I am sorry to say Graphene has fallen to the slop.

[piffio@mastodon.social]

@haverholm @davidgerard

Great resources, will check them all out!

[davidgerard@circumstances.run]

@zimzat also it never, ever stays true. It's another "i'm totally not vibe coding, I'm" [list of things anyone else would call abject vibe coding]

[zimzat@mastodon.social]

@engideer @davidgerard Thanks, that specific message is a lot closer to an explanation of what's going on. It's the libertarian "personal responsibility" CYA.

I agree with all the rest too, it's a slippery slope. I'm not ready to throw out the baby with the bathwater, but it's going suspiciously sideways fast.

Honestly I've been suspicious of Graphene for a while because its proponents have a tendency to be shills. Top results for "Degoogle" on YouTube are often fear mongering.

[engideer@tech.lgbt]

@zimzat Yep def same, wouldn't know of an alternative that checks all my use cases as of now, unfortunately. Same goes for the Linux kernel etc after all. But at least one can adjust their expectations accordingly knowing this

[helpstertee@social.helpsterte.eu]

@davidgerard@circumstances.run oh nice they are not even arguing against the social and environmental problems and just talk about tool efficiency.

guess as long as Soylent Green meets the nutritional macros…

[lettosprey@tech.lgbt]

@shimst3r We have AI tools at work doing these kinda "review tasks" automagically, and report potential issues.

My experience with them is that they do work, they highlight things that is wrong and should fix, things that can be a potential security issue, etc. But so far, I have yet to see it highlight something that a non-AI static analysis tool would find with a fraction of the resources.

It can bug about things that isn't an issue, this is something that the "old" tool does as well. So, all in all, higher resource use for about the same output.

What can LLM tools do that the old tools could not? Of course, they can suggest a fix.

That part is the dangerous one. Using LLM to detect potential issues should not come at a high risk, though the benefit compared to existing tools now seems low. This might improve over time, I dunno...

Using them to suggest patches has a high risk of patches being applied without enough critical evaluation. Not ok.

@davidgerard

[dysfun@social.treehouse.systems]

@FediThing @contrasocial @davidgerard yes but have you considered the new models change everything?

[haverholm@radikal.social]

@piffio Happy to share! @davidgerard

[davidgerard@circumstances.run]

@dysfun @FediThing @contrasocial literally their excuse is, we have to use the chatbot or we'll be left behind

[slotos@toot.community]

@shimst3r @davidgerard

There are two possibilities with AI: either it’s a shitty coder or a really good one. The latter is way more terrifying, when you consider the small circle that controls it.

Funny how people that should’ve known why that is the case turn out to have been running a theater all along.

[elexia@catcatnya.com]

@davidgerard really wish I wasn't reliant on having a smartphone, it's all so grim

[pndc@social.treehouse.systems]

@davidgerard @dysfun @FediThing @contrasocial Why should GrapheneOS even care that they're being "left behind"? "Left behind" by whom? Is the point of the project merely to one-up similar projects on features at any cost, including at the expense of quality and security? That's not a great proposition for its users, who should perhaps instead start looking at the competition to see what GrapheneOS is so desperate to catch up with.

[frutiger@mastodon.social]

@davidgerard If ycombinator is the source on grapheneos issues, then it's not true. Some of you sound like religious nutjobs when talking about ai with words like 'tainted', a horseshoe version of twitter users who are psychotically pro ai.

Edit: you blocked me, but this was my first fedi account, so yes it's masto.social. Is that the best you've got, mister terminally online who replies in 30 seconds?

[rogerbw@discordian.social]

@pndc @davidgerard @dysfun @FediThing @contrasocial They want some of that sweet VC money (that has already been thrown away elsewhere)?

[davidgerard@circumstances.run]

@frutiger@mastodon.social mastodon.social, 14 followers, thanks slopbot

the source is reading their fucking AGENTS.md you dipshit

[davidgerard@circumstances.run]

@RogerBW @pndc @dysfun @FediThing @contrasocial yeah @zzt points out the fucking weird Motorola deal as the obvious proximate cause

[iain@kolektiva.social]

@davidgerard “The output is filled with hallucinations and incorrect details but an experienced developer can sift through it, identify the real issues it uncovered and get those fixed.”

Because every developers favorite part of using code analysis tools has always been sifting through all the false positives.

[davidgerard@circumstances.run]

@iain yeah that bit was very obviously someone pasting in a marketing excuse

[sharpcheddargoblin@reclusive.blog]

@davidgerard So how fucked up is /e/os with respect to LLM use? GrapheneOS is out of consideration, and I considered them to be the best de-Googled Android option, unfortunately.

[barubary@infosec.exchange]

@iain @davidgerard Reminds me of the C standard, which requires a conforming implementation to output at least one diagnostic for invalid code, but says nothing about valid code. So if your compiler simply prints "your code is bad" in all cases, even if the code is fine, it is conforming.