Why Firefox for eCS-OS2 didn't went up from version 10?

[Martin Iturbide]

Hi

I have this question since I didn't followed the mozilla OS/2 port forum/newgroups.

What was the technical limitation because the Firefox ports didn't went from version 10 to 11 and follow on?

Regards
Martin

[Andy Willis]

Primarily IPC, interprocess communication.  It was in 10 for everyone else be we could back it out.  IPC makes it so that each tab runs in its own process so that each tab is protected from the other.  There are a few other things that Dave is working on too.

[ivan]

I had assumed it was because our versions are following the commercial ESR line which makes it up to date.

[Dave Yeo]

What was the technical limitation because the Firefox ports didn't went from version 10 to 11 and follow on?

Short story, by FF10 I was the only one left building and halfway through 11 it broke and I didn't have the skills to fix it.
Long story, after FF4 they dropped support for non-ipc (interprocess-communication allowing tabs to run in their own process so when Flash crashes it would only take out the tab it was running in, Flash crashing is a problem on all platforms) and mostly Walter kept the ipc patch alive. This was a very tedious process as the patch grew to 280KBs (980+ lines) + supporting patches of about 75kb which had large amounts that needed to be applied by hand. Walter did an excellent job of keeping the patches up to date but had personal problems and got tired of rebasing the patches and moved on.
The other problem was development moved on under the assumption of IPC existing so more things didn't work. I had to remove various things like battery status that were aimed at mobile devices anyways. Eventually most of the HTTP code was rewritten in such a way that I didn't have the skills to fix it.
Also I go through phases of working on it, then getting sick of hours of typing #ifdef MOZ_IPC #endif, then trying to figure out the remaining unresolved symbols.
At this point I feel that without IPC support which Rich estimated at a man-month of work for a skilled developer we won't go on. It is being worked on though not by me.

[Dave Yeo]

I had assumed it was because our versions are following the commercial ESR line which makes it up to date.

That was as much luck as anything though I did have the goal of getting 10 out for the reason that it would be supported for over a year.
The rapid release cycle made it very hard for a couple of people to keep up. Perhaps one person working on it full time would have but after Peter moved on and Mozilla switched to the release every 6 week schedule...
Mozilla has a lot of people employed developing Firefox and we've never had more then 3 or 4 since IBM moved on. And it can be tedious.

[Paul Smedley]

And it can be tedious.

As can working on any software be

[ivan]

That was as much luck as anything though I did have the goal of getting 10 out for the reason that it would be supported for over a year.
The rapid release cycle made it very hard for a couple of people to keep up. Perhaps one person working on it full time would have but after Peter moved on and Mozilla switched to the release every 6 week schedule...
Mozilla has a lot of people employed developing Firefox and we've never had more then 3 or 4 since IBM moved on. And it can be tedious.

Dave, if that was luck then we must thank our lucky stars you made that decision.  The ESR branch is the most stable and much less prone to problems, so much so that I have converted all my friends to using it - even if they are using windows.

Thank you for your luck and all your hard work.

[Ben]

Thank you for your luck and all your hard work.

Oh yeah, for sure, Dave.

While I seldom hear it voiced, I have no doubt that all of the OS/2 community feel only kindness, and think the best of thoughts for you, because of your generous work.

Everyone here knows that  OS/2 could only be less useful without your personal efforts.

For myself, I say "thank-you", though I know that many more are thinking it too...

... and I have no doubt that we all feel the same for all of the other programers who voluntare their time as well, like Paul Smedley and Steve Levine.

.

[Silvan Scherrer]

The other problem was development moved on under the assumption of IPC existing so more things didn't work. I had to remove various things like battery status that were aimed at mobile devices anyways. Eventually most of the HTTP code was rewritten in such a way that I didn't have the skills to fix it.
Also I go through phases of working on it, then getting sick of hours of typing #ifdef MOZ_IPC #endif, then trying to figure out the remaining unresolved symbols.
At this point I feel that without IPC support which Rich estimated at a man-month of work for a skilled developer we won't go on. It is being worked on though not by me.

IPC will gome very soon we hope. We start working on FFox >10 (probably 17 or 18 direct) at the beginn of February. IPC was done since very long for VPC, but not many ppl know of it and Mozilla refused to include the patches back then.
So stay tuned for an official announcement.

regards
Silvan (team leader)

[Ben]

IPC will gome very soon we hope. We start working on FFox >10 (probably 17 or 18 direct) at the beginn of February. IPC was done since very long for VPC, but not many ppl know of it and Mozilla refused to include the patches back then.
So stay tuned for an official announcement.

That is great news, Silvan.

I for one, (of many no doubt), am looking forward to it.
There are some good improvements in later versions... like not reloading all saved tabs on startup at once, but waiting until clicked.

But with FireFoxs new, messed up, numbering system it is hard to tell a significant upgrade from an incremental increase.

And how long will it be before they get into three digits?.. already being at 18... a few years?