, that exactly. It takes forever, and so if
you're relying on that, they're going to definitely not rank
you as higher as other websites.

Speaker 3 (10:01):
If that makes interesting, That does make sense. Yes, I
think to understand kind of your response, which thank you
so much, I think like when it comes to SEO,
I kind of put it like put it into like
two big buckets. You have like off page optimizations, and
those could be like inbound links and even you know,
marketing or paid marketing or organic and that kind of stuff.

(10:25):
And then you have like on page optimizations. And so
when it came to angular and on page optimizations, we
were kind of like pushed to the back of the pack,
right just in terms of the way that the DOM
was rendered, just in terms of the performance. It's time
to interactivity. And it's funny that you mentioned the skeleton thing.
At first, I was thinking of like the ghost skeleton elements,

(10:48):
but I think you were actually referring back to like
there was this idea many years ago of like just
render and then don't hook up all, like don't actually
run the JavaScript yet, but just render the DOM with
your CSS and and then later come back and kind
of wire up all the events and the closures and stuff.
And I think that was something that maybe Mishkov talked
about a long time ago, and I know that fast

(11:10):
forward to today, that's something that's kind of almost easily
to obtain, and I'm interested to learn from you more
in terms of what that looks like today. So we've
kind of at the stage, right, So we've talked about,
like kind of historically speaking, Angular and SEO not a
great thing. Other frameworks were trying to solve this in
other various different ways. But let's kind of fast forward

(11:32):
to like now, ish right, So we've got Angler is
a very mature framework. We've got things like service side rendering.
We've also got and I want to give a shout
out to Brandon Roberts with analog JS where I can
do static site generation. We've got hydration and partial hydration.
We've got a lot of different things, a lot of
different tools that we've got available to us. Tell us

(11:54):
a little bit like, what does Angular and SEO look
like today? As an Angular developer, Even if I'm not
the latest V twenty Let's say I'm I'm still in
nineteen or something. Kind of is what is the landscape today.
Am I still at the back of the pack or
what are some of the things that I can do
to kind of move things forward?

Speaker 1 (12:13):
No, I definitely not at the back of the pack,
And in fact, I would say that especially in the
past three years through a series of the path from
Angular sixteen I think it was when they started to
launch some of like the initial hydration and incremental hydration
and up through now with v twenty with route render mode,

(12:37):
which is another big update that I can talk about
in a minute, but basically a bunch of stuff to
put them pretty on par with other frames are out there,
and really the key go to model for the ultimate
initial load performance and ultimate user experience, which translates to

(12:57):
good SEO because the search engine see that, and user
see that, and whatever is instant rendered, that's it's all
a server view that's instantly rendered and then like almost
no job script downloaded initially like very very very very minimal,

(13:18):
and then as the user interacts to the page that
it starts to come alive. So that's like part of
the incremental hydration thing. And once you have that basic setup,
which it took Angular a while to get there, it
reacts had that a little bit you know before in
the past. Once you get there, then from that point
forward all of the different advantages and the way that

(13:39):
the framework can help is essentially in developer experience to
try to make it easier, to ensure that you can
do that, you know, for as you kind of continue
to build and scale out your application, because what happens
a lot of time, frankly, is that you if you're
building something like a one page or like simple test app,
even back before these features, you could you could hack,

(14:02):
you know, around the issues that Angler had to achieve
these different things, but it require a lot of effort,
right Like I've done it. I've hacked around the inabilities
of Angular in the past to still achieve the same
level of stuff, but it just required a lot of work.
So the key is once you kind of get there,
is how easy can Angular make it? And it seems

(14:25):
like every single release now they just make it easier
and easier with like little tiny updates.

Speaker 4 (14:31):
What kind of updates are we talking about, because we
so far mostly focused on like the performance aspect, and
now I feel like we're more switching into like the
meat attack kind of things and other developer experience optimizations.

Speaker 6 (14:42):
So so what's on your mind.

Speaker 1 (14:44):
Yeah, So, like let's say, with defer an industry came
out to have like deferred views, which has has This
isn't all just SEO and source I rendered stuff, like
a lot of stuff has advantages just for general performance
and user experience, even if you aren't doing any of that,
like just to client side app. But like deferable views,
they've added like extra features too deferable to make it

(15:08):
to the point where now you can have kind of
like essentially like server only components in essence, even though
they don't call them that, and only hydrate when the
user starts navigating around. And then like another good example
is I mentioned the route render mode, so like you
can already do at the router level lazy loading, which

(15:32):
helps on the client side so that you can kind
of cut up your package into little pieces, which is
a very good practice to do, and then you only
load them once they go to that route. But the
thing that was missing was to say that for one
particular route you just don't need client side at all,
not even forget about even like deferred views or whatever else.

(15:53):
Just it's like straight static And there were ways to
accomplish this in the past, but again and it was
kind of like a pain or whatever. But now it's
like a very simple API like it made like literally
a bunch of like I deleted a whole bunch of
code that I had in my school app that I
did in order to like try to achieve this now
and it's like a very simple like one line for

(16:14):
each for each route that now you can achieve the
same thing.

Speaker 3 (16:18):
So that that's cool.

Speaker 6 (16:20):
I have not used that.

Speaker 4 (16:21):
So that basically just means that does like a static
generation or like hydrates at once and does a static
generation of that snapshot then basically and just push out
that that road.

Speaker 1 (16:32):
So so it makes sure first of all that like
for routing, you know there there it is a little
there is a little bit trickiness that has to go
on in order to enable this because if you're on
the client side and you're only doing client side routing,
it's going to always look to the client side jobascript
to load the next page for route. But if you

(16:54):
want to do this mixed mode where you have like
a server only pages and client only pages, it has
to know for the server only pages, that has to
kind of reload and go back to the server, you
know what I mean, because you want it to to
just have no client ClientSide JavaScript at all. And that
is actually important for I mean certain number of use

(17:16):
cases that like I've had in the past that it's
super beneficial. Even like the little tiny amount that is,
you know, for doing some of the hydration stuff, you
get rid of that it makes it even more optimize.
So I mean these are all like little tiny, like
one percent like little things or whatever, but it all
it adds up. It does make a difference when you

(17:37):
are kind of like trying to like have like the
most optimal site.

Speaker 3 (17:43):
Yeah, that's really cool man, thank you for bringing that. Yeah,
so it looks like I have not used this yet,
and so that's something I want to play with. Yeah,
right in your wrap route configuration, you can put the
render mode and then you can have pre render server
or client so you can instruct you know, and I
think a good example this might be like an about
page or something. It's like, listen, there's no interactivity here.

(18:04):
It's just some images and you know, I'm sure it
looks great, but there's there's no need that this needs
to go through additional like bootstrapping or hydration and so
just render this as like static and so that's fantastic.
What are some other opportunities for developers with kind of
the newer modern Angular tool set to target SEO.

Speaker 1 (18:28):
Yeah, so, I mean so a lot of this is
again performs later, like there's a whole bunch of stuff
like on what content you are actually producing, which we
could have a separate podcast, and that that isn't necessarily
Angular related. Although actually one thing on this I will
say before I talk about some any of the specific
stuff is that with a lot of the changes that

(18:52):
have gone on recently, it is beneficial to not just
have like a content only sites. So even the traditional
sites that have been only content only in the past
have realized that Google, Google and other like especially AI
related a UIs, are favoring the sites that have like

(19:20):
not only good continent and a certain type of content
with certain characteristics, but ones that have some unique feature,
some unique capability. So that's why like Angular and React
based apps are in some ways performing a lot better
in this new world than just like the straight static

(19:42):
websites of the past which used to like dominate you know, SEO.

