The funny thing is, the techniques shown here are the ones that were once considered something only advanced front-end developers or publishers could do. Seeing that a former symbol of skill has now become a subject of satire makes me think that what we call 'high-level' ultimately comes from what others can't do. I personally never even thought about how to implement ASCII art animation.
I've worked on several projects where people looked at the site, which was simple and straight to the point, and people would straight up tell me they didn't take it seriously because it didn't have these performative UI things on it.
It's like when a Youtuber's audience complains about how they're constantly asking you to subscribe. The reason it happens is because the statistics say it works.
I don't think the commentary being made here is that startup websites should not be flashy. Just that, maybe they don't all need to look exactly the same as each other.
I think homogeneity is an unavoidable end game for the internet (unfortunately).
At work we’ve been discussing whether to migrate off our home grown component library to Material UI. I shudder at the thought, personally. However, a compelling reason to use a ubiquitous framework is that the ubiquity means folks intuitively know how to interact with your product.
Like many of us I was born into a deeply customizable Internet, all of my websites were green or red on black. They were a glorious amalgam of fixed width fonts and <blink> tags. With occasional wingdings characters for fun and games and complex <table>/<tr>/<td> tags for really epic layouts. They were l33t, honestly ^_^
But, as time goes on and more and more people use this thing, converging on the one-true-UX feels like a net good thing assuming the fundamentals are right. To some degree the LLM-ization of the Internet is essentially the end game of squashing the personality out of the Internet which bootstrap started.
We’re on the cusp of spoken word being the core UX of computers with a fall back to reading the LLM transcript, neither of which benefits from <blink>
> a compelling reason to use a ubiquitous framework is that the ubiquity means folks intuitively know how to interact with your product
Not that I disagree with you, but I'll also offer a tradeoff.
When people expect to pick up your app intuitively, it can also just mean them using the app absent-mindedly, which can mean them skipping the manual and jumping straight to trying to tie up the support lines. Whereas if your ui asks for a user's full focus up front, yes there are downsides to that but they're also more engaged.
It seems to me the parent commenter is saying the opposite: looking exactly like each other _is_ the point. It's a form of social signaling, to indicate that a project "belongs" to the in group of high-flying successful AI hype projects.
Note I'm not arguing that this is a good strategy. But given that so many people follow it I imagine it's not as bad as it appears on the surface.
It really comes down to first impression. Your website design is your company’s first impression. If the design is clean, people will believe the product is clean and robust as well. Similar to how people think things that cost more and probably high quality and better overall.
As for this website, the best component is the ASCII animation in the hero and you can’t even copy that component. In fact, that nice ASCII hero is what gave me a good first impression to go thru all the components.
Sometimes utility can be so good, users don’t care about design. I was also thinking of it as a business coming to a SaaS website. B2C is filled with so many dark patterns, first impression probably plays less of a role.
In the marketing world this is called revealed preference. This stuff is A/B tested to death. Anyone trying to sell something is best served by watching people's behavior instead of listening to what they say, as the two are often different if not polar opposites.
The perspective marketing world seems toxic. From the perspective of the "consumer", it sure does feel like we are being "ignored", "tricked", or "bamboozled" when our stated preferences are ignored in favor of "revealed preference".
It isn't that we have a "preference" for these things, it is far more likely that a user just doesn't have their guard up 100% of the time, and these psychological manipulations are designed to cut through that.
Sure, these strategies probably net clicks, but they aren't from people who "chose" your product, they are clicks from people who were manipulated into clicking.
I suppose whether you think that is okay depends on your industry and ethics.
It's behavioral marketing, vs status/aspirational marketing.
A stated preference isn't necessarily current or situational (I will choose to run instead of watching another 45 minutes of Youtube videos).
A situational preference is often inertia, and behavioral marketing will directly hinder the meta cognitive processes that usually give us the agency to override our default mode choices (John has been on YouTube for the last 20 minutes, what next suggestion is not likely to keep him there?)
