I managed to find the exact 2010 bug report: https://issues.chromium.org/issues/40502297
Turns out I personally responded. (Did not expect that)
It was a bug in WebKit that was resolved a year later.
I have no memory of the bankruptcy closures 14yrs ago. He probably has a fair point. But anyway, details there for the curious.
IMO the title should probably have a "(2012)" suffix.
I have the opposite experience.
Only once have I taken the time to submit a Chrome bug report.
Chrome Canary updated and then refused to launch entirely, on Windows 11. I dumped the Event Viewer and application crash logs into the bug report.
It was fairly promptly fixed, and an overall positive experience.
- https://issues.chromium.org/issues/465823612
- https://chromium-review.googlesource.com/c/chromium/src/+/72...
I actually turned off bug reports for my open source project, after some loser got abusive with me because I told him “that’s not a bug report, that’s a support request, and I won’t support you”.
The problem with bug reports is that the majority of them are feature requests. Or they are support requests. Very rarely are they actual bug reports.
So, I turned off bug reports.
The way I handled them before the age of bug tickets was to have a mailing list, and to tell people who emailed me privately to instead send an email to the list if they wanted support. This way, the community as a whole could either support the person, or we could have a discussion about why I wasn’t going to implement their favorite pet feature “for fun and for free”.
Right now, in the age of AI-automated attacks based on 0-days, developers need to pause adding new features and make anything Internet-facing as secure as possible (I’m very lucky my little open source project has never had a remote code execution bug).
Any large commercial organization wants people to fill out surveys or give feedback on their new app or their call center person.
A genuine bug report is a professional contribution. These are commercial enterprises.
They should be paying people for reports in appreciation at least.
Vote with your feet.
> it came up as a "broken image" icon (which isn't a bug) and failed to display the alt text from the img tag (which is)
The test page he provided http://perl.plover.com/NPC/test.html renders correctly for me in Chrome on Mac (Firefox renders the alt text, but no broken image icons), maybe the bug report _is_ obsolete?
I've been on internal teams with internal bug trackers that have done the same thing. It's unbelievably short sighted. Same with closing a bug as "wont fix". The bug still exists, why close the ticket? Who does thst benefit and how?
I remember how I submitted bug in Gmail. I've checked it 6 years later, and it still was there.
This needed a (2012) appended to it. Old things of note can be neat but this is clearly confusing people.
I work at a place where if you have too many "open tickets", you get a message from someone in a managerial position telling you that you have too many open tickets ("thanks, I did not know that!"). The kicker is, however, our ticket assignment works on a best-fit basis, and I dip my hands in enough stuff a lot of cases are assigned to me, and I don't mind. What I do mind though, is having to close tickets just to close tickets to fill someone else's absurd and abstract and useless bottom-line. But the managers don't seem to understand the difference between closing tickets and helping people. I don't close tickets before the case is resolved, and I don't care that there is a ticket assigned to me five years ago about a GPU virtualisation problem that may have gone stale 2 years back when NVIDIA, bless their heart, updated their drivers. I can't attend to the ticket on a whim, and it's on a long back-log, but I _know_ the problem may still be there which is why I keep the ticket assigned to myself. And so on.
I am pretty sure Google has the same problem as I attribute to our work place (just to be clear, no I do not have a problem with too many tickets, we're storing garbage 100x the volume these tickets require as upkeep, and in a year it will be 100x more for the former).
I tested systems for years. Declaring bugs not fixed but making them features to work around is an old practice that Microsoft and other large companies are famous for.
Big corps appreciate your open source free as in lunch work. Their profits and bonuses got fat fat
...this really would be the perfect use case for AI, wouldn't it?
Of course actual code changes to something like Chromium need to be driven by humans, for all sorts of reasons. And I'm not suggesting Google should hand over all bug triage to a machine, or anything like that.
But if you've already decided, for better or worse, that you're going to auto-close old issues—maybe at least have an AI look at them first? The bug was clear and straightforward; an agent should be able to confirm that it still exists, either by reproducing it or by looking at code changes since the problem was reported. It doesn't even need to be a particularly strong model.
Surely Google of all companies has the compute for that?
In the analogy the neighbor is also a billionaire, making the interaction even worse.
[dead]
I feel like this was posted disingenuously to goad the assumption it's "AI files too many bugs so Google..." but this happened 15 years ago and is not relevant to anything. There's no reason to share this on HN.
Also every large tech company I've ever worked at has done this periodically. Especially when I worked on small teams that built internal tools. We'd have years of bugs and feature requests that we'd never be able to address. Half the open tickets were from people that had left the company. So every few years management would mass-close things with a "If this is still relevant re-open it."
I'm confused. Is this something that just happened or was it 2012 and the story was just posted yesterday for some reason?
> [ I wrote this in August 2012 and then forgot about it. ]
Edit: I think it is old based on the addenda
> A Mastodon colleague was complaining today about a similar practice, and I wanted to refer him to this article, but couldn't because I had completely forgotten to publish it.
Fair enough I guess, but odd choice to start the article with "18 months ago" and date it 18/11/2026.