Live data from Hacker News

Where Are the Vibecoded Photoshops?

indiepixel.de

321–330 of 410 posts

Re: Where Are the Vibecoded Photoshops?

#321

Earlier quoted context omitted.

For others who are confused. The author is someone in the demoscene who has been accused of "vibe coding" large pieces of work. His counter is that if such things are possible, where is the vibe coded Photoshop? He then goes on a tangent that "vibe coding" is a kind of easy accusation mostly levelled at "neurodiverse" developers Which is why he ends the post: So. Where are the vibecoded Photoshops? WHERE IS THE THREA…

AH! That brings the whole thing into context. So to charitably paraphrase the argument, it's something like: "If I used AI to produce something technically impressive, my human input was still an important part of the process. Nobody has vibecoded (AI-generated without human guidance) a photoshop, which is evidence that the human input/oversight is still important. Therefore my AI-assisted work is important/impressiv…

Yes, I think so. Although it's a poor argument in my view: just because nobody has vibe coded Product X, doesn't mean his Product Y was praiseworthy

Re: Where Are the Vibecoded Photoshops?

#322

Earlier quoted context omitted.

Exactly this. I have an acquaintence who is a wine connoisseur and collector. He has done technical project management, but does not program himself. Over the course of several months, he has produced an app that manages a database of the wines he has. It's a lot more that just a CRUD-app. In addition to maintaining the obvious data (name, year, winery, notes, etc.), it can take a photo of the label, parse it, and fi…

> 6 months of fiddling with a couple of different AIs in his spare time Six months of fiddling and $X00 in subscription/token fees to make a DIY inventory management app that's going to need regular attention and revision, with ongoing service fees, to accommodate not-quite-right implementations and hidden technical debt. That's a toy for wealthy hobbyists, not a revolution. The industry needs to deliver on a lot mor…

> That's a toy for wealthy hobbyists, not a revolution.

Isn't this exactly how many technologies get started?

A PC used to cost as much as a very nice used car, and had few practical uses cases, as one example.

Re: Where Are the Vibecoded Photoshops?

#323
post #163

Earlier quoted context omitted.

what about the data? is it locally hosted? if he drops his phone everything will be lost? or are servers and databases involved? if so, where are they being hosted? how did he manage those? even these sorts of stories are incredibly shallow and hard to believe for me personally.

If someone has any curiosity, they can ask the AI about this and it will engineer a solution, like use iCloud or some free tier service. After all, it's basically how us software engineers arrived to where we are today. It's hubris to think nobody else has the interest nor attention span to walk a solution incrementally to its conclusion, esp when they don't know what the final solution will look like ahead of time.

> After all, it's basically how us software engineers arrived to where we are today

An increasingly smaller minority of us. The vast majority have gone through a bootcamp, or an undergrad or similar to gain specific skills that they can apply writing software for a corporation.

There's hardly any reason to believe the percentage of the general public that reach that level will be any higher than in the profession itself.

Re: Where Are the Vibecoded Photoshops?

#324
post #34

I don't think that's the direction this is going to take. Replacing a mature app with an incredibly wide audience and a million use cases may or may not be possible, but what's actually happening is that people are making an app that does exactly what they want, and using it to solve their own needs. Previously you might use Excel to take raw data from various places and analyze it, create charts, reports, extract fi…

Yeah, the pain/reward ratio is against vibecoded replacements for mature tools. Piracy is cheaper than tokens.

But over the next 5 years I expect a growth in Blender-like open-source projects aiming to take on the big closed-source elephants. Code is cheaper now. The main downside of LLM coding, unmaintainable spaghetti code, can be mitigated effectively with discipline and coordination.

You still need maintainers to uphold contribution standards, but people will throw tokens at you. A small, disciplined team can go a long way, make a decent enough product, and then attract the institutional money (like Blender did) and hit that growth curve where everyone rallies you and you've won.

Lots of companies would have a vested interest in reducing these dependencies to Adobe et al., or have a more customizable product. Competitive professional tools, more like Blender and less like GIMP, but in other areas, like DAWs, CADs, and others.

Re: Where Are the Vibecoded Photoshops?

#325
post #96

Earlier quoted context omitted.

Funny how that works both ways. "I am 100x productive with AI, I can ship in days what took months before" and "oh no, you don't understand, it's not really possible to vibecode Photoshop" found elsewhere in this thread. If I were magically 100x productive, the first thing I would do would be to recreate Photoshop.

Can I ask why? Specifically, why Photoshop and not Canva, or Figma, or Lightroom? I ask because Photoshop probably wouldn’t have a market if it was introduced in 2026. What would you be hoping to accomplish by recreating Photoshop?

There is nothing close to Photoshop (or the Affinity suite) that runs natively on Linux. Despite how niche that market is, one could make big bank if they could magically create a valid alternative out of thin air. I’d pay hundreds of dollars for it myself. I don’t even care about PSD compatibility.

(Nope, Krita and GIMP are not good enough)

Re: Where Are the Vibecoded Photoshops?

#326
post #268

Earlier quoted context omitted.

There’s zulip https://zulip.com/

That's a SaaS / Product. https://zulip.com/self-hosting/ Similar to the GitLab model. Example limitations: > Mobile notifications without limitation for eligible communities > Mobile notifications for up to 10 users for other organizations

The app is open-source, so I guess it would be OK to deploy it yourself and thus register for your own notifications pipeline.

Re: Where Are the Vibecoded Photoshops?

#327

Earlier quoted context omitted.

Just to be clear, I spent Billions of tokens last month, dick'ing around with AI, do I think it can do my job, sure if I was as checked out as I am feeling being mandated to only use AI for work... Trust me this version of me can't write code better than an LLM. I think even the LLMs from 2 years ago could do as much. Only thing we have solved with LLMs is how to generate code, as in literally converting what you typ…

Same mistakes Indian IT Service firms already did, trusting third-party AI Service providers across every division and department in companies. Now Open AI and Anthropic are launching own Service firm/wing for maximizing ROI. Direct Competition. Huge Loss for Indian IT Service Firms. They paid AI providers to Train own automated competition end to end so much so that they are learnt the gaps and how-to own the market…

I couldn't parse all of it, but largely I honestly think software is a sum of it's details you write better software the more details you add to the picture while maintaining simplicity. The art of software is balancing the two.

You can also find this in science, I feel like software is about finding the Gauss’s law for the dozens of electrostatic principles the masses keep coming up with.

AI strips away the need for simplicity by making brute forcing details via verbosity that ruins the need for "good" software, and makes all software by definition unmaintainable since maintainable product as a unit has to be simple enough for a LLM or a human mind to understand.

Currently a human context window is in 100s of billions of tokens, while LLMs seem to max out at 1M.

For anyone thinking but humans can't produce all of the text verbatim, that's the difference b/w humans and machines we can parse and take the most essential parts from 100B token context it's just are speed of prefil and recall is slower. And we memoize learning so if we know X then all variants of X are learnt on the same foundation.

Human learning is so very different from LLMs is so very fascinating as a trad "AI engineer", well some person who has worked and built/shipped ai agents and models at scale.

Re: Where Are the Vibecoded Photoshops?

#328

I think it's an interesting conversation; there's an assumption that "vibe coding is complete and incapable of creating Photoshop alternatives". But it's more like, still open to debate but sounds likely: how long until vibe coding is powerful enough to generate Photoshop alternatives quickly? So yeah I think people would concede the point that vibe coding isn't quite all there yet; but is it there at the size of a f…

Where are the vibe coded Photoshops? Here is a good place to start: https://github.com/light-and-ray/awesome-alternative-uis-for... AI has obliviated both the need and the desire for discrete pixel manipulation outside of the existing professional disciplines. It has been replaced by the explosion of diffusion models with a direct human language --> pixel image loop.

Nobody who wanted to erase a tree or fix their smile paid for Photoshop anyway. and Photoshop does a a lor more than pixel manipulation. So, a moot point.

Re: Where Are the Vibecoded Photoshops?

#329

Earlier quoted context omitted.

Can you share a link to your app?

I recently had a coworker open my eyes to why vibe coding, or AI-assisted coding is so popular. He likened it to a slot machine, where pulling the slot's arm is like asking an LLM to code something. You get crap most of the time, but when it works, it's like getting a payout. That dopamine hit keeps them pulling, hoping for another hit, and they then believe it's a better way to develop software.

Oh that’s definitely it. Even Steve Yegge mad with vibecoded powers has said it feels a lot like gambling. I wouldn’t trust the judgment of gamblers talking about their ‘winnings’, so to speak.
Post reply on HN