Why is Firefox abandoning its strength in customizability?!

[toolong]

Jody, think of it this way. No snark intended here.

The world is moving on and changing. Like a world train passing by in a terminal.

You have two options. Get on the train and join up with the changes. Or? Stand by and watch the world pass you by. And be left in the past.

Think about it.

[JodyThornton]

No question. I agree. I've been experimenting with the Nightly 57 release.

[Frank Lion]

If it suited the Mozilla agenda they'd say 99% of users use CTR.....
FYI: https://blog.mozilla.org/addons/2011/06 ... -on-users/

Interesting page that.
Funny how that now longer is the situation today .

The 'ol numbers game.

...and when it doesn't suit, the numbers go down and that goes back over 10 years now.

Here's the latest - the public stats of CTR - https://addons.mozilla.org/en-US/firefo ... s/?last=90

Third week of June, all stats on all legacy extensions just, er, fell overnight by 20 - 25% or so. Firefox Complete Theme stats fell 80 - 90% ! Bizarrely, even Thunderbird theme figures changed, but there the drop was around 66%.

Not the first time by far and you don't hear about it, I suspect, because most extension/theme devs are just past caring, so it's only when someone waves stats around as 'evidence' of something, that it gets mentioned.

There's a sort of interesting footnote to this - whenever I've pulled my stuff off AMO in the past and hosted it myself, I often change the GUID of the theme. I forgot to do that with Metal Lion Vista, which means that AMO still track its Active Daily User figures. AMO forgets to fiddle those figures and to this day, I have a theme that was last updated in 2009 with around 3,000 ADU...which is more than the now stated stats of a couple of the current 2017 themes I have on AMO.

[DanRaisch]

"There are three kinds of lies: lies, damned lies, and statistics." (Attributed to Benjamin Disraeli but actual author unknown.)

[billyswong]

...and when it doesn't suit, the numbers go down and that goes back over 10 years now.

Here's the latest - the public stats of CTR - https://addons.mozilla.org/en-US/firefo ... s/?last=90

Third week of June, all stats on all legacy extensions just, er, fell overnight by 20 - 25% or so. Firefox Complete Theme stats fell 80 - 90% ! Bizarrely, even Thunderbird theme figures changed, but there the drop was around 66%.

Not the first time by far and you don't hear about it, I suspect, because most extension/theme devs are just past caring, so it's only when someone waves stats around as 'evidence' of something, that it gets mentioned.

There's a sort of interesting footnote to this - whenever I've pulled my stuff off AMO in the past and hosted it myself, I often change the GUID of the theme. I forgot to do that with Metal Lion Vista, which means that AMO still track its Active Daily User figures. AMO forgets to fiddle those figures and to this day, I have a theme that was last updated in 2009 with around 3,000 ADU...which is more than the now stated stats of a couple of the current 2017 themes I have on AMO.

Wow, I can only say, wow.

[GHM113]

Opinion from the other side of the barricades:
https://www.reddit.com/r/firefox/commen ... s/dllybzq/

CTR has about 300k users. We can't justify adding that many features directly into Firefox (which incidentally, will slow Firefox dramatically) for 300k users.

So basically bye-bye if your addon has less than 300k users

[Mark12547]

I guess you have to put me in the "show me" category.
Show me how MP improves my browsing experience.

A year ago I stopped using Firefox because whenever I opened several tabs at once (by using a folder in my bookmarks, right+click, "Open All in Tabs"), Firefox would freeze, any video or audio stream playing at that time would stop, and all of Firefox would be frozen until the biggest resource eater, loading my overly large Netflix DVD queue, would complete.

I would try the same thing in Chrome, a multi-process browser with a process per tab (up to 20 content processes, and after 20 it starts using the processes for multiple tabs), the video or audio stream would continue playing, most of the tabs would load quickly and would start responding, and my overly large Netflix DVD queue would load in less time than it took in Firefox. In other words,

• The UI in Chrome would remain responsive, and Chrome even has a workable task manager that I can pop up at any time to see what tab(s) is (are) using which process and the amount of CPU, memory and network resources it is using. (I filed a bug report with Mozilla about their about:performance to reporting CPU-intensive processes, and even when it works it's less helpful.)
• As soon as each tab finished loading, I could interact with it (whereas with Firefox all tabs would have to had finished loading),
• Any tab that was playing a video or playing an audio stream would continue to work without the least bit of stutter,
• The longest loading page (Netflix DVD queue) would take less time to finish loading, and
• Chrome would make full use of the four cores of my CPU when it had four or more tabs it was trying to load simultaneously, whereas Firefox would use just slightly more than the equivalent of one core even when all those tabs were being loaded at once.

(It didn't hurt that every extension I was using in Firefox had the same or equivalent extension in Chrome.)

But starting in Firefox 53 (with experimental about:config options and careful choice of multi-processing-compatible extensions), then in 54, and especially now in Nightly 57, Firefox has become almost as fast as Chrome, and every bullet point above now also applies to Firefox (other than the Netflix DVD queue still takes longer in Firefox than in Chrome, but the time difference has narrowed quite a bit, and I still prefer Chrome's Task Manager over Firefox's still a bit unreliable about:performance).