Speaker 3 (19:47):
Like interactivity or like like on page time, that kind
of thing. Because if I've got a got a web
app or something like that, maybe that's got more advanced
functionality and interactivity. Let's spend more time on it. Okay,
kind of the metrics. So we're looking at I'm just curious.

Speaker 1 (20:04):
Yeah, So I mean the short answer is yes, but
it's it's helpful to kind of understand the background there.
And so let me let me quickly mentioned about like
what this big change was that I sort of alluded
to earlier, like the past couple of years, right, so
it's like it's all a related or whatever. And so
when chat GBT came out, right, which was like the

(20:25):
first leader in this space in terms of like people's
able to utilize LM based technology. Uh, there's always been spammers.
There's always been people who are trying to hack search engines,
the system in order to gain the system, in order
to get users. Yeah, yes, in a constant battle. But

(20:51):
as soon as they were able to utilize these tools
and then quickly other companies for all Gemini and everything else,
spammers it was like they on steroids. They started generating
like so much content. Yes, and to a point where
in the in August of twenty twenty three, Like there

(21:12):
was like a month where like you guys may or
may not remember it, where like Google search results went
really crazy, like they were just showing like the amount
of spam that was being shown for like top results
just like went off the charts because through different techniques
utilizing a lot of generated content, they were able to

(21:33):
get around a lot of the safeguards they Google had.
So yeah, like hit the panic button sort of changed
drastically changed their algorithm and it started favoring a lot
of things that it had in the past, and a
lot of it was for the intention of trying to

(21:53):
not put a lot of this just AI generated low
quality content higher up in the search ranking. And so like,
I mean one thing which is not angry related or whatever,
but just to give you an example, is like they
relied on more like brand authority, like so like they
cared more about your reputation of your entire brand instead

(22:14):
of like a signal individual content. But the thing getting
back to Okay, so circling back to the thing that
I was starting to say, how it's related to this
is that one of the ways that they knew or
they or things that they seem to favor more. Is
like if your site was not just like a bunch
of content, but it actually had something unique, some like

(22:36):
special functionality that users were able to like sign up
for and to use and like you know, like pay
for like even whatever anything that gave them like extra
signals that like Okay, this isn't just a spammer putting
like up garbage on the web or whatever. So people
noticed that like okay, you have to have some sort

(22:59):
of sort of yeah, unique feature, like something that goes
beyond content. So it's both those. You still need the content,
you still need good content, but but having both of
those is helpful now for SEO. Does that make sense?

Speaker 3 (23:14):
It does make sense. And I do remember like vaguely
that timeline because you know, I think you're previous to
like lllms. You know, content was really important in fact,
like I remember I went to like an SEO conference
a long long time ago, and I remember like one
speaker was like content is king, you know what I mean,
like on stage, like it's all about like you got

(23:36):
to get that blog and you know, get other people
talking about your product and of course off page optimizations
and inbound links, but you had to have like this
really active content and constantly like signaling to Google, there's
something new here. Come back and check me again, you know,
send the Google bot. I want to be. I want
to be, you know, I want you to index my
site again and see how I have all this new content.

(23:56):
So I continue to move up the rankings just in
terms of like freshness, but also like other areas and
like getting kind of that long tail, and I've got
all this great content. And then I do remember, like
all the gen Ai stuff came out, and then I
even remember like my realtor, like the newsletters that I
used to get for my realtor, which I don't think anyways,

(24:18):
this is a random story, but like you know, they'd
be like these, like they'd write things on their blog,
you know what I mean of like what's happening in
like the local community or broadly speaking, and then quickly,
like pretty pretty quickly, it was like you'd click the
link or maybe you go look at it or something.
Not every time, sorry about that, but like occasionally I
look at it and I would literally like come look

(24:40):
at it, like, oh, obviously just generated clothes. I'm out
of here, Like I'm not going to read this. This
is junk. It's it just like looked and smells like
junk and generated. And I think a lot of like
marketers in the beginning were like, GENI you know, this
is this solves our content problem. Like I can just
like brank out loads of content and you're right, the

(25:03):
spammers and the bots all kind of like pick that
all up. And so it's interesting that Google like had
to make a drastic shift in terms of like, Okay,
well maybe content isn't king anymore because content now has
become cheap. It's really easy to generate this content. And
so now I'm going to look at like legitimacy and
interactivity and like what what actual value prop do you

(25:26):
have in terms of like you're exactly your your you're
offering here? Because before they could kind of infer value
prop maybe through some of these other signals, but now
those have gone down. Yeah, so what does that mean
for the Angular developer? I guess is what I'm wondering,
like kind of steer it back a little bit, Yeah,
what does that mean for the Angular developer? So these

(25:47):
things change, the interactivity is there, you.

Speaker 1 (25:50):
Know, I mean, among other things, I think it encourages
us to use something like Angular. Like in the past, Oh,
you'd be like, oh, you're your your site requires like
your using as a channel, like yeah, don't use don't
use Angular. But like now I think it's actually an asset.
It's like valuable to you know what I mean.

Speaker 3 (26:09):
It's cool. Yeah, that's really cool. And we've got all
these tools with uh, service side rendering and static site
generation and all of this that we can we can
kind of play happily with with the SEO space. So
I've got a question. Maybe this is a change in topics.
So I want to see if Child or y'all wanna

(26:30):
wanna chime in here before I kind of move the
conversation forward in a different direction.

Speaker 7 (26:36):
No, I'm just learning here. I uh I. I am
aware of all of the new stuff related to suicide
rendering inside of Angler, but I have not had a
chance to actually use them or work with them.

Speaker 3 (26:49):
Yeah, And I think I think part of that there
was Jeff's point, right, Child, was like historically Angular wasn't
the tool that you picked up, right, I think you know,
Gatsby or Next or maybe some of these other tools
that really like out of the box quote unquote like
and I know Gasby wasn't a react, but anyways, kind
of like out of the box, it's like, hey, I
can really easily like render like I can have like SSG,

(27:12):
and I can have service, I render SSR, and I
can like mix all these things together and I can
make it really easily and everything just kind of work together,
whereas like traditionally Angular that was kind of a challenge
to put that all into into play. Go ahead, you on,
what do you got for us?

Speaker 4 (27:28):
I'm first of all too stupid to under with myself,
but what I was wondering is if you're seeing like
an advantage point and using Anger over one of those
more traditional frameworks that are like intended for like SEO
have the application like Astro next, those kind of things
gets me, even though that's kind of a curse worth
these days.

Speaker 3 (27:46):
Yes, sorry about that Astro.

