Rapid Release Follow-Up

October 3rd, 2011

Rapid Release

My recent post on the rapid release cycle generated a lot of response, some very thoughtful and some also very frustrated. Many of the comments focus on a few key issues listed below. We’ve been working on how to address these issues; I’ll outline our progress and plans here.

large deployments that certify software before permitting use can’t manage a 6 week cycle
add-on compability issues
update notices and fatigue
frustration that we didn’t get these things addressed better before making the change.

1. Large Deployments. We’ve made a proposal for extended support for large deployments.This proposal is under discussion now in the relevant newsgroup and in our Enterprise Working Group. We are incorporating feedback and expect to come to closure on this proposal shortly.

2. Add-On Compatibility. There are a couple of related issues that have made add-on compatibility difficult. First, we have historically assumed that add-ons are incompatible until proven to be compatible. This is a very conservative assumption which creates work for all add-on developers and notifications to all add-on users. We’ve corrected this for the add-ons hosted by Mozilla. Work is underway to correct this for the remaining add-ons. Here is a more detailed explanation of the topic; feature planning details are also available.

3. Update Fatigue. In the past we have been very careful to make sure people know something is changing with their web browser before it changes. We did this to make sure people are aware and in control of what’s happening to their environment. Our position was to err on the side of user notification. Today people are telling us — loudly — that the notifications are irritating and that a silent update process is important. This work is underway. The first set of improvements should appear in the next Firefox release, with more improvements appearing in the next few months. Also, one main reason people are notified of updates is due to incompatible add-ons which will be addressed by the work on add-on compatibility. More details can be found in this blog post: http://www.brianbondy.com/blog/id/125/mozilla-firefox-and-silent-updates

4. Frustration. The comments also registered frustration that we didn’t get these issues better addressed before making the shift. The change was abrupt and we should do better in the future. We focused very effectively on making sure we could make the core engineering aspects of a rapid release process work. We focused well on being able to deliver user and developer benefits on a much faster pace — we’ve already brought major memory improvements to make browsing faster, Do Not Track to Firefox for Android, developer tools and HTML5 support. But we didn’t focus so effectively on making sure all aspects of the product and ecosystem were ready. We believe we have plans in place to alleviate the issues that resulted, with improvements rolling out in in the coming weeks.

Categories: Mozilla

91 comments for “Rapid Release Follow-Up”

1
Corp. IT said on October 3rd, 2011 at 3:11 pm:

Mitchell,

Thank you; this message is more important to me than the technical details. As you know, businesses are concerned not only with the product but with having a reliable partner. The earlier disjointed messages, some unprofessional, along with the lack of official Mozilla response to that situation (no clarification, no accountability, no clear strategy, etc. — if there was a response, it wasn’t well publicized) created some doubt. What you just wrote is the kind of communication we need.

Congratulations on winning the Tom’s Hardware review. It’s just one publication’s unscientific review and I don’t want to exaggerate its meaning, but it’s a sign that the rapid updates are paying off.

Corp. IT
2
J. McNair said on October 3rd, 2011 at 3:55 pm:

Well, it’s not an apology, but perhaps one doesn’t need to apologize for quality (why, hello there, Firefox 7). Thanks for the admission of responsibility, and for the clear descriptions of what your fine (though sometimes frustrating) organization will do to make things better for ALL users.
3
sysKin said on October 3rd, 2011 at 6:41 pm:

I think the part where mozilla representatives told community members to leave the community if they don’t agree with current approach (whatsmybrowserversion com) didn’t help either.
4
Mitchell said on October 3rd, 2011 at 9:33 pm:

Corp IP, J McNair, SysKin — As you note, there are also improvements in how we communicate that would help both people interested in Mozilla and those of us trying to make it work well. Another thing to work on.
5
davide ficano said on October 3rd, 2011 at 10:41 pm:

Honestly I’m tired to receive from my extensions’ users bug requests in the form “please update form Fx 7.x, 8.x”.

I host my extensions on AMO but also on my website, the problem occurs on my website and I can’t be reactive at every new Firefox release because I’m not a commercial developer but I would care for my users.

Please don’t tell me to use the AddonSDK because it isn’t (yet) productive like the XUL environment and converting complex existing extensions can be very hard, introducing regressions without significant benefits.

Developing for a platform must be a pleasure not a pain.

Please apologize me for the rant but I really love the Mozilla world, I totally agree with its manifesto, I believe into the motto “Make the web a safe and better place” but as volunteer developer the new release cycle is really really frustrating
6
Axel Hecht said on October 4th, 2011 at 4:06 am:

Davide, I recall someone mentioning that firebug got their problems around updates and local vs AMO served figured out. Mind poking the add-on community for a post on that?

Also, if you want to serve updates locally, Mike Kaply has a post on http://mike.kaply.com/2011/06/29/creating-a-universal-updateurl/ on how to serve them to be “compatible by default”.
7
skierpage said on October 4th, 2011 at 5:21 am:

While a vocal minority complained, Mozilla made three Firefox releases that reliably, rapidly, and regularly improved the browser. And judging from Nightly, so will the next two. I hope you’ll find that improving communication is an API-compatible documentation tweak by comparison

I think 2011.N would be a better version number, but I’m not complaining.
8
Jay said on October 4th, 2011 at 5:41 am:

I was aghast (pleasantly) when installing FF 8 beta for the first time when the new addons compatibility dialog appeared with check boxes for those addons that are compatibile and which ones are not as well as which ones I want to keep and make compatible. After making my choices the install proceeded and ALL of my previous addons worked as intended.
9
Ronald Hunter said on October 4th, 2011 at 6:00 am:

