bigblind

If I'm totally honest, this feels like somethign that'd be quite low on my priority list. I'd muchrather focus on the on-screen UX and accessibility if I have limited time. I think making things printable makessense in certain contexts, and I think could actually replace a lot of the pDF generation sites do for things like tickets, invoices, etc.

So, it's not launching in the EU, again because of DMA rules and Apple not giving access to the same tools to 3rd-party AI companies.

I get this, but I find the difference in application of the DMA pretty weird. In august, the EU came to an agreement with Apple on their developer policy in a way that still really feels like it goes against the spirit of the law, and still puts an insane tax on thousands of European developers.

Yet they seem to play hardball on enforcing a rule that currently just negatively impacts a handful of AI companies but leaves consumers waiting for features. The prioritization feels weird.

I don't think this was only about implementation being the bottleneck. It was also because implementation forces you to think about the thing you're building at a very grannular level of detail, at which point you might discover flaws or opportunities that weren't obvious at the higher level.

And I find that, when it reasons through and encounters some of those, it just makes assumptions rather than asking the user.

There's somethign very black mirror about an office floor full of people softly whispering to their computer

Sheesh... what's in a name? :p

Same here, that's the main value I get from especially music streaming services: I hear about anartist, and can immediately go check otu their discography.

i find the writing style a bit peculiar at times. Take this lecture on bloom filters for example: https://academa.ai/lectures/bloom-filters

It suddenly starts talking about URLs, but it never really sets up the idea that we're using some kind of URL lookup service as an example. The way it introduces the memory limit feels similarly strange. At some point, it just says that we have 8GB of memory, but it does that halfway through the explanation, not as part of the setup of the problem we're actually solving.

Quite fun to play with.Nice work!

This looks interesting, but I'm also getting to the point wherethings like the video having a clearly AI-generated voiceover turns me off a bit. Like, if you're excited to tell me about your product, tell me yourself.

I think you can be present in the moment, but still find value in a tool like this. I may not be "the norm" because I'm legally blind, but in some sense, a tool like this could make me even more present. I currently use my phone a lot to zoom in on signs to read them. I'd love it if I wouldn't have to hold a phone in front of my face, just to know whether the place I'm standing in front of is, in fact, Jimmy's Barbershop, or whatever place I'm trying to navigate to for the first time.

Thinking completely selfishly, as a visually impaired person, I want these to succeed, because they can be a great assistive tool. And assistive tools being general consumer products makes them cheaper.

Let's start a $1.3bn goFundMe, and invite all billionnaires with a conscience (do they even exist?) to put some money in. + an extra $20 in postage to mail a turd to donny

I feel like they do and they don't at the same time. The buildings may look different, but city center rents driving out a lot of small local businesses, and leaving the same brands everywhere.

It doesn't do great on recent events it seems. The amusement park Walibi Belgium recently announced a company called RMC is doing a makeover of their wooden roller coaster, so I did a search for "walibi belgium rmc", and it found one very out-of-date article about earlier rumours, and a bunch of less relevant stuff.

Then you become a poletician, though I guess that could also apply to lying to people who want to be lied to.

I'm curious whether people feel like UX could at all be a differentiator here. like, sure, i can build a very quick crud app with Ai in an hour, but there are lots of UX decisions that, if not prompted in the right direction, AI just handles badly.

i guess the problem is that once a teamp goes through the process of figuring out good UX for a certain flow, which can take time, that UX then becomes trivial to copy.

I think you've just summed up late stage capitalism.

This looks interesting! I understand not wanting to put out a narrated tour as the video, but being visually impaired, i find video demos without narration, that constantly move around/focus on different things hard to follow. It still might be worth putting a short screencast with you actually walkign through usijng the product and narrating it.

It's unfortunate that this is necessary. It should be obvious that wearing noise cancelling headphones in trafic, including as a pedestrian, is a bad idea.

I'm legally blind, so I have my own bias here, but I think people really over-rely on sight. If you do want to listen to something while walking around a city, I can highly recommend bone conduction headphones, that keep your ears unblocked.

If you just bell once or twice, and don't aggressively keep ringing, I'd never consider a bicycle bell in a shared space rude. I even consider it good manners, though as others have said, that varies between cultures.

Being visually impaired, though, I'm grateful for cyclists who use their bell. It's immediately clear. For some reason, my brain takes slightly longer to process someone yelling "on your left!" or similar, than just a quick "ring ring".

It is a benefit if you're a stakeholder in those companies, or your friends are stakeholders and will pass on some of the winnings as a "thank you."

I guess? At least there you can review the plan, but is this planning mode any better at making architectural decisions than when you prompt an LLM and let it make the changes directly? (it might be, just not sure.)

> On projects where I have no understanding of the underlying technology (e.g. mobile apps), the code still quickly becomes a mess of bad choices. However, on projects where I know the technologies used well (e.g. backend apps, though not necessarily in Python), this hasn’t happened yet, even at tens of thousands of SLoC. Most of that must be because the models are getting better, but I think that a lot of it is also because I’ve improved my way of working with the models.

I wonder whether at some point we'll get a translation model, that translates relatively vague requests into sound architectural decisions, with some embedded knowledge of the environment you're building in, and that can ask clarifying questions when there are multiple options with different tradeoffs.

Here's an example of one of their dev logs: https://www.youtube.com/watch?v=pdym24sg1HQ

I guess it's too late for an "Eternal sunshine of the slopless mind"?

I like this a lot!

One thing I found as a somewhat experienced solver of cryptic crosswords, is that it was confusing that I couldn't just type an answer. I skipped the tutorial/lesson, because I didn't want an explanation of how cryptic clues work, I just wanted to solve one. So a little, inline explanation for first time visitors on how to solve them on this site might be good.

This is an interesting idea, and I'd love to see whether people come up with interesting stories/imaginations with this, but I feel very strongly pushed to engage before I've even properly explored the platform to see the potential. Like, all the stories seem to have their images locked, I assume until I make an account. Then there's also this mention of a pro subscription and some deal for the. firlst N subscribers. I'm not even sure yet if this a platform that's worth enough for me to make an account, let alone pay money for. I understand that image generation isn't cheap so you need to think about monetization early, but I suspect you'll need more ways for people to see the value of it without the commitment of even creating an account to have this become successful.

I never considered the volume of output tokens from dev tools, but yeah, I like this idea a lot.

I can certainly believe that this is really an agent doing this, but I can't help that part of my brain is going "some guy i his parents' basement somewhere is trolling the hell out of us all right now."