Speaker 1 (27:48):
Yeah, I mean so, the primary advantage I see, to
be honest, is that Angler allows you to add the
super rich functionality when you need it. So like those
things are are great, perfect for doing the the content
stuff and the stuff that's geared specifically for search engines,

(28:11):
but when you have your entire application, it's going to
be a mix of that it's going to be a
mix of landing pages that are geared for people coming
in and then you know, some rich content and rich
functionality and being able to do all that in one framework.
That's where the major advantage comes. You know, in the past,
like tons of different sites that I've worked on in

(28:33):
different setups and I've worked in somewhere you kind of
split things up and you use like different routing to
have you know, let's say one of those more traditional
SEO friendly frameworks handle landing pages and then it's completely
separate for this app Acmeinc.

Speaker 3 (28:53):
Dot Com is your anger and dub is your word
press or your squarespace or your Wix or whatever it
is that you're using.

Speaker 1 (29:01):
Like you're but but the noalo chy is it's not
it's not that it head to head for for that stuff.
It's like better per se, but it's now that it
brings it all together and you could just use one
thing for everything in no contact switching and whatever.

Speaker 7 (29:15):
Here's uh, here's my approach, and I guess like, here's
my perspective to uh, should I use Angler Silvi side
render for today's applications not necessarily se O s c
O S cee O right.

Speaker 4 (29:29):
Uh.

Speaker 7 (29:29):
The thing that I I see in Anglo Silicide rendered
right now, it's still lacking in out of frameworks is
the ability ability to use the silver that comes with
Cilicide render to actually set up your your routings for
your data. And that's like a huge bummer to me.
Let's say that I want to be Yeah, I know,

(29:51):
I know, right, I'm just saying, like Anglo c O,
I like the box I want to make.

Speaker 3 (29:58):
Uh.

Speaker 1 (29:58):
Let's say I.

Speaker 7 (29:59):
Want to make content to promote Angular SSR. And in
order to do that, in order to build like a
full full slack application by just using Angler I was,
I would still have to stand up like a no
express application for my back end separately. And I think
that's great, like friction for content creators to put stuff
out there to promote this. And it's like, I think

(30:20):
it's like a two ways kind of thing. People are
looking for stuff, looking to try stuff, but not enough content.

Speaker 6 (30:26):
But isn't that kind of the same in other frameworks too.

Speaker 4 (30:29):
I mean all of them have an SSR story by now,
but not necessarily server story. I mean that's where next
comes in. That's where next comes in, That's where analog
comes in.

Speaker 6 (30:38):
Like you and stuff. We can do server rendering two.

Speaker 7 (30:41):
But yes, but because those are manstreamed and like in Anglar.

Speaker 6 (30:46):
Okay, that's point.

Speaker 7 (30:48):
Yes, the thing is like, so let's say if Angler
on the server side rendered page on the documentation is
say we endorse Analog. Yes, that's a different story. Yeah, yeah,
Like these these different libraries saying react, Hey, if you
want something so Iran to go use next jays Angler
does not say that.

Speaker 1 (31:10):
Yeah, and I think they should. I think that Analog
covers like a really great space and they do like
an amazing job. So and maybe they will in the future.
I don't know. I would say that Originally in twenty
sixteen when Angular two came out, Initially, Patrick Stapleton and
I created Angler Universal as a separate library, so it

(31:33):
wasn't part of Angular first and it took years, like yeah,
at least four years before they started, like they integrated
everything to like inside the Angular library itself. Yeah, yeah,
well it was actually technically we did add it. It
was on the Angular repo, but it wasn't part of

(31:54):
the main code base. It was like a separate module
that wasn't part of the main Mono repo or whatever. Yeah,
so it's slightly different or whatever, but but whatever, like
it made sense. I don't know if that will necessarily
happen with like something like analog, just because like it's
not like I think it would be probably too many features.

(32:15):
So it's like, in a way a good thing to
keep it separate. But I do I would advocate for
them endorsing it, like for like actively being like this
is the solution for that, you know, I'd be advocated
for it.

Speaker 3 (32:28):
Yeah, it's a great it's a great framework. Yeah for
hash Brown we built it all was analog, and yeah
I was. It was my first like big project with analog,
and I was impressed. It's just how well it worked.
Like it's so for me anyways, and maybe just another
shot out to Brandon and Chow and team, Like everything
I wanted was already there, which showed signal to me

(32:51):
as an Angular developer that this is fairly mature, if
that makes sense. Like I wanted to build an interactive experience.
I wanted to have static pages. I want to have docs.
I want to have an API reference. I want to
be able to have you know, all these different things.
I want to be able to pull them together. I
want server routing, I want all of this, And it
was just like every time I went back to the documentation,
I was like, oh cool, nice, that's there. I like

(33:13):
that great. Like everything I wanted to pull off the
shelf was there for me. And so a big shout
out to Analogue. I think if you're you know, potentially
considering building something like what we're talking about, where like
this is a hybrid marketing slash e commerce like kind
of public facing application, and SEO is important. Seems like

(33:36):
at the base Angular plays very well with it today.
And then if you're looking to use something like an
open source project like Analog where you kind of have
some additional tooling around Angular, it's a great option. It's
a really great option. All right, let's nuke this conversation.
Let's get to the next conversation. Jeff, does SEO matter

(33:56):
today with AI?

Speaker 1 (33:58):
So, yeah, this is a big topic because if you
talk to some people, some SEO professionals, especially some people
whose businesses have been hurt by a lot of the changes,
they feel that everything is killing the web to go
Like I'm saying to go to an extreme viewpoint here,

(34:21):
because if you have and I don't necessarily subscribe to this,
but let me lay it out or whatever. Yeah, yeah, yeah,
So if you have AI giving you the answer, you
can make a good argument that that is the best
possibly user experience. But then they get that answer from

(34:41):
other websites, right, but they don't give any traffic to
those websites. So that breaks the sort of fundamental economics
of the web, Like at least what it's been for
a long time that it used to be these websites
would give stuff away for free in exchange for users
coming back to them, that these other whatever, other systems

(35:04):
that utilize their free content would give them back users
like give them links or whatever. Now, I mean it's
debatable weather to what degree, Like that's not been fully
the case for a long time because Google has always
had certain things.

Speaker 3 (35:19):
I was gonna say, wasn't There wasn't. There was like
a big lawsuit back in the day with Google because
they were rendering some of the content from a website
like right on the top of Google, you know, they
stitch in all the various answers. Yeah, there you go.
That was it, right, right, And I don't remember I
don't think it was The New York Times, but that's
with open AI. There was a lawsuit you know, years
ago where they were you know, they were negative company.

(35:41):
This organization was negatively effected because their content was being
like lifted up off of their site onto the actual syrup.
And so in the SERP page, people were like, oh,
there's my answer. I don't need to like go down
click on this organic regal or perhaps even paid result,
land on this page and then visit them and then
have an opportunity to learn more, to explore their website,

(36:03):
maybe sign up for that product, you know, whatever that
funnel looks like. But it was a big lawsuit back
in the day because of that content being lifted up
under the SERP And I know that you know that
was an issue back then, but it's almost like this
is coming back up again now with AI. So yeah,
I think there's some legitimacy legitimacy to the concern.

Speaker 1 (36:22):
Yeah, And so I don't think the answer in the
long run is necessarily like, oh, we got to add
links back to everything. And in fact, I so if
you are someone who kind of like believes that these
AI systems are gonna get better, which people you should,
you know, because they're going to continue to improve the

(36:43):
way they do things. You have to assume that more
often people are going to get their answers within the
context of that user interface and not necessarily be bounced
around all over place. So I think the answer is
not necessarily adding links back in, even though it's it's okay,
and it's a good thing to have the referral links

(37:04):
at the bottom, but I just don't think that many
people click on them, and less people click on them
over time as they trust the AI more so it
more is okay. A couple things.

