4 ms·

I'm sure there are plenty of bugs like this in other browser but yeesh... Unfixed for 20 years because it was an implementation bug no one in authority was wil

I'm sure there are plenty of bugs like this in other browser but yeesh... Unfixed for 20 years because it was an implementation bug no one in authority was willing to fix. No other browser even supports it! Ugh.

thrdbndndn 5y ago
There are plenty old bugs yes but does this one really count as a bug? It categories as "enhancement" and I personally agree it's more like a feature request.
yoz-y 5y ago
Feature requests like this is precisely the place when saying "it's open source, just do it yourself" is a valid remark.
- Cthulhu_ 5y ago
  Don't you first need some kind of consensus where whoever is responsible says "Yes, we will accept this"? I mean these 20-year bugs are a smell where there is indecision, I don't think it's a matter of nobody wanting to go into the code. Decide if it has to be fixed or changed, or explain why you won't, and close the issue. (on that note, I don't like the "this issue is stale -> closed" on GH either, unless the maintainer(s) replied or asked for additional information and the OP abandoned the issue)
  - thesuitonym 5y ago
    If your change is good, you can create consent by making the patch available. People will install it and eventually Mozilla will implement it. In this case though, I doubt you'd get much traction because nobody really cares.
- Kuinox 5y ago
  It took me more than half a day to get the repo cloned & make it compile. I won't attempt to contribute to firefox again.
  - senkora 5y ago
    Fun fact: Firefox was the first program I ever compiled. I’d just finished an online Python class and wanted to get involved in this open-source thing. Mozilla had some Python tasks but they still required a dev build of the browser. I was on Windows at the time. It took me three days to figure out, and thinking about cygwin still makes me shudder. Then I saw that the Linux instructions were just a one-line apt command for dependencies and then the build command. It looked magical. This is what convinced me to move to Linux and exposed me to so much cool software and ideas. I never did end up contributing to Mozilla.
- AshamedCaptain 5y ago
  And then you do it yourself and your patch is never accepted because it disagrees with the other developers. So you try to make an extension which implements the behavior you'd like, and it works pretty well for a couple years. But then they decide to completely remove their extension API, and instead give you one that doesn't even offer 0.1% of the hooks that the old extension API offered. So at the end of the day you're forced to maintain your own private fork. Fortunately, the other developers attention is focused on other areas of the codebase these days (hopefully ones that are irrelevant to you), so you don't have to do a lot of conflict resolution. Eventually, you switch to a source-based distribution where you can pile patches upon patches on each of the packages you use, and they are auto-reapplied every time you upgrade the package. Such is the life of the "open-source software user with an itch" these days. It's still better than the alternative... at least on paper.
dathinab 5y ago
Is that even a bug? It's more like a unwanted feature hardly anyone uses and as such hardly anyone ends up caring about in either direction. On the other hand changing the icon through JS is something people do care about (unread notification indicator) for example. I seen one or two cases from "still icon" -js-> "animated icon" -js-> "still icon2" which where ok (or did they just do a lot of js refresh??). I can't remember any side abusing changing or animated icons, tbh. So if I would manage Firefox I probably wouldn't fix it either because it's not worth the time investment needed, while on a project the size of a browser there are basically always relevant issues to fix. (Exception maybe if I have to touch the code in question anyway and remember the issue.) And the age in the end doesn't matter if there are always docents of way more relevant bugs, they always will get priority. It only would matter if I care about irrelevant stats.

thrdbndndn 5y ago

There are plenty old bugs yes but does this one really count as a bug? It categories as "enhancement" and I personally agree it's more like a feature request.

yoz-y 5y ago

Feature requests like this is precisely the place when saying "it's open source, just do it yourself" is a valid remark.

Cthulhu_ 5y ago
Don't you first need some kind of consensus where whoever is responsible says "Yes, we will accept this"? I mean these 20-year bugs are a smell where there is indecision, I don't think it's a matter of nobody wanting to go into the code. Decide if it has to be fixed or changed, or explain why you won't, and close the issue. (on that note, I don't like the "this issue is stale -> closed" on GH either, unless the maintainer(s) replied or asked for additional information and the OP abandoned the issue)
- thesuitonym 5y ago
  If your change is good, you can create consent by making the patch available. People will install it and eventually Mozilla will implement it. In this case though, I doubt you'd get much traction because nobody really cares.
Kuinox 5y ago
It took me more than half a day to get the repo cloned & make it compile. I won't attempt to contribute to firefox again.
- senkora 5y ago
  Fun fact: Firefox was the first program I ever compiled. I’d just finished an online Python class and wanted to get involved in this open-source thing. Mozilla had some Python tasks but they still required a dev build of the browser. I was on Windows at the time. It took me three days to figure out, and thinking about cygwin still makes me shudder. Then I saw that the Linux instructions were just a one-line apt command for dependencies and then the build command. It looked magical. This is what convinced me to move to Linux and exposed me to so much cool software and ideas. I never did end up contributing to Mozilla.
AshamedCaptain 5y ago
And then you do it yourself and your patch is never accepted because it disagrees with the other developers. So you try to make an extension which implements the behavior you'd like, and it works pretty well for a couple years. But then they decide to completely remove their extension API, and instead give you one that doesn't even offer 0.1% of the hooks that the old extension API offered. So at the end of the day you're forced to maintain your own private fork. Fortunately, the other developers attention is focused on other areas of the codebase these days (hopefully ones that are irrelevant to you), so you don't have to do a lot of conflict resolution. Eventually, you switch to a source-based distribution where you can pile patches upon patches on each of the packages you use, and they are auto-reapplied every time you upgrade the package. Such is the life of the "open-source software user with an itch" these days. It's still better than the alternative... at least on paper.

dathinab 5y ago

Is that even a bug? It's more like a unwanted feature hardly anyone uses and as such hardly anyone ends up caring about in either direction. On the other hand changing the icon through JS is something people do care about (unread notification indicator) for example. I seen one or two cases from "still icon" -js-> "animated icon" -js-> "still icon2" which where ok (or did they just do a lot of js refresh??). I can't remember any side abusing changing or animated icons, tbh. So if I would manage Firefox I probably wouldn't fix it either because it's not worth the time investment needed, while on a project the size of a browser there are basically always relevant issues to fix. (Exception maybe if I have to touch the code in question anyway and remember the issue.) And the age in the end doesn't matter if there are always docents of way more relevant bugs, they always will get priority. It only would matter if I care about irrelevant stats.