←back to thread

1125 points CrankyBear | 1 comments | | HN request time: 0s | source
Show context
woodruffw ◴[] No.45891521[source]
I’m an open source maintainer, so I empathize with the sentiment that large companies appear to produce labor for unpaid maintainers by disclosing security issues. But appearance is operative: a security issue is something that I (as the maintainer) would need to fix regardless of who reports it, or would otherwise need to accept the reputational hit that comes with not triaging security reports. That’s sometimes perfectly fine (it’s okay for projects to decide that security isn’t a priority!), but you can’t have it both ways.
replies(13): >>45891613 #>>45891749 #>>45891930 #>>45892032 #>>45892263 #>>45892941 #>>45892989 #>>45894805 #>>45896179 #>>45897077 #>>45897316 #>>45898926 #>>45900786 #
ryandrake ◴[] No.45892032[source]
> But appearance is operative: a security issue is something that I (as the maintainer) would need to fix regardless of who reports it

I think this is the heart of the issue and it boils off all of the unimportant details.

If it's a real, serious issue, you want to know about it and you want to fix it. Regardless of who reports it.

If it's a real, but unimportant issue, you probably at least want to track it, but aren't worried about disclosure. Regardless of who reports it.

If it's invalid, or AI slop, you probably just want to close/ignore it. Regardless of who reports it.

It seems entirely irrelevant who is reporting these issues. As a software project, ultimately you make the judgment call about what bugs you fix and what ones you don't.

replies(1): >>45892816 #
vacuity ◴[] No.45892816[source]
But if it's a real, serious issue without an easy resolution, who is the burden on? It's not that the maintainers wouldn't fix bugs if they easily could. FFmpeg is provided "as is"[0], so everyone should be responsible for their side of things. It's not like the maintainers dumped their software on every computer and forced people to use it. Google should be responsible for their own security. I'm not adamant that Google should share the patch with others, but it would hardly be an imposition to Google if they did. And yes, I really do intend that you could replace Google with any party, big or small, commercial or noncommercial. It's painful, but no one has any inherent obligations to provide others with software in most circumstances.

[0] More or less. It seems the actual language is shied from. Is there a meaningful difference?

replies(2): >>45893319 #>>45895167 #
ryandrake ◴[] No.45893319[source]
Well, it's open source and built by volunteers, so nobody is obligated to fix it. If FFmpeg volunteers don't want to fix it or don't have the time/bandwidth to fix it, then they won't fix it. Like any other bug or CVE in any other open source project. The burden doesn't necessarily need to be on anyone.
replies(1): >>45893540 #
lenerdenator ◴[] No.45893540[source]
They aren't obligated to fix CVEs until they're exploited, and then, suddenly, they very much were obligated to fix the CVEs, and their image as FLOSS maintainers and as a project are very much tarnished.
replies(2): >>45894619 #>>45895878 #
saagarjha ◴[] No.45895878[source]
If they are unable to fix CVEs in a timely manner, then it is very reasonable for people to judge them (accurately!) as being unable to fix CVEs in a timely manner. Maybe some people might even decide to use other projects or chip in to help out! However, it is dishonest to hide reports and pretend like bugs are being fixed on time when they are not.
replies(2): >>45898244 #>>45899518 #
walletdrainer ◴[] No.45898244{3}[source]
Please don’t use “CVE” as a stand-in for “vulnerability”, you know much better than this :)

Most vulnerabilities never get CVEs even when they’re patched.

replies(1): >>45898417 #
saagarjha ◴[] No.45898417{4}[source]
Only using it because the comment I replied to is, of course I agree that most vulnerabilities are patched without one
replies(1): >>45898492 #
1. walletdrainer ◴[] No.45898492{5}[source]
While this feels like it’s perhaps bordering on somewhat silly nitpicking, the trend of conflating vulnerabilities with CVEs is probably at least mildly harmful. It’s probably good to at least try not to let people get away with this all the time.

The way many (perhaps most) people think of CVEs is badly broken. The CVE system is deeply unreliable, resulting in CVEs being issued for things that are neither bugs nor vulnerabilities while at the same time most things that probably should have CVEs assigned do not have them. Not to even mention the ridiculous mess that is CVSS.

I’m just ranting though. You know all this, almost certainly much better than me.