It's like finding out the open source community's got black mold
-
@mcc@mastodon.social oh this is just a github moment. well the one commit is an obviously correct one-line change that was PR'd
@mcc@mastodon.social i have no idea why the PR author apparently needed claude to change literally a single function call from
panictozfs_panic_recover, but. -
@mcc@mastodon.social uhh are you sure
-
@mcc@mastodon.social oh this is just a github moment. well the one commit is an obviously correct one-line change that was PR'd
I deleted because if there was any chance of error, I didn't want to leave it up, but a PR'd one line change is still enough for me to complain.
The thing is, my understanding is you have to use the tool in a particular way for it to actually create the committed-with header. For more complex uses I believe the header does not show up.
-
I deleted because if there was any chance of error, I didn't want to leave it up, but a PR'd one line change is still enough for me to complain.
The thing is, my understanding is you have to use the tool in a particular way for it to actually create the committed-with header. For more complex uses I believe the header does not show up.
-
It's like finding out the open source community's got black mold
@mcc their vibe-coded slop (zfs) versus my honorable, ornately written tapistry (xfs)
sgi stays winning (i am gonna be shattered if it turns out xfs ALSO uses ai code, my goats had failed me if so) -
@mcc their vibe-coded slop (zfs) versus my honorable, ornately written tapistry (xfs)
sgi stays winning (i am gonna be shattered if it turns out xfs ALSO uses ai code, my goats had failed me if so)@det I'm a bit concerned about RHEL.
-
@mcc@mastodon.social @algernon@come-from.mad-scientist.club (it appears to have been the only commit by that user in 9 years, fwiw)
-
@mcc@mastodon.social @algernon@come-from.mad-scientist.club (it appears to have been the only commit by that user in 9 years, fwiw)
@mcc@mastodon.social @algernon@come-from.mad-scientist.club but also i mean in this case not merging it would be weird, would the alternative be to leave the bug there out of spite?
-
@mcc@mastodon.social @algernon@come-from.mad-scientist.club but also i mean in this case not merging it would be weird, would the alternative be to leave the bug there out of spite?
@leo @algernon If it were my project i would ban the contributor. One line or not they are not following the DCO as written which this project requires. So either the DCO is violated, the DCO has a "you can do anything as long as it's through a hole in a sheet" exception for AI, or the DCO is not enforced at all, or most likely this crept up on OpenZFS and they have never thought about this and did not notice the claude line.
-
@leo @algernon If it were my project i would ban the contributor. One line or not they are not following the DCO as written which this project requires. So either the DCO is violated, the DCO has a "you can do anything as long as it's through a hole in a sheet" exception for AI, or the DCO is not enforced at all, or most likely this crept up on OpenZFS and they have never thought about this and did not notice the claude line.
-
-
It's like finding out the open source community's got black mold
Growing in the cruft.
-
@leo @algernon If it were my project i would ban the contributor. One line or not they are not following the DCO as written which this project requires. So either the DCO is violated, the DCO has a "you can do anything as long as it's through a hole in a sheet" exception for AI, or the DCO is not enforced at all, or most likely this crept up on OpenZFS and they have never thought about this and did not notice the claude line.
@mcc@mastodon.social @algernon@come-from.mad-scientist.club for what it's worth ibm (who is admittedly definitely a biased party here) has espoused the position that the DCO does not ban AI but if you include a Signed-off-by line on an AI-generated commit you are explicitly saying that you take personal liability if a court ends up ruling that AI output infringes copyright
-
@mcc@mastodon.social @algernon@come-from.mad-scientist.club for what it's worth ibm (who is admittedly definitely a biased party here) has espoused the position that the DCO does not ban AI but if you include a Signed-off-by line on an AI-generated commit you are explicitly saying that you take personal liability if a court ends up ruling that AI output infringes copyright
-
@leo @algernon Anyway IBM's position on is AI is enough to cause me to not use their linux distribution even though I think it would work on my laptop better
… I mean also their position on the GPL wrt RHEL and whether RHEL is following it. Basically my view of their lawyers right now is about the way I'd view a crowbar wedged into the community center cashbox
-
@mcc@mastodon.social @algernon@come-from.mad-scientist.club for what it's worth ibm (who is admittedly definitely a biased party here) has espoused the position that the DCO does not ban AI but if you include a Signed-off-by line on an AI-generated commit you are explicitly saying that you take personal liability if a court ends up ruling that AI output infringes copyright
@mcc@mastodon.social @algernon@come-from.mad-scientist.club linux upstream explicitly allows AI contributions as of december and there have been quite a few since https://kernel.org/doc/html//next/process/coding-assistants.html
-
It's like finding out the open source community's got black mold
@mcc I hate it, and I immediately knew what you meant, and I hate that too.
-
It's like finding out the open source community's got black mold
@mcc I like that the mold is now visible rather than hiding under the carpets.
These people were always who they are now. And we treated them as rockstars or at least as reasonable bosses.
-
@mcc Not to mention people who are very, very excited to huff black mold spores.
@xgranade@wandering.shop @mcc@mastodon.social huffers: “I got you a smoothie! It’s black mold spore, my favorite!”
“… wait, what?” -
@mcc@mastodon.social @algernon@come-from.mad-scientist.club linux upstream explicitly allows AI contributions as of december and there have been quite a few since https://kernel.org/doc/html//next/process/coding-assistants.html
@leo @algernon I thought this was only for documentation, for now?
To be clear, it is my intent to eventually move entirely to software which is either no-AI by license or forked from pre-AI code. If it means I have to throw away Linux I throw away Linux. Linux is an inferior product I am *only* using because Windows added AI. Why the fuck am I using Linux if it's also AI?