Yeah, it is highly toxic. I'd assume that in most cases those "revealed preferences" are specifically engineered, not organic. It's taking advantage of biological reflexes and calling it a true preference.
I was going to say that too. Some of these I definitely am guilty of. I have a few dozen that aren't on the list but it's a breath of fresh air to see it so well organized even though, we all know what it is :D fantastic job to the author(s).
The most extreme virtue-signal is to go completely browser-default and have no styling whatsoever. Like lowercasing because your pinky can't be arsed to reach for the shift-key even though you've a billion dollars in series A.
I've mostly stopped caring about using using proper capitalization, commas, grammar and spelling in my writing of comments, primarily as a signal that i'm not an llm.
If you turn on HN's "Show Dead" setting, there are tons of LLM-generated comments on stories related to AI. You can see the human(s) behind the LLM trying to fiddle with the style of comment by making them skip proper grammar, capitalization, use or avoid certain phrases, and so on. The biggest tell for LLM content, though, is just the content as a whole: it sounds fake and ungenuine, like it passed through a committee of hostage negotiators to remove the speaker's own attachment/expectations.
They can configure it to use all lowercase letters, skip em-dashes, make grammar mistakes, stop saying "it's not X, it's Y", or whatever, yet the content itself just has a fake quality to it that makes it stand out, which is why those comments still get flagged IMO.
I had this conversation the other day. I'm a native German speaker originally, which is why I hand out commas like it's candy and capitalize things unnecessarily. Sometimes I notice these things and leave them in when I write something, since at least it gives you a good indication that a human wrote it... for now.
The problem is that omitting capitalization, commas, and so on signals, in addition to "not AI in default settings", but also "I'm part of the San Francisco AI in-crowd and Altman is my spirit animal".
Honestly I can just swap these bad boys in and ship in less than a couple hours if it'd be funny enough. I don't think they're bad designs at all, and I don't think every aspect of my business needs to be unique and obsessed over.
IMO this is like judging landscaping companies for all using similar looking shovels.
I get the whole trope thing and maybe I'm just an old man but I still am kinda impressed when Claude sh*ts out this type of UI 100 times faster than I ever could. It might also be that I never could have made UI even of this quality before AI. (˶ˆᗜˆ˵)
It's still better than the sh*t developers produced three years ago.
Some people just like to feel superior by shaming others' work. You can easily tweak the visual output if you want to, but it's good enough for most use cases and better than what developers used to produce.
“TokenStream – Server-sent events (SSE) were added to the HTML5 spec in 2008 but never used until 2025.”
I remember chunked transfer encoding shipped in 1997. It's been possible since then to readily and easily stream bytes of text or chunks of html the way everyone sees LLMs do today.
I used this to write a web based telnet client in 1997, and later a text moo / chat for the web. In both cases used a frameset so your line to send was at bottom of screen, the incoming lines were server-sent as things happened server side, and scrolled the client as new lines came in.
There were other things you could abuse before that, but less reliable.
The author should have AI set up a simple deployment to EC2 and Azure and make an endless series of semantically meaningless AI companies with web sites and submit them everywhere. The web sites should also do this themselves.
> Is this bad? Not really, just uninspired. After all, validating a business idea was never about fancy design, and before the AI era, everything looked like Bootstrap.
There is a difference between trying to craft your own design and just shipping with whatever defaults the LLMs output. And the same has been the case pre-LLM when using CSS/HTML templates.
I guess people will get back to crafting beautiful designs to stand out from the slop. On the other hand, I’m not sure how much design will still matter once AI agents are the primary users of the web.
The thing about humor is that you don't have to tell people when you don't get a joke, you can just quietly continue to live your life while you wait for your next chance to be temporarily happy.
It's like when a Youtuber's audience complains about how they're constantly asking you to subscribe. The reason it happens is because the statistics say it works.