Speaker 8 (37:15):
One is.

Speaker 1 (37:17):
The SEO people being a little bit more smart, or
people that are using search and AI as a channel
like that that you want users, you actively are trying
to get users through search or search engines slash AI
user interfaces, having first of all restructuring your content to

(37:40):
that in a way that appeals to a lot of
those AI interfaces. But there'll be all sorts of new
tricks of like, you know, okay, whatever you're writing including
your brand in it, so that like when they when
the AI is displaying the answer, that you make sure
that it has your brand somewhere in there. That like

(38:01):
there's something along those lines that can be done there
is with MCP, which is like a way of having
these AI use functionality within websites. There's a lot of
flows that people have started to like figure out or
think about it should say, because it's so new like
that it's not really being used at scale or anything

(38:24):
like that. But there's flows that like a lot of
discussion within the SEO channels or whatever else of like
how you basically leverage an MCP service that is utilized
through AI and sort of transition that to somehow get
the user to either recognize your brand and or like
get them to then you come to your site afterwards.

(38:44):
So there's there's different paths along those lines that are possible.
Like again that's like super super new and like who
knows how that's going to develop. And then the last
way that a lot of people are really pushing on
is trying to push for or systems that or processes.

(39:05):
But some type of setup where these AI companies are
incentivized actually straight up pay content creators and other creators
because they provide such valuable stuff, and that doesn't really
exist right now, but there's a lot of discussion in
different ways that could potentially happen in the future as well.

Speaker 3 (39:26):
Yeah, I can see that evolving, just like we saw
like YouTube or whatever evolve from like just a free
website with you know, basically, I think Google is losing whatever,
five hundred million dollars a year on YouTube too. Clearly
thinks have shifted over the last five ten years in
terms of what that looks like and how the content
creators have been really instrumental to the growth and really

(39:49):
included in part of that. And that's I'm sure there's
a lot of debate and nuance within that. I'm not
in that community, so I don't fully understand it, but
I think that's really interesting. So the the short answer
I think that you said is like SEO still matters
today if I understand correctly, but things are certainly shifting
and changing, and it'll be interesting to kind of keep

(40:10):
an eye on the market as things change, MCP kind
of how these AI providers are going to be shifting.
But content still matters, and good content still matters, if
I understand correctly, because the AI companies are still, you know,
to a certain extent, built on the content of humanity.
And so yeah, I think that's really interesting. I have
a ton of questions, but I also don't want to

(40:31):
dominate the heck out of this conversation. I want to
give Chao and Yan an opportunity to kind of chime
in here as well. I'm sure they've got some thoughts.

Speaker 6 (40:39):
Good morning.

Speaker 4 (40:40):
You know that moment when your coffee hasn't kicked in yet,
but your slack is already blowing up with Hey, did
you hear about that new framework that just dropped?

Speaker 6 (40:48):
Yeah?

Speaker 1 (40:49):
Me too.

Speaker 6 (40:51):
That's why I.

Speaker 4 (40:51):
Created the Weekly Depth Sproove, the newsletter that catches you
up on all the web def chaos while you're still
on your first copy. Oh look, the anger feature was
just released, and what's this typescripts doing something again? Look
also through the poor requests and change shot gramma. So
you don't have to five minutes with my newsletter on

(41:14):
Wednesday morning, and you'll be the most informed person in
your stand up. That's better the Weekly Desperate because your
brain deserves a gentle onboarding to the week's tech mathness.
Sign up at Weeklybrew dot dev and get your dose
of deaf news with.

Speaker 6 (41:30):
Your morning caffeine.

Speaker 4 (41:31):
No hype, no clickbait, just the updates that actually matter.

Speaker 6 (41:35):
Your Wednesday morning self will.

Speaker 4 (41:36):
Thank you one thing I'm wondering then, is like, how
do you, for instance, evaluate now githubs move to have
like an MCP on the one hand, but also gethub
being like particular for developers like very content heavy, like
we at least I use it a lot for like
I don't know, I look very often like anger at
TCB engine. It takes me to the right file and

(41:58):
get up directly, because that's the thing that I need
to look at every soft And so how do you
evaluate such a move versus like because it definitely takes
away from the actual side over to like whatever chat
interface you use, a tool you use to interact with
an MCP.

Speaker 1 (42:20):
So it's hard. I've actually talked or tried to ask
a bunch of Googlers about something along these lines at
Google io in like how should we be thinking about this?
You know, you people that have traditionally relied on this
channel and or want to in the future, and you know,

(42:41):
especially when the value coming back is not necessarily there
right now, which is sort of what you're alluding to
and the common message. And I do think this is true,
even though it's sort of annoying to like hear about.
Is that, like, ultimately, if you make something valuable, like

(43:03):
you will eventually be rewarded for it, because like if
you think of the incentives on their side, So like
let's say you create this because this is so new,
like this mc because we talked about that, like this
MCP service that is you're really valuable. It's it's it's it.
People are using it a lot, but like you're losing

(43:25):
money like crazy or whatever. So like there's a question
of how how long you could support that. But if
if the AIS start using it more and more, like
eventually they're going to realize and then and then, like
if you think about in a larger scale things that
this happens. But then a lot of these services just
start going away because it's not being funded. Eventually, the
the the market will dictate, like they'll there'll be a

(43:50):
realization that there has to be some sort of it's
it's beneficial for these ais to have that capability as well,
and so that they have to somehow back. So I
don't know how that's going to happen. I don't know
the answer, but it's so it is more of a
faith thing, which I feel very uncomfortable with. But if

(44:10):
you look at the history, I would say, like, okay,
forget about the current environment. Let's bring it back to
the twenty years of spammers trying to hack and get
around like search engines. It's worked to some degree, like,
but but the thing that's consistent is whatever thing people

(44:31):
figured out to make it work, it's always temporary, always temporary,
like eventually it goes away.

Speaker 3 (44:38):
Yes, yeah, yeah, or whatever. Yeah.

Speaker 1 (44:40):
But if you do create just valuable content and is
like something that's truly used by other people, that is
almost always consistent, like because Google is incentivized to try
to make sure that you're like whatever changes that they make, like, Okay,
temporarily it might like there even with my various websites,

(45:02):
we've had dips and whatever else, even though it's been
like really good content, are really good, well very popular
app but it bounces back quickly or whatever eventually because
again it helps them as well. So maybe that's the
best answer to give you.

Speaker 3 (45:17):
I know, but I think that that is that that
has long been the answer, right, And I mean that's
like if you build something of value and you continue
to kind of play the long game on it, like
hopefully the market responds appropriately and those signals will come
around but you're right, I think there's a lot of
like jostling in the short term.

Speaker 4 (45:36):
Go ahead, young, But if I play Devil's applicate, then
what does SEO a matter of content is king at
the end of the day our quality?

