all 6 comments

sorted by: hot top controversial new old
[–] 11 points 1 year ago* (4 children)

https://metr.org/blog/2025-07-10-early-2025-ai-experienced-os-dev-study

Nobody ever adresses that using AI might actually be less efficient (while feeling more productive to the users)

  • source
  • hideshow 4 child comments
  • [–] 10 points 1 year ago (3 children)

    It might. It might help too (something your link explicitly states). It may be circumstantial.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 7 points 1 year ago* (2 children)

    Yeah thats the point I was trying to make.

    Why is it talked about so onesidedly when it is actually multi-facetted? The blog to me reads like: "there are legal reasons against AI and there are fanatic reasons against AI, so we yive you the option to disable it (but actually it's totally great as we all know and agree)."

    I wish they'd addressed it in a more nuanced way or not at all. Just saying "It is now a config" would have been enough, but they went out of their way to point out only the subset of the arguments that can be easily dismissed.

    Guess they have a product to sell that rides on the succsess of AI coding.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 3 points 1 year ago (1 child)

    I don't see why they should need to address all the concerns specifically if they've provided the option people want anyway?

    It seems they don't really force you to use the AI anyway so people could just ignore it if they wanted?

  • source
  • parent
  • hideshow 1 child comment
  • [–] 6 points 1 year ago*

    You can do that just now? "More configurable" my ass, everyone who keeps wanting me to switch to zed.

  • source
  • 3 comments in c/privacy@lemmy.ml view that post
    [–] 15 points 1 year ago

    This really should've been there from the start once they began adding AI tools. Or at least a pick_a_model: None option.

  • source
  • [–] 6 points 1 year ago (1 child)
    2 comments in c/zed@programming.dev view that post
    [–] 1 point 1 year ago*

    I've been using zed for nearly a year now, it works well, quietly.
    I don't need AI, I have a complex project written in scala, whose compiler and tools ('metals') are already intelligent and much more logical than LLMs, many context-relevant suggestions pop up, and when it compiles it generally just works.
    However I do wonder about the longterm business model of Zed without AI, and in the current geopolitical context I'd be interested in a european or global alternative.

  • source
  • [–] 0 points 1 year ago

    While I like this I am still in fear of the upcomming enshittification we have seen with so many software projects. So many "ideas" have a shit motivation, like said AI that can be deactivated..yeah. But I think there is something wrong with the devs. Why do it in the first place? Why slap AI on and just later allow to disable it?

    Same for the TERRIBLE integration of facist data exfiltrator Gravatar. We all hate gravatar. Why share data with crooked gravatar....run by the guy that forces gutenberg on anyone? why share data with crooked US companies in the first place?

    I think the devs are on a wrong path like vibecoders just adding every shit because they think they can but in the end it was a waste of everyones time.

    I do no longer have high hopes for Zed. In Zed i click on help...what do I see: "Zed on Twitter". On their page: Twitter↗ Bluesky↗ YouTube↗ Discord↗ Reddit↗

    This is the ppl you do not want to play with - everything US based, everything giving a shit about your privacy. No fediverse, matrix, mastodon accounts...just the usual scumbags. So when I read their values: https://zed.dev/values https://zed.dev/team

    I think we should call a doctor to get them help.

  • source
  • 1 comment in c/technology@lemmy.world view that post
    [–] 11 points 1 year ago

    I swear "How to disable AI in x..." is all our lives now. 😬

  • source