At work we’ve been discussing whether to migrate off our home grown component library to Material UI. I shudder at the thought, personally. However, a compelling reason to use a ubiquitous framework is that the ubiquity means folks intuitively know how to interact with your product.
Like many of us I was born into a deeply customizable Internet, all of my websites were green or red on black. They were a glorious amalgam of fixed width fonts and <blink> tags. With occasional wingdings characters for fun and games and complex <table>/<tr>/<td> tags for really epic layouts. They were l33t, honestly ^_^
But, as time goes on and more and more people use this thing, converging on the one-true-UX feels like a net good thing assuming the fundamentals are right. To some degree the LLM-ization of the Internet is essentially the end game of squashing the personality out of the Internet which bootstrap started.
We’re on the cusp of spoken word being the core UX of computers with a fall back to reading the LLM transcript, neither of which benefits from <blink>
Not that I disagree with you, but I'll also offer a tradeoff.
When people expect to pick up your app intuitively, it can also just mean them using the app absent-mindedly, which can mean them skipping the manual and jumping straight to trying to tie up the support lines. Whereas if your ui asks for a user's full focus up front, yes there are downsides to that but they're also more engaged.
Note I'm not arguing that this is a good strategy. But given that so many people follow it I imagine it's not as bad as it appears on the surface.
As for this website, the best component is the ASCII animation in the hero and you can’t even copy that component. In fact, that nice ASCII hero is what gave me a good first impression to go thru all the components.
It isn't that we have a "preference" for these things, it is far more likely that a user just doesn't have their guard up 100% of the time, and these psychological manipulations are designed to cut through that.
Sure, these strategies probably net clicks, but they aren't from people who "chose" your product, they are clicks from people who were manipulated into clicking.
I suppose whether you think that is okay depends on your industry and ethics.
A stated preference isn't necessarily current or situational (I will choose to run instead of watching another 45 minutes of Youtube videos).
A situational preference is often inertia, and behavioral marketing will directly hinder the meta cognitive processes that usually give us the agency to override our default mode choices (John has been on YouTube for the last 20 minutes, what next suggestion is not likely to keep him there?)
And I'm gonna be honest, I kind of want to use a few of these components for real (the ASCII art is fantastic).
https://www.berkshirehathaway.com/
They can configure it to use all lowercase letters, skip em-dashes, make grammar mistakes, stop saying "it's not X, it's Y", or whatever, yet the content itself just has a fake quality to it that makes it stand out, which is why those comments still get flagged IMO.
It's really strange... I see some text with obvious tropes and sometimes I read something and there's no obvious AI trope... but it's just not human?
you're literate smart... poetic; because
you read e.e.cummings
and william carlos
williams
...
fin.
IMO this is like judging landscaping companies for all using similar looking shovels.
Some people just like to feel superior by shaming others' work. You can easily tweak the visual output if you want to, but it's good enough for most use cases and better than what developers used to produce.
So, it's progress.
I remember chunked transfer encoding shipped in 1997. It's been possible since then to readily and easily stream bytes of text or chunks of html the way everyone sees LLMs do today.
I used this to write a web based telnet client in 1997, and later a text moo / chat for the web. In both cases used a frameset so your line to send was at bottom of screen, the incoming lines were server-sent as things happened server side, and scrolled the client as new lines came in.
There were other things you could abuse before that, but less reliable.
But yeah, talk about things nobody used....
When in Rome!
My conclusion from that experiment:
> Is this bad? Not really, just uninspired. After all, validating a business idea was never about fancy design, and before the AI era, everything looked like Bootstrap.
There is a difference between trying to craft your own design and just shipping with whatever defaults the LLMs output. And the same has been the case pre-LLM when using CSS/HTML templates.
I guess people will get back to crafting beautiful designs to stand out from the slop. On the other hand, I’m not sure how much design will still matter once AI agents are the primary users of the web.
[0] https://www.adriankrebs.ch/blog/design-slop/