Speaker 1 (45:44):
Well okay, okay, okay, okay. Yes, So this has also
been a question that there's been asked even you know,
like a decade ago or whatever else. And so I
guess the thing is in order to like, it's one
thing if you want to gain some sort of value back,
and you will eventually, but if you want to be

(46:05):
competing for like the top rankings and like the top
of the tier for different areas, you always need both.
Like you need to have like a super valuable service
that is good, but then you also need to have
that SEO part in order to like ensure that all
the right signals are communicated from Like it's because because

(46:29):
you can have the best service and like because nobody's
talking about technically or yeah, some technical glitch that you
have that Google isn't able to like detect that, you know,
because they don't they're not able to like see into
the minds of every person on the planet or whatever.
So it is a combination of both. It's needed for sure.

Speaker 4 (46:49):
Okay, last question from my end because the thing that
I always feel with like SEO, it kind of feels
like playing the system in a way, right sure, because
if you if you look at it like okay, right, well,
but that that's again like saying like okay, comping doesn't
matter as much.

Speaker 8 (47:09):
So it's like, yeah, how do you maybe I just
need would be interested in your perspective on that. Do
you see on the one that I agree with you?

Speaker 6 (47:21):
Okay? Easy?

Speaker 1 (47:23):
That No, it's it's so whenever you have like s
c O is a marketing channel. It is a way
of acquiring users at the end of the day, and
there are other marketing channels. Any marketing channel you have,
so like uh, you know, traditional sales or trying to

(47:45):
do virality, which is a stupid channel, or whatever, any
of the different marketing marketing channels, you have to put
effort into it, like you have to like it's yes,
your your product is good, but then you have to
put effort into the channel itself. And that's what I
view SEO as is like that's it. So for some
of the other ones, it's a different type of effort
for SEO, it means understanding those different weird rules and

(48:09):
ever changing rules like it changes over time, and trying
to like keep up with it as it changes over time.

Speaker 4 (48:16):
So in terms of like analogies, you kind of see
it as like making a stupid face for YouTube video,
where it's like, okay, it makes the content more discoverable,
more approachable, maybe yeah, but then at the end of
the day, the content still matters more because otherwise people
will look at it, identify at as garbage and leaf
within ten seconds.

Speaker 1 (48:33):
I agree. Yeah. In fact, it's funny. One of the things, like,
so I'm part of this SEO slack that people always
tell their different stories of different clients and whatever, and
the most common thing like that they always talk about
is like whenever they go into a new client is like,
oh yeah, okay, so we did filell this different stuff
and like we can work on these different technical things,

(48:55):
but okay, this will help. But like your core thing,
like none of this is gonna matter if like you
don't you know, improve this. This their core, like your favorability,
you know, the the actual feedback that you get from
real users human users for your your main thing, you know,

(49:18):
like you can't put lipstick on a pig. Is that
the analogy?

Speaker 3 (49:22):
I think that's it. I think that's it. It's interesting.
I mean, like some of the largest growth companies we've
seen in the last you know, two to three years,
their marketing strategy has been largely build something that people want,
and they'll just tell other people about it, right. I mean,
I don't want to get political. I hate to even
say it, but like Tesla is like an example. I know,

(49:43):
I'm sorry, I don't want to talk about Elon, but like,
you know, to a certain extent, like they've been broadly successful,
certainly in the last it's arguably around taxes and the
financing side of it, but there was a certain extent
they've been successful just because they had no marketing strategy
other than just build a really great product and people
will tell there are people about great products.

Speaker 4 (50:01):
Right.

Speaker 3 (50:02):
Open Ai is another example of like I mean, talk
about a hockey stick just in terms of like month
over a month or ar r. Right, I mean just massive, right,
But yeah, go ahead, going to challenge it.

Speaker 4 (50:15):
Yeah, Open my Eye is a little difficult because they
have like all the Microsoft marketing butts with in the background.

Speaker 6 (50:20):
But Curson is a great example.

Speaker 3 (50:21):
Yea is a great example.

Speaker 6 (50:24):
I don't even have a person's department.

Speaker 3 (50:26):
Yeah, but their marketing is like developers telling other developers like, hey, dude,
you got to check this thing out, watch this agentic mode,
and watch what happens when I use Claude four and wow,
I can do this thing and this is really powerful
and look at me. Go so yeah, I think at
the end of the day, I think Jeff's like one
of our key takeaways here is like whether we live
in the world of SEO, which was maybe kind of

(50:47):
still alive and well today, or in the world of AI,
and how these two kind of bridge and kind of
depending on what happens with the research engine, result page
and people's user behaviors and all of this. At the
end of the you're building a great product, like hopefully
the market response to that great product, whether it's through
whatever channel that's through. I think the other takeaway for

(51:08):
me is also that like Angler is a great choice
for building a marketing website and for building an app,
and where SEO and AI matter, Right, So if you're
building whatever it is, you know we're building hash brown
open source project, you know we built it with Angular
whereas like, honestly, I got to say, like, if this
was like three or four years ago, I don't think

(51:28):
I'd be building it with Angular. I think it would
be like an astrocite or a next JS or something
like that, you know what I mean. And so that's
just it's cool that it's you know, as as folks
that have kind of followed the Angular story for a
long time and are invested in the community, it's cool
to see that Angular is kind of growing in this space.
But yeah, a lot to learn in terms of how

(51:50):
this all impacts kind of where AI goes and stuff.
I think I was shared before the I'll be sure
this real quickly before the podcast, before we started recording.
You know, we're a tiny little consultancy. We do a
lot of stuff with banks and financial firms and and
that kind of thing, and a lot of folks use
ag grid and we had like our first client lead,

(52:11):
like somebody emailed me and I was interested in doing
a project, and I was like, oh cool, you know,
how did you hear about us? And they're like, you know,
I asked Givity and Gibbety was like, I talk for
the folks that live love app like they're they're the
people to talk to about angular and ag grid and
so I was like, all right, dude, nice, very cool.
I like this. So you know, it's like, what do

(52:33):
I need to do to maybe further this along? And
like kind of what are going to be what's kind
of the long term play here, because I do anticipate
it'll depend, it'll be multi it'll be I think, kind
of variations based on any kind of direct generational and
what kind of product you're building and all that kind
of thing. But I do anticipate that more and more
folks are going to go straight to something like Giviety

(52:53):
and say, hey, you know, I'm looking for a plumber,
I'm doing a project, or I'm looking for our landscaper
in my backyard in Kansas City. Here's my budget. I
kind of want a rock feature with a waterfall. Who's
are some companies that I should reach out to And
they're going to be like, here's the landscaping down the
road that you know near you that you recently did

(53:14):
a rock fall feature in somebody's backyard. You should call them.
And you're like, great, I'll call them, you know or
whatever that is, but go ahead. Yeah. So you're doing.

Speaker 4 (53:23):
That's hilarious because you're actually lining up contractors for backyard
READINGSIGN right now, so are you.

Speaker 3 (53:30):
That's so funny. I just yeh.

Speaker 4 (53:33):
On that note, though, you kind of alluded on that
that the browser will eventually die from your perspective, not
the web. I still think that the way you sounded like,
there will still be like web content used for like
indexing and training lms. But what is your perspective do
you think people? That's and I don't want to put
words in your mouth, but that's kind of like a
picture that you hinted me.

Speaker 3 (53:56):
Yes, you know, I think I would have to answer this.
I'm intrigued as well.

