15
Are artcraft apps considered slop or they have potential? (lemmy.ml)
submitted 1 day ago by to c/opensource@lemmy.ml
26 comments fedilink hide all child comments
 

I just tried pdf craft on fdroid and really liked it, but its on beta and only 2 versions released.

Is it going to be well maintained and become an alternative or just stay unstable ai dlop? It had more stars on github than some other pdf viewers.

Whats your take on that. I am excited to see some decent apps on linux. Especially for .appimage and launching instantly unlike photoshop.

you are viewing a single comment's thread
view the rest of the comments
[–] 21 points 1 day ago (15 children)

The current business model of the big LLM providers is unsustainable. They are burning money to get people hooked. Which means maintaining and developing these applications will become a lot more expensive, once the providers need to make money. If these projects don't have a solid plan for this event, they'll probably just get abandoned.

  • permalink
  • fedilink
  • source
  • hideshow 15 child comments
    •
  • [–] 12 points 23 hours ago* (last edited 11 hours ago) (3 children)
  • permalink
  • fedilink
  • source
  • parent
  • hideshow 3 child comments
    •
  • [–] 1 point 14 hours ago (2 children)

    Can you point to which part specifically are you referring to here? https://github.com/storytold/photocraft

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 11 hours ago* (1 child)

    Almost by definition, maintainability doesn't refer to a "specific part"

    https://www.softwareimprovementgroup.com/blog/artcraft-adobe-clone-ai-code-quality/

    I'm not interested in getting into an argument about this any further, look through their github issues, there's plenty to learn.

    As a note, if you think they're genuinely reviewing this much code at this speed, I think you need to consider the logistics of that.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 2 points 10 hours ago*

    Any project is a sum of its parts, and if you're talking about maintainability, then the parts you could easily point to would be things like high coupling, overly large file sizes, functions that are too big and should be broken up.

    Meanwhile, your own link shows that it's roughly within the ballpark of industry average. And if you look through issues for any large project on github you'll see similar types of issues. Given how new the project is, none of this is really surprising or out of the ordinary.

    As a note, nowhere did I suggest they do manual code reviews of everything at this speed. What I do think they have is a regression test suite which they build out as issues come up and they fix them.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 3 points 14 hours ago

    That's only true for the US though. Electricity costs in China are a third of what they are in the states, and open models are now becoming very competitive. We can also see here that DeepSeek is already close to breaking even. The other factor is that models continue to get more efficient, and even local models like Qwen 27b can do serious work. Especially, when combined with good tooling. A coding harness that uses Qwen with five different roles scores as well as Anthropic Fable on the benchmarks.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 0 points 23 hours ago (9 children)

    there's deepseek and a lot of other local models.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 9 child comments
    •
  • [–] 5 points 22 hours ago (8 children)

    Yes sure, local models exist. But somehow the big flashy projects use the big flashy models. Local models are usable, but not as capable as the big cloud models, to which they have tied their development practice.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 8 child comments
    •
  • [–] 4 points 22 hours ago (7 children)

    if you look at benchmarks the difference is almost negligible, and there is nothing tying your development practice to claude or whatever. i constantly switch between agents and nothing stops me or changes my workflow when i do this. a new model can pick up instantly where another left, and once the western model is doomed deepseek will still be there running on your local machine.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 7 child comments
    •
  • [–] 3 points 21 hours ago (2 children)

    My local machine is dead and I can't afford new ram.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 21 hours ago (1 child)

    and you owe that to big llm western models run on massive data centers, deepseek is here to disrupt that. once the imperial monoliths fall consumer grade products will become cheaper and you will hopefully be able to get a new machine. i say this from argentina my friend, my country besides being fucked has import restrictions on technology which make buying hardware even more expensive than for people in the anglosphere.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 1 point 10 hours ago (3 children)

    What if we just learned to code and wrote if ourselves and took some personal interest in the lines we produce?

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 3 child comments
    •
  • [–] 2 points 8 hours ago (2 children)

    why don't you learn to harvest your own food, tailor your own clothes, etc, etc. your argument is one against automation. without keeping in mind both are not mutually exclusive. you can do whatever you want while others do what makes them happy!

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •