The user diligently updates Firefox, optimistically hoping that this new version will be as awesome as the previous ones that vastly improved performance. Firefox starts and, suddenly, a bunch of things that used to work stop working.
Why?
Well, essentially, something that they can't hope to comprehend and that they never asked for and had no input on just ruined their browser.
What do you think they are going to conclude? That extension developers are to blame?
Nope.
Their conclusions are going to be something along these lines:
"Those idiots (Mozilla) broke Firefox... again"
"Firefox 57 is a buggy piece of Turd"
"I'm tired of dealing with this shit. I'm switching to Chrome"
Users won't blame the extension developers. They will blame Firefox.
Mozilla should have done a better job with this transition. "Upgrading" an extension to use the new API requires a complete rewrite, AFAIK.
It is TOTALLY unreasonable to expect that a bunch of developers - who probably are doing this for free - are going to be able to quickly migrate their extension to the new API - especially considering that the API isn't even stable yet, and is missing a ton of functionality.
This is the mother of all breakages. Mozilla should've tried to smooth this transition, not just simply pull the pin on the web extensions grenade and yell "Catch!".
They screwed this up big time. Time will tell what this will do to their ever shrinking market share.
> Users won't blame the extension developers. They will blame Firefox.
What's funny to me is that these 'users' you describe are apparently on HN - I thought this place was mostly software devs, but it's striking how many posts seem to fundamentally misunderstand the decisions made by Mozilla.
> This is the mother of all breakages
I would be that somewhere above 90% of FF users will be unaffected. Given Firefox's market share that's still a lot of people, but let's not pretend like they broke everything overnight.
> Mozilla should've tried to smooth this transition, not just simply pull the pin on the web extensions grenade and yell "Catch!".
You're implying that this was an unexpected change that Mozilla was not forthcoming about. It is the opposite. We've all known about this for months.
The value prop of no longer being tied to an extremely old system is significant and you're not giving it any of its due credit.
the only plugin that broke for me was NoScript and I switched to uMatrix which I like even more. i'm not exactly a firefox power user i guess but i am a web developer who is definitely an advanced user, and i was barely affected by this.
The users' perspective, in short: before 57, I had one browser that was fast but lacked all the UI tweaks that I had grown accustomed to over the course of a decade or so (Chrome/Chromium), and one browser that maybe wasn't so fast but had all those UI tweak (Firefox, which I used all the time).
Now I have not one but two fast browsers with a UI that I do not like.
It's perfectly understandable that Firefox devs would love Firefox to be more like Chrome, but that has nothing to do with what the (existing) users want: they already have Chrome available, they are not holding their breath waiting for Firefox to become Chrome.
> "I'm tired of dealing with this shit. I'm switching to Chrome"
That was already happening, where "this shit" was crippling performance issues, and a lack of sandboxing and modern security features.
I expect Mozilla believed that the user fallout from breaking some extensions would be less than the existing continuing fallout of the ongoing issues.
> Mozilla should've tried to smooth this transition, not just simply pull the pin on the web extensions grenade and yell "Catch!".
You do realize that this transition has been going on and has been publicized for years, right?
1) Mozilla isn't retarded. They've done market research on this.
For example, in Sep. 2015 around 40% of users did not use any extensions at all. Another sizeable number of users is going to have their ad blocker and nothing else. Even with 2, 3 or 4 extensions, it's unlikely that you're going to experience a breakage, and if you do, it's likely that you'll find a replacement.
Average users rarely use unpopular extensions and popular ones either are maintained or will have a replacement made. There are some semi-popular ones that currently can't yet be fully recreated, but those are the types of extensions that change so much about the browser that average users won't be using them anyways.
2) Users aren't retarded. I know, we like to act like they are, but only the most cynical are going to switch browsers, because of this. Out of spite. It does not make any sense to switch to a different browser, just because the browser that you're used to has become different. Nor does it make sense to switch to Chrome, which is still by far less extensible than Firefox 57, just because Firefox has become somewhat less extensible.
3) The core of the API is more than stable. It's Chrome's extension API, that's been battle-tested in Chrome for years. Most extension developers will not need more than that. And it's most definitely not missing a ton of function, especially not things that non-power-users need.
4) Their market share is not anymore shrinking. It's been growing again since the release of Firefox 48. That was the release which shipped the first iteration of multiprocess. They could not have shipped multiprocess as early as that, if they did not know legacy extensions to be deprecated now with 57. Because the majority of legacy extensions are not multiprocess-compatible and neither would have been updated to be.
AMO would be reverse Russian Roulette where only roughly 1 out of 6 extensions will not kill your performance. That's just as well something you can't expect average users to understand and it would be like that for the next few years still.
So, yeah, they did rush this, but it was to save their market share. Had it continued to drop like in the half year before Firefox 48, we'd now be deep into negative user numbers.
> Even with 2, 3 or 4 extensions, it's unlikely that you're going to experience a breakage, and if you do, it's likely that you'll find a replacement.
This is simply not true as there had been many popular UI-centric extensions that just can't be replicated as webextensions due to lack of API support. "Advanced UI features belong in extensions, not in main" had been the Firefox mantra for many years and negativity about 57 is the logical consequence.
Which are the type of extensions that change so much about the browser that average users just won't use them.
When I discovered Classic Theme Restorer, Tab Mix Plus and similar, I was already a semi-pro user and I still found them intimidating.
There were a lot of checkboxes and they changed around a lot of things and I didn't yet know how to create a separate Firefox profile where I could've actually just wildly tried different options without the fear of something breaking irreversibly.
The user diligently updates Firefox, optimistically hoping that this new version will be as awesome as the previous ones that vastly improved performance. Firefox starts and, suddenly, a bunch of things that used to work stop working.
Why?
Well, essentially, something that they can't hope to comprehend and that they never asked for and had no input on just ruined their browser.
What do you think they are going to conclude? That extension developers are to blame?
Nope.
Their conclusions are going to be something along these lines:
"Those idiots (Mozilla) broke Firefox... again"
"Firefox 57 is a buggy piece of Turd"
"I'm tired of dealing with this shit. I'm switching to Chrome"
Users won't blame the extension developers. They will blame Firefox.
Mozilla should have done a better job with this transition. "Upgrading" an extension to use the new API requires a complete rewrite, AFAIK.
It is TOTALLY unreasonable to expect that a bunch of developers - who probably are doing this for free - are going to be able to quickly migrate their extension to the new API - especially considering that the API isn't even stable yet, and is missing a ton of functionality.
This is the mother of all breakages. Mozilla should've tried to smooth this transition, not just simply pull the pin on the web extensions grenade and yell "Catch!".
They screwed this up big time. Time will tell what this will do to their ever shrinking market share.
reply