Speaker 6 (54:01):
And you picture it. But I want I want to.

Speaker 3 (54:03):
Just as like a kid of the nineties.

Speaker 6 (54:07):
Like me, and you didn't ever have a browser at
that time ninety.

Speaker 3 (54:11):
Three point zero. We had we had nets, the Navigator
dogue I used to it was on the shelf at Staples.
I could buy my browser and take it home with
me and install it. You know, I don't know, like
browsers are just like right here inside me, Like I
hope we never lose the browsers.

Speaker 6 (54:26):
Like a walk next to it.

Speaker 3 (54:29):
And I hope we never lose like the openness of
the browser, right, you know, like view source has been
like or whatever. It was like the openness of the web.

Speaker 4 (54:37):
But yeah, but you're talking about the openess up the web,
and I don't think that will go away, but really
just a gateway. It's basically like you the entery to
like discover that. So and I think that will eventually
go away because it will be just more convenient, as
like in your example, like why should I go to
Google and figure out what other best landscapers in my
area when I can also just go to chetch you

(54:57):
pet to figure that out.

Speaker 6 (54:58):
So I would be rid of curious to hear your
spectacle now.

Speaker 3 (55:01):
And I think maybe I'll just pile on before je.
I'll give more time to Jeff to answer this question.
So you come up with a really good one for us,
let me tee it up a little bit more for you, Jet, Like,
I think it's really interesting because you're onto something like
let's just back out of the conversation just real briefly,
like think about like the workflows of the web today,
Like here's a really popular workflow in the web today,

(55:22):
Like I need to set up I just started a
new client or a new project. I need to set
up a daily stand up, and I need to set
up in my calendar. I want this to be a
repeating event on Monday, Wednesday, Friday at nine o'clock. I
want it to be on Thursday. For our schedules, we're
gonna do ten am stand up on Thursdays and whatever
it is. So I go into Google Calendar and I

(55:42):
hit ad event, and I have to do this custom
thing and I have to go down here and have
to click this button and this button. Ah, shoot, I
got this wrong. The validation is wrong. I got to
come back heret fix this thing. And it takes me
like minutes to just create this like stand up event. Right,
But the power of natural language with an AI model
is I can say, hey, I need to create a
stand up Monday, Wednesday, Friday at nine o'clock Thursdays at

(56:04):
ten o'clock. Send the invite to my coworkers Ben and Mike. Like,
natural language is really powerful, right, and so it'll be
interesting to see what that does with the browser, what
that does with all of this. But that's just my ramblings. Jeff,
You've had plenty of time to come up with an
amazing answer. The question that presented to you, Jeff, is
our browser's dead. I think it's the question up on

(56:29):
the fly, and we're gonna put browsers on it and
say that the browsers are dead.

Speaker 1 (56:33):
We're all coming short answers. No, there will always be
use cases and whatever. But the real question is more
so like how many actual humans are going to be
on the web, on your website, on your domain, viewing
your web interesting versus consuming your services in other ways.

(56:53):
And I do think that over time people can consume
your services potentially in other ways, either giving your service
like through like a MCP type thing where it's just
like the functionality, and or maybe even in the future,
like there's UI elements that you're able to somehow influence
in some of these AI interfacing well you know, web

(57:17):
components live again or whatever.

Speaker 3 (57:18):
I don't know.

Speaker 1 (57:19):
But one one interesting keeping point with a lot of
that that I'm not sure how it's going to go
is will these AI interfaces have ads at some point?
And that may seem like a small thing, and it
may seem maybe people have opinions one way or another,

(57:42):
but that actually will have a pretty dramatic impact on
a lot of different stuff. Because yes, so it's one
thing if everything is subscription based and you can sort
of the user can trust that like, Okay, they paid
their money and everything in here is not there's no
like ads snuck in or whatever. And in that case,

(58:02):
you know, they'll know if they see a certain piece
of functionality that like okay, this came from this other site,
like there wasn't like money exchange, I don't know whatever.
But like as soon as you introduced ads to it,
it might make things weird, Like it might both for
the user and for you as a service that's trying

(58:22):
to be utilized within this, Like I don't know how
that will work out, but that's going to have a
dramatic impact on how this all plays out as well.

Speaker 3 (58:33):
I think it's a really interesting point, Chef. I mean,
but I read yesterday opening I lost like five billion
maybe last year or something like that, and like that's
a bit of a burn. I'm sure they're not concerned
about it, but I think they still have plenty of cash.
But nonetheless, like it'll be interesting to see how this
all plays out, and and ads, it's going to be
one of those things, right is you know, now can

(58:53):
I place ads? And what does that look like? And
how does that change the whole dynamic and the relationship
between me and because right now my friend, you know,
like we hang out, we talk, we chill, you know,
like I didn't bite Giberty over to my backyard barbecue
with my nice rock feature and water landscaping.

Speaker 4 (59:11):
You know.

Speaker 3 (59:12):
But it starts.

Speaker 1 (59:13):
Selling as if they add if they put ads in
an unobtrusive way, sort of like Google circa twenty oh two,
like the early two thousands, where it's like in this
sort of isolated area where like everybody knows, like, oh,
that's the AD area. Then actually that I don't think
that's necessarily an issue per se. I mean, especially for
free service or whatever. The problem really comes if they

(59:36):
start integrating ads, you know, you know where.

Speaker 3 (59:39):
They generate in the generator.

Speaker 4 (59:41):
Yeah, you're a star, or I wouldn't next surprise if
we see something like kind of like the sports event
where it's like this half time report is sponsored by
Pepsi or something, and so like this massive's generated in
cooperation with.

Speaker 3 (59:58):
My that's interesting. Yeah, it certainly changes the relationship and
it certainly impacts a lot of how how people are
going to be interfacing them at these things. Yeah, what
do you got cha?

Speaker 7 (01:00:11):
I just I just want to give my two cents
on the like moving away from the web from a
perspective of an individual contributor to a depth tools company,
I did think building a terminal UI applications right now,
at this point, with the current technologies, is much better

(01:00:32):
than a web based application for a depth tools company.
Interesting for interesting, for example, I'm just saying, let's just
take GitHub Actions for example, I personally would use a
GitHub Actions terminal UI application then going to their web
UI because on the terminal, I can, I can I

(01:00:55):
have I already have my GitHub CLI or authenticated and
everything right, I can pull up an action that failed,
I can pull the law from that fail, spin up
another CLAW section with MCP servers loaded or whatever. Right,
it's just right there on my terminal. I don't have
to live the terminal at all to go to Google

(01:01:16):
and they search for the bugs and look at the
logs and all of the stuff.

Speaker 6 (01:01:19):
I think.

Speaker 4 (01:01:20):
I think, and terminal your eyes are like have a
very defined way of navigation and stuff. In a way,
there are more accessible than web applications that don't care
about that often don't care about accessibility, like you know
exactly in the terminal application what a tab and an
antwer will do and how they interact differently.

Speaker 7 (01:01:39):
Exactly, and what I mean, like in terms of the
availability of the MCP's technologies right now on the terminal
is much more vibrant than what is available on that
web I don't think it's available.

