infixed

I've been thinking a lot about building onboarding experiences, and was reminded of this presentation that I read about 10 years ago.

The gist, that the same principles that make a video game where you save a princess fun to learn (e.g. super mario bros) can be applied to building products, seems so obviously true - yet difficult to put into practice.

The most influential part of this presentation is this quote:

When you build applications that let users be smart, they love you for it. The secret to good game design is simple. Set up situations where there is a problem that must be solved and let the user solve it. Give them subtle clue, but don’t take away that ‘aha’ moment.

Scary part: you have to believe the user is smart.

infixed

3 anni fadiscuss

I think the prose in the pre-amble is a bit over-flowery and heavy handed (e.g. LLMs really aren't that expensive, I very much doubt the WSJ claim that Copilot is losing money per user, LLMs aren't always "painfully slow", etc.)

Having said that, the actual recommendations the article offers are pretty reasonable:

- Do as much as you can with code

- For the parts you can't do with code, use specialized AI to solve it

Which is pretty reasonable? But also not particularly novel.

I was hoping the article would go into more depth on how to make an AI product that is actually useful and good. As far as I can tell, there have been a lot of attempts (e.g. the recent humane launch), but not a whole lot of successes yet.

infixed

3 anni fadiscuss

This topic is about fluid simulation, but it also references shaders as a way to performantly implement fluid simulation.

I've played around with shaders in the past to build particle simulations with millions of points, and it's always really tickled my mind. You basically write functions that operate against a big 2d grid of colors. But because colors are represented by a 4x1 vector [r,g,b,a], you can repurpose this pattern to do general purpose computation (e.g. you can represent a point in a 3D coordinate space as a [r,g,b] color).

You can see this in the codepen in this post. It's literally creating "materials" and "render targets" that are actually just intermediate computation steps in the fluid simulation.

  class Fluid {
      constructor(context) {
          this.context = context;
          this.speed = 4;
          this.forceInitMaterial = this.createShaderMaterial(forceInitFrag);
          this.divergenceMaterial = this.createShaderMaterial(divergenceFrag);
          ...
          this.divergence = this.createRenderTarget();
          this.advection = this.createRenderTarget();
          ...

infixed

3 anni fadiscuss

Is it poor advice though? I don't think the article is suggesting to skip designing a nice landing page -- the article is saying that customizing the _login page_ is not worth the effort, which I agree with.

I can say that at my last company, redesigning the login page was bottom of the stack of things we wanted our first designer to work on. We just had a centered "Login with Google" button, our logo, and a email + password form. What else do you need?

infixed

3 anni fadiscuss

Agreed with this sentiment. I personally find it funny that Craigslist is still very much alive and kicking, and is my go-to for apartment hunting amongst other things, despite an underwhelming mobile experience and a design that hasn't been updated in literally decades.

infixed

3 anni fadiscuss

In their conversation, they bring up mobile as a recent platform shift that caused a lot of disruption. But I think it's interesting to remember how slow that took. The iPhone was released in 2007, but the real winners of mobile were launched years later -- Uber (2010), Snapchat (2011), and TikTok (2016) -- and those winners took several more years to even start to gain true traction in the market.

I don't think a lot of people back in 2007 could have predicted that the biggest thing to come from mobile would be an app that let teens remix music videos and share with their friends.

This is why I think it is a little pointless to try and create mental models for what products and features to build to capitalize on AI (though it can be fun). It's so early that we're not capable of understanding what's possible yet. If anything, we're probably at the viral "fart app" stage that mobile was in for its first few years.

infixed

3 anni fadiscuss

I think the main argument of this essay is that if you're an early stage company with no customers, it's not a bad thing to churn out features to see what sticks in the market because you don't have the luxury of customers to talk to. And that in this case, people might falsely complain that they are a "feature factory."

I feel like that's a bit different from what most people think of when they hear "feature factory." When I hear feature factory, I think of an engineering team that has zero input into the product process and just builds whatever PMs or leadership says is important.

In the case of an early startup with no customers, I think if engineering teams get no justification, aren't involved in talking with customers, they are completely within their rights to complain about feeling like a "feature factory." The right solve isn't to say "actually -- we have no customers, so shipping a bunch of stuff isn't feature factory mentality" but to actively engage the team in product discovery conversations with users.

infixed

3 anni fadiscuss

There was an interesting thread that did the rounds on HN a couple years back about using Faker in a related (but different) product.

https://news.ycombinator.com/item?id=27252066

Personally, I think your demo has a pretty neat / novel UX. I couldn't find a way to generate a large number of examples at once -- did I miss that somewhere?

infixed

3 anni fadiscuss

One of the issues with AI products is that they often require a good amount of input to use. It seems to me that a great AI product would save me time and do things for me without being asked to.

For example, at our company we have a quite a few of alerts set up. Datadog also automatically detects anomalies. It would be neat if this (or something else) could automatically do an initial triage without being prompted and give me a free headstart on issues that come in.

Otherwise, it feels like it's "work" to learn how to use the product, which seems to miss the promise of AI (doing things for us!).

Submissions

Mixing Games and Applications [pdf]

Evidence for the earliest structural use of wood at least 476,000 years ago

Earliest Carpenters

State of AI in Production

Interview with Humane Founder Imran Chaudhri

Google Cloud TPU Multislice Training

Retool Workflows is now generally available

Haskell: Syntactic Heroin

comments