Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Firefox bug gets fixed after 25 years

bugzilla.mozilla.org

Post reply on HN
Print view View original

101–110 of 176 posts

Re: Firefox bug gets fixed after 25 years

#102

Post by aeontech »

johannes1234321 wrote: post #92

With such old bugs I often wonder whether it's not best to simply close them. Few people will browse them and many will be "accidentally" closed by reworking of different subsystems. But if they are still relevant somebody will file it, again. If not reopened it's probably not a problem anymore.

The problem with that is the dismissal of the time investment by the original filer who may have spent a significant amount of time getting reproducible steps, screenshots, and other relevant information into the bug. It treats users' time as free.

I think there's a subtle but important distinction between _closing_ the old bugs outright, and leaving them open but marking them as "Verify" with a comment like "we're not sure but it may have been fixed with recent work - can anyone confirm if it's still reproducible?"

Re: Firefox bug gets fixed after 25 years

#103

Post by mjevans »

beeboobaa3 wrote: post #74
chungy wrote: post #26

Earlier quoted context omitted.

It looks weird to me too, but it's hardly indecipherable. You have a four-digit year (2000), there's only 12 months in the year, so the "28" is the day and "03" is the month. In ISO 8601 format, it'd be 2000-03-28. In American written format, it'd be March 28, 2000.

What's weird about the most common date format in the world, dd-mm-yyyy?

ISO8061 or RFC3339 are the only correct date formats; though if for some forsaken reason your ledger starts accounts on the right hand page and treats the left as part of a prior record, then placing the least frequently changing part of the date to the edge of the page is good as a primitive index.

https://xkcd.com/1179/

https://ijmacd.github.io/rfc3339-iso8601/ https://stackoverflow.com/questions/522251/whats-the-differe...

Re: Firefox bug gets fixed after 25 years

#104

Post by bmicraft »

SECProto wrote: post #94
sapiogram wrote: post #85

Earlier quoted context omitted.

Earlier this years, Github would routinely say "2 years ago" for comments made 11 months ago.

I've experienced (not on GitHub) things that happened 729 days ago being "one year ago", which is similarly incorrect

It's not correct, but at least not unexpectedly so

Re: Firefox bug gets fixed after 25 years

#108

Post by fuzzy2 »

johannes1234321 wrote: post #92

With such old bugs I often wonder whether it's not best to simply close them. Few people will browse them and many will be "accidentally" closed by reworking of different subsystems. But if they are still relevant somebody will file it, again. If not reopened it's probably not a problem anymore.

I've opened a few bugs, mostly surrounding extensions. Bugs from the XUL days still exist today, even though the UI is different and the API is different and everything. It is somewhat astonishing. I filed most of them over 10 years ago.

So no, just closing them because they'll be accidentally fixed isn't the correct argument IMHO. Instead, they should just admit that most stuff will never get fixed and just clutters the list. Even though that hurts.

Re: Firefox bug gets fixed after 25 years

#109

Post by hedora »

mg wrote: post #8

The one long term open bug that irks me the most is that you cannot properly date-format the x-axis of scatter plots in LibreOffice: https://bugs.documentfoundation.org/show_bug.cgi?id=54132 12 years old. According to the comments, fixing it is discouraged as the code is so "horribly complicated", it would likely cause regressions.

If you just need a spreadsheet and graphing package, gnumeric is a good choice under Linux.
Post reply on HN