Speaker 3 (01:01:53):
It's interesting, Yeah, I think like two thoughts to that.
I think chiao and maybe a little bit of a
challenge included in this is like you could view and
one perspective is you could view chat GPT as that
terminal right of like, yeah, I don't need to go
look at the issues or the pull requests or whatever
it is on GitHub or consume it through the gethub

(01:02:15):
like web interface. I could consume that all directly through
like a terminal chat GPT and like a natural language
element and say, hey, how many open issues are there?
What issues do I need to review? Whatever, like what's
happening or whatever that is or whatever that interactivity looks like.
And then arguably I do think that we will see
Maybe this is just my own personal perspective, but I

(01:02:38):
do think that we will see web applications become more
fluid in natural language and the ability to act more
like a terminal. So you'll you'll eventually maybe potentially go
to websites and it's going to be more of a
terminal like experience on a website where you'll interact with
the website. Because if you think about it like today,
like we spend a lot of time as web developers,

(01:03:00):
engineers and designers determining the proper flow and navigation for
a user in my web app, you know, even GitHub
it's a great example. You know, I'm going to put
these links at the top, and I'm going to show
the files here and overhears these little graph and the contributors,
and I've determined, and I've put tons of money in
thought and user studies and design and worked into designing

(01:03:22):
an optimal navigational experience for the user. However, with the
natural language and with the M approach, what if why
don't have to design that optimal navigational experience for the user? Rather,
I let them tell me what they want and what
they want to be presented with and how they want
to navigate through this, and then I'll generate that UI
based on their what they're looking for. What they need,

(01:03:44):
what answers they want, and so my UI is completely
fluid based on who is it the who, like the law,
authenticated user or whatever it is. And like, yeah, maybe
i still have some investment in optimal navigation and I'm
gonna put a Hamburger menu up here. I'm going to
have a left column here, right, But that might actually
just go away more in the future as we begin

(01:04:06):
to interact with web applications in a more natural language
l M optimized world. But that's just my thinking and
how I think about things, So I don't know if
that helps, but I'm just yeah, I think this is
a really fun conversation. I appreciate. Yeah, I think the
terminal UI thing is really cool, and I like the
young last point I know wants to get it here.
That's party like the like the known navigational aspect is

(01:04:29):
really important right as a user, like I know, like,
but again, that's almost dictated by the terminal UI rather
than dictated by myself. It's still a pre determined navigational flow,
which is great. But maybe there's option another optionality around it.
But yeah, I all shut up.

Speaker 4 (01:04:47):
I love but I want to completely go back to
our original topic because yes, so I already said that
five times that I just have one more question.

Speaker 6 (01:04:54):
I do have one more question, Jeff, you do I do?

Speaker 3 (01:04:58):
How do you?

Speaker 4 (01:04:59):
How important do you think like lll M text formats
because there's the submerging web standard of like having a
slash l ll M path in your web.

Speaker 3 (01:05:10):
Yeah.

Speaker 4 (01:05:10):
Yes, that basically summarized like all the content. And I
personally could not find assertative statement that says, oh may
I uses that or cursor uses that or someone actually
uses that.

Speaker 3 (01:05:26):
It reminds me of robots that TXT.

Speaker 6 (01:05:28):
Right, someone knew that.

Speaker 3 (01:05:32):
Something started that didn't somebody Like is it the history
of that?

Speaker 9 (01:05:35):
Jeff?

Speaker 3 (01:05:35):
Like, is the history that like somebody started this and
then it kind of like took off and then now
it's like, well not now, but you know, fifteen years ago,
if you didn't have a robots TXT that you got
like a red check on your se O audit or whatever.
You know, like how could you not do this?

Speaker 1 (01:05:52):
But eventually once people, once people start doing something in
mass it's funny. There's there's certain things you try to
get people to do, and you try put so much
out to it and they just will not do it
or whatever, and then other things take off out of
nowhere and you're like, where did this come from? And
usually when it's the latter, like some some random thing
you know has taken off and people are doing it,

(01:06:16):
it's both very difficult to stop them and people, like
most people realize like, well, it's probably not our best interest,
like let's just use this as best we can, because
it's so hard to get If you tried to stop
them and get them to do something else, that that
would probably fail, you know what I mean, Like like
for robots TC, it's like, oh, like, oh, we want

(01:06:37):
to do this other format, it's like, well, nobody's going
to do that or whatever they you want to do it,
just use the thing that's already there that's working. But
as hard as the text, yes, I have not tried
to use that. I don't know much about that. I'll
have to look into it because but I suspect that

(01:06:58):
I can't imagine that unless out of nowhere, like robots like,
it sort of takes off on its own. I wouldn't
think it would, just because it's sort of unnecessary.

Speaker 3 (01:07:12):
Like crawling content.

Speaker 1 (01:07:17):
Yeah, and they're already good at like discerning what your
stuff is about.

Speaker 3 (01:07:22):
And all text. The LM text though, is more for
like my context stuffing, right, and so like if you
got to like the AI Engineer conference that was down
in San Francisco last week, if you got AI dot engineer,
you click on the top of their header, they've felt
like a link to their ll M dot t x
T and so I can snag that. And it's just
a just a system prompt, we'll call it that, right,

(01:07:42):
It's not you're not actually putting it in the system prompt,
but it's a system prompt that I can basically copy
paste all that lll M t XT and it has
all the schedule and the speakers and the agenda and
what's for lunch and everything in it. So I'd grabbed
this huge body of text. And your l M context
window now is massive, So I grab these, you know whatever,
ten thousand tokens worth of content. I go, hey, jippity,

(01:08:07):
I am attending AI Engineer conference next week in San Francisco.
I am staying at the Hyatt Downtown. Here's the l
M dot TXT. What are some fun things that I
can do on Tuesday night before the conference starts on
Wednesday morning. And it'll take all of that and be like, great, Brian,
the Hyatt looks like a great choice, because that's what
it always says. And you know, you're going to really

(01:08:28):
enjoy this pre party that's going to be down the street.
It's going to be a bunch of Angular developers getting
together talking about AI. It's right up your alley, you know.
And I parsed it from this giant context and so
here's this nice answer for you. But I don't know
of any like automated around that, right, yahn, we're like

(01:08:49):
automatically in that context or anything I do that, don't I.

Speaker 6 (01:08:53):
That's what I thought.

Speaker 4 (01:08:54):
It wasn't attendant for at least, but like basically to
define what the l M can train on from like
a website perspective. I have the website open and we
can put it in the show note. But l M
text is a proposal to standardize on using to provide
information to help LMS use website and inference time.

Speaker 3 (01:09:16):
Yeah, and I thought, yeah.

Speaker 4 (01:09:17):
I couldn't find it now on top of my head.
But I thought even Angular thought deaf, not use it.

Speaker 3 (01:09:23):
I didn't even know that. Let me check it out,
I thought, So I couldn't see I couldn't.

Speaker 6 (01:09:34):
But I thought I saw something the other day.

Speaker 3 (01:09:37):
Could you imagine like for like on the Angler that
dev instance, like they could stuff all the docs, but
they could also just stuff like the whole API ref
right into like a giant TXT file, generate it statically
or dynamically or whatever, and you know, or build And
now I can use that to stuff my context and
ask questions about the new.

Speaker 6 (01:09:59):
Window.