So I am now back on Firefox and several times a week I really like Firefox still working, still rendering an audio or video stream without any stutter, while at the same time it is opening a collection of tabs that includes my big Netflix DVD queue. This was possible due to multi-processing.(One could argue that Firefox could have been written to multi-thread and let each thread run in a different core, but it wasn't; but future upgrades to Firefox are likely to increase parallelism by doing just that as part of the attempt to get pages to render even faster.)

But if one is working with just one tab, or interacting with just one tab at a time and the other tabs don't have any work to do until you switch to them, then multi-processing isn't making maximum use of the CPU, but it would still allow the UI/main process to function even when the content process(es) is (are) locked up or crashed, and that could lead to better diagnostics or at least better isolation of problems, and if using multiple content processes when a web page in one tab causes one process to crash (such as when I scroll down a never-ending pages of pics), tabs rendered by the other processes are unaffected and still continue to work.

All the above is what I have observed, not theorized about. Firefox Nightly 57 with multiple content processes when accessing multiple pages while at the same time playing a YouTube video or a DAB audio stream works far better than back a year ago when all of Firefox was one single process with the exception of the Flash container (which, if playing Flash content, was launched in a second process).

[Aris]

Regarding add-on stats in June 2017, Mozilla announced this back then: https://blog.mozilla.org/addons/2017/06 ... tatistics/

I suspect Mozilla heavily manipulating add-on stats since a few years now to support their own course for sure. Ever since themes user numbers started falling massively even before lw-theme/Personas became "themes" and actual themes, that are more than just two images, became "complete themes", something smelled fishy.

When stats tell the story Mozilla wants to tell, the explanation to drop any feature because of too few users is always valid. Its not like manipulating stats and polls is a new thing and its definitively not just restricted to user stats or Mozilla, but that is a different story.

I remember a Firefox dev telling "an add-on with over 400k users can not just be ignored and just dropped", but we past that now. There was even a time CTR reached almost 500k users for a while between September 2015 and June 2016. http://i.imgur.com/NVcLUV1.png

[impar]

Problem with current web-ext addons is that they offer very poor range of abilities compared with current addons. Its feels like a race to the bottom.
Some of the current "legacy" (:lol: at the name) addons will never be made under web-ext because they just cant (missing APIs or whatever), and even if a current addon can be made to work in web-ext the developer might just not care. There are a bunch of addons with very low userbase that wont be updated and still have no current web-ext replacement.
A perfect example for this is the time saver addon "tumblr hi-res" (works in v55, even if AMO says it doesnt)
https://addons.mozilla.org/en-US/firefo ... lr-hi-res/
For the time saved it is worth the change to Chrome\Vivaldi\whatever, ending 13+ years of Firefox use.

[flaneurb]

Just to say a big Thank You to Aris for all he's done for Firefox users. It's a pity what's happening now.

[Frank Lion]

Regarding add-on stats in June 2017, Mozilla announced this back then: https://blog.mozilla.org/addons/2017/06 ... atistics/I

The user count is a very important part of AMO. We show it prominently on listing and search pages. It’s a key factor of determining add-on popularity and search ranking.

They using this one again as a reason for 'adjustment'? Must be 10 times in as many years now and they still haven't, apparently, managed it.


https://blog.mozilla.org/addons/2017/06 ... ent-224029

Lightweight themes are the majority, yes, but their usage stats are calculated separately. They won’t be affected by this change (I failed to mention that). Usage for complete themes will fall significantly, though, but there aren’t many active themes now.

Now, there's a surprise.

Btw you read this from the Dynamic Duo? -

http://forums.mozillazine.org/viewtopic ... #p12595781

Hi guys, I'm a member of the Mozilla add-ons team. I just want to say now that we're as unhappy about the decreased visibility of full themes as you are and are working to do something about it. We just found out about this change today.

Post Posted 12 Jan 2013 12:13 am
To expand a little on Kris's post, the changes were based on an old design and we weren't aware they were going to be pushed live this week. We're making some adjustments that should be applied to the site soon.

kmaglione? I guess the k stands for Kellyanne or something. For the record, here's what that page looked like in 2013 - https://addons.mozilla.org/en-US/firefox/themes/ and here is what it looks like in 2017 - https://addons.mozilla.org/en-US/firefox/themes/

Opinion from the other side of the barricades:
https://www.reddit.com/r/firefox/commen ... s/dllybzq/

If you worked for Mozilla and had to put out that crap, where would you rather do it? On reddit, where Game of Thrones and console games are endlessly discussed with rabid ferocity by the idle or here?

Mozilla employees are not, almost, but not totally daft.

[Grumpus]

The newer addons page looks similar to the Amazon new software downloads Canonical applied to Unity which ran every time you went on line with Ubuntu only it's not on your system (yet).

[Aris]

...
kmaglione? I guess the k stands for Kellyanne or something. ...
...

Kris Maglione
Add-ons Technical Editor
Mozilla Corporation

[Frank Lion]

...
kmaglione? I guess the k stands for Kellyanne or something. ...
...

Kris Maglione
Add-ons Technical Editor
Mozilla Corporation

Yep, I know. It is my 'topical humour'.

I often throw a few gags into my posts and that was one to amuse our U.S. members here.

[Aris]

Ah OK, didn't get the joke as I'm from Europe