Google Chrome dev team declares bug report bankruptcy (2012)

84 points by speckx 4 hours ago on hackernews | 39 comments

Google Chrome dev team declares bug report bankruptcy

[ I wrote this in August 2012 and then forgot about it. ]

About 18 months ago I filed a report for a bug in Google Chrome. It wasn't a very important bug. I reported that when Chrome was asked to inline an image, and the image turned out to be in the obscure and seldom-used X-bitmap format, 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).

I then forgot about it until this week, when I got a notice that the report was being closed because it was "obsolete". I was invited to resubmit the bug report if I considered it valid.

My first draft of this article continued politely "The Chrome dev team botched this in a number of ways." That draft came out boring, weak, and vague, so I'm going to try continuing a different way.

That was a really jerky thing to do. I put in my time and trouble to construct a clear bug report with an easy test case; the bug report contained a link to a page that demonstrated the problem. To see if the problem is still present, all you have to do is click the link in the bug report.

But after I spent my time and effort to submit this bug report, Chrome dev decided to ignore it. After they had ignored it for a year and a half, they decided to pitch it in the trash. But if I want, I can submit it again. Now why on earth would I want to do that? Does Chrome think I enjoy assembling bug reports? Maybe I do it for fun? No. The bug report has no value to me. It was an expense. It was a donation to someone I thought would appreciate it. Suppose my neighbor has a boat. Sometimes he lets me use the boat, but it leaks. One day he asks if I will help me fix the boat. I spend the day in his garage painting and sealing some new boards to put into the bottom of the boat. Some time later, I see he has put the new boards into the trash. "What did you do that for?" I ask. He shrugs. "I don't really care if the boat leaks," he says. "But if you want to come over and paint and seal some more boards, how about next week?" Sorry, what am I supposed to get out of this transaction? Chrome dev says they have 40,000 bug reports, too many to deal with. So what? Autoclosing them isn't dealing with them either; it's throwing them in the trash. If having you don't want to get bug reports, yo udon't have to solicit them. They say that a lot of the reports are useless. I'm sure that's true. Most people don't file very good bug reports. A few people do. Who gets pissed off when you throw all the reports into the trash? The people who put in the most effort, that's who. So if you're trying to adopt a strategy that will ensure that your next 40,000 reports are even worse, this is the strategy to adopt. What should Chrome have done here? There is an easy thing and a hard thing.

The easy thing not to say "We threw away your report, but if you think it was a good one, please start over and resubmit it," which is likely to piss off anyone with any sense. Instead, you stick a button on the web site and say "We threw away your report, but if you think it was a good one, please click the button." I would not object to checking if the report was still valid, and clicking a button to reactivate it. The harder thing that they should have done was to not close the reports at all. Managing bug reports is hard. Perl was years behind on bug reports for many years. Many of those reports were indeed obsolete. But there is no cost to keeping them around.

Addenda

  1. 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.
  2. JWZ complained about something similar in 2003.

[Other articles in category /tech] permanent link