Speaker 3 (01:10:00):
But yeah, yeah, yeah I would. But then maybe my
editor is not going to keep constantly using like at
view child and they'll actually use the view child function instead,
because it could be.

Speaker 4 (01:10:12):
Could be properly version like version four, like oh, you're
using Anger seventeen. Here's all the apl for Anger.

Speaker 3 (01:10:18):
Seventeen TXT for seventeen. Stuff that in your context, and
now you can ask questions against your editor the jibbity
or whatever and say, you know, only use Angular seventeen
signal or whatever eighteen signals or whatever it is. And
so they'll do that and it knows the Yeah, go ahead, Jeff,
you want to sneak into.

Speaker 1 (01:10:37):
On the SEO slack, which is like all the big
name SEO people, including you, I'm just a lurker, so
they so I didn't look at this before, but they
seem to be going back and forth saying it's a

(01:10:59):
gammick slash fad. No system is gonna want to expand
additional unreliable requests. Why there's no sight wide Schema standard yet,
it's just a proposal yet it's not a standard. So,
like I think in general, thing they're saying is there
nobody's really jumping on that as far as the SEO
professionals until if and when it becomes.

Speaker 3 (01:11:20):
A standard of motipcial signals. Yeah, yeah, I mean that
was It's interesting they said schema because I gotta think
like Schema dot org and some of the stuff that
you know, the data Jason or whatever it is, Jason D.
And you know, I'm putting on the attributes on your
website to really clarify to the Google bot that you know,
this event happens at this time and all that kind
of stuff. Go ahead.

Speaker 6 (01:11:40):
Yah.

Speaker 4 (01:11:41):
Just so the l M text side has a directory
with all the sites that uses them. There are a
pretty couple of big players on there with Bond cloud
Flare that do use it. Yeah spelled Yeah, Well I'll
definitely keep an eye on it.

Speaker 1 (01:11:58):
We'll see it.

Speaker 3 (01:12:00):
It's an interesting thing like whether it'll become part of
the tooling though, and I don't know if it would
go into the training data necessarily. And maybe that's the case.
Maybe that was more of your question here on. But
also like say, I'm using just like the gibty client,
and I asked an Angular question. What if I could say, hey,
you're just in natural language. Hey use the following LM

(01:12:20):
dot txt file for version nineteen. I'm using Angular nineteen.
I want you to write only code with modern single based,
signal based approaches. Use this as a reference, And now
I can ask you that question and I've got that context.

Speaker 4 (01:12:34):
I made the proposal at least for Jeff Bryan and say, I,
when we have a slash web tool, we call it that,
it also checks the slash LM's text or whatever the
proposal is forction.

Speaker 3 (01:12:46):
Yeah, yeah, yeah, yeah, yeah, it'll be interesting to see
because so well the tools, like, I don't know if
it'll go into training data. And maybe that's the Jeff's point,
just in terms of like the SEO and the trust factor.
I can stuff anything I want into that, or I
guess I can stuff anything I want to my website.
But yeah, it'll be interesting to see whether that picks
up as a standard or where in the where in

(01:13:07):
the tooling that gets pulled in, you know, kind of
morbid run time or is that more of like a
training like data set that goes into it. I mean,
Opening Eye can't get their hands on enough data. I
can't imagine why they wouldn't wouldn't include it at some
point or something like that. But fascinating stuff, fascinating stuff.
I feel like we could just keep going, but I

(01:13:28):
think the listener is probably like, man, one of these
guys going to wrap up. Yeah, Man, it's been such
a privilege to just chat with you and to learn
from you and all this stuff. I think, you know,
I already kind of outlined some of the key takeaways,
but I think the other key takeaway is that for
me anyways, is that Angler continues to push the boundary forward.
Angular will be a framework of the future no matter what,

(01:13:50):
and I think Angler will continue to play well in
that space, whether that's a mixture, hybrid, SEO AI world
or whatever that looks like. I think, you know, Angular
has been has been a long committed standard around just
like moving the web forward, and so it'll be very
interesting to see how Angular kind of continues to progress
in this way. But a huge kudos to the Angular

(01:14:13):
team and of course to the Analog team and some
of the stuff that is in the Angular space today
and the broader ecosystem around just building really great interactive
user experiences using a framework that we all like and
enjoy to use, and we don't get penalized for that,
which is really nice because it did feel like, you know,
years ago, you almost got penalized to a certain extent

(01:14:34):
for using Angular for a marketing website. And so most
of the time, like my experience was, especially with like
being in the consulting space, is that you know, Angler
never really powered the marketing website, right, you know, the
marketing website was was word Press or whatever, and that's
a great choice. There's nothing wrong with WordPress. It's very powerful.
But now Angular is in terms of building your entire

(01:14:56):
ecosystem on a singular framework, that's a possibility today. You know,
can build my marketing website and I don't have to
like reship assets and have additional domains and routing and
everything's set up to kind of fork off like the
marketing side from the apps app highly interactive side, and
there's a lot of great technologies and tools available to

(01:15:18):
Angular developers to do that. So man, thanks for coming
on and thanks for taking some of the.

Speaker 1 (01:15:25):
Thank you guys for writing me. Yeah, great time, Jeff.

Speaker 3 (01:15:29):
If people want to follow you online or they want
to get to know you, and and definitely please do
plug your company and the things that you guys are doing.
We'd love to hear more about that.

Speaker 1 (01:15:39):
Uh well, definitely. If you have customer service problems, you
can visit get human dot com. We have tools so
you don't have to wait on hold, and you could
have our AI agent try to solve your problem on
your behalf call the company and try to speak on you.
We have to solve the problem or whatever if you want.
And then the other big thing I'm doing right now

(01:16:01):
is I have my own channel YouTube channel tech Time. Today.
In fact, Friday, I am recording an episode with Steven Fluin,
if you guys remember him, then we're gonna be talking.
I'm sure we're gonna be talking about a lot of
other AI stuff, a lot of AI stuff, a lot
of Angular stuff, a lot of SEO stuff. So in

(01:16:21):
I do a short form content, I try to edit
it down to like ten minute pieces each so definitely
check that out.

Speaker 3 (01:16:29):
Cool, very cool. Thank you so much. Yeah, appreciate it.
Thanks for your time, Jeff, and to the listener, We'll
see you next time. Bye.

Speaker 1 (01:16:36):
Yeah.

Speaker 9 (01:16:38):
Hey, this is Presstol. I'm one of the ng Champions riders.
In our daily battle to crush out code. We run
into problems and sometimes those problems aren't easily solved. Ng
comp broadcasts articles and tutorials from NGI champions like myself
that help make other developers' lives just a little bit easier.
To access these articles, visit medium dot com, forward Slash ngcomp.

Speaker 2 (01:17:00):
Thank you for listening to the Angular Plus Show, an
ingicoff podcast. We would like to thank our sponsors, the
ngcoff organizers Joe Eames and Aaron Frost, our producer Gene Bourne,
and our podcast editor and engineer Patrick Hayes. You can
find him at spoonful ofmedia dot com.

All Episodes

S9 E13 | Angular SEO 2025 |Jeff Whelpley

Episode Transcript

Popular Podcasts

New Heights with Jason & Travis Kelce

Dateline NBC

24/7 News: The Latest

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}S9 E13 | Angular SEO 2025 |Jeff Whelpley