I certainly hope that no plan which includes silent updates will change the policy allowing users to require permission for any update. I REALLY hate this feature of Google Chrome.
Otherwise, I am 100% in favor of the changes. One thing, I think the update should take place when the program terminates, or when it next starts so that the annoying ‘an update is available’ doesn’t interrupt my work. And, of course, no update should occur when the program is in use.
10
Pingback from MOZILLA AIMS TO ADD SILENT UPDATING TO FIREFOX 10 » MOZILLA, FIREFOX, CHROME, WINDOWS, GREGG, STRONG » GADGETTECHNEWS.CO.CC

[...] &#959f th&#1077 Mozilla Foundation, Mitchell Baker, acknowledged wh&#1072t &#1109h&#1077 called “update fatigue.”“In th&#1077 past w&#1077 h&#1072&#957&#1077 b&#1077&#1077n extremely careful t&#959 [...]
11
Pingback from Firefox Keeping Rapid Release, But Adding Silent Update Option For Irritated Users : Test Drive

[...] is changing with their web browser before it changes,” Mozilla Chair Mitchell Baker said in a blog post yesterday. “ Our position was to err on the side of user notification. Today people are telling us [...]
12
Sailfish said on October 4th, 2011 at 12:09 pm:

The “in-the-wild” add-on detector and authorization process should help as I’ve already seen it in action using 8b, 9a2 and 10a1. Also, while many extension add-ons continue to work from one release to the next, since the underlying architecture changes in the new release do not break the extensions code, this has not been the case with theme add-ons. In every release, there have been significant changes in terms of styles, yes, but also (new stylesheets, new/changed files and added/removed images. While a previous release theme may “appear” to visually work, in fact, in a number of cases, they are missing visual changes, some of them very important. Integrating these changes into a new theme takes a significant effort on the author’s part in that they have to do a careful review of each change to insure that they’ve properly accounted for it.

So, my recommendation would be to consider another approach to how 3rd party themes (3PT) are handled in up-releases. Before a previous release AMO-hosted 3PT version number is auto-incremented, a careful review of the new theme changes added since the last release should done to determine if it is wise for a 3PT to have its max-version number bumped by the AMO team and, if not, then ACR should also not allow the theme to be used on the next release.
13
Phillip M Jones, CET said on October 4th, 2011 at 3:47 pm:

Why not do away with the need for max version in extensions.That seems to be the biggest stumbling block for majority of extension addons.

Also in my case I’ve noticed some of my Internet plugins (not extension) have been made non working on versions past 3.6.x FF (and 2.0.x on SeaMonkey) due to the changes. You changes should not affect Plugins because they are unrelated.
14
Franadora said on October 4th, 2011 at 4:26 pm:

Will the automatic updating require administrative permissions? On XP I am usually in a user account with limited permissions. From once a week to once a month, I log in to an account with Administrator permissions to perform maintenance. I noticed that the recent upgrade to FF7 (and FF7.01) only occurred when I logged in as administrator. Other than XP, I’m not sure what other OS’s are affected. It is currently possible that I could surf for almost a month with an insecure browser.

Very early version of Firefox (or Firebird or Phoenix?) included a change to the throbber if an update was available. Perhaps that should be brought back?
15
Pingback from Mozilla aims to add silent updating to Firefox 10 | National Cyber Security

[...] Mozilla did not say it was copying Chrome — it’s denied doing so with other features — but the chairman of the Mozilla Foundation, Mitchell Baker, acknowledged what she called “update fatigue.” [...]
16
Pingback from Links 4/10/2011: Parted Magic 6.7, Red Hat’s Latest Takeover, Fedora 16 Beta | Techrights

[...] Rapid Release Follow-Up My recent post on the rapid release cycle generated a lot of response, some very thoughtful and some also very frustrated. Many of the comments focus on a few key issues listed below. We’ve been working on how to address these issues; I’ll outline our progress and plans here. [...]
17
P. Smith said on October 4th, 2011 at 5:59 pm:

Seems like the development of main product (Firefox) in MoFo is quite chaotic. No one thinks of consequences at first step of any new change.
You had to prepare the “ecosystem” for the rapid release cycle BEFORE you actually switched to it.
That is a fail, and actually it is just one of many.
While Mozilla remains deaf to users’ voice it will regress more and more.
My opinion is that Mozilla now is in the crisis. The reason of that crisis is a bad management model. I don’t mean personally you, but I could mention creative lead Aza Raskin, who’s destructive decisions don’t find users’ approval.
Please, Mozilla, here my roar.
18
Pingback from Mozilla aims to add silent updating to Firefox 10 | lubuklinggau.org

[...] Mozilla did not contend it was duplicating Chrome — it’s denied doing so with other facilities — though a authority of a Mozilla Foundation, Mitchell Baker, concurred what she called “update fatigue.” [...]
19
sup said on October 4th, 2011 at 7:01 pm:

Version 7 is on Rapid Release. 6 on LTS.
After X months, 7 becomes LTS, 8 on RR.

Simple. Why not?

I get to use LTS, and don’t have to constantly monitor changes.
New features – configure (WebGL etc.).
Removed features – compensate (Stylish, if it’s working..).
Broken addons. Simply bumping max version doesn’t make me feel at ease. I would prefer to know author tested it. Specially a security related addon, like Perspectives.

I don’t want Firefox to be a chore. I don’t want to constantly attend to it. I want to use it.

I don’t want RR, and i’m not an enterprise. I just want a stable browser. Stable in my sense, to be clear and not invite an argument, not in any other.
If anyone would use RR, i’m fine with that. Give me an LTS, preferably as i described above.

A frustrated, loyal Firefox user.
20
sup said on October 4th, 2011 at 7:08 pm:

I forgot to add, omiting the changes is also not what i want. I do want to know what’s new.
The reason I and probably others hate to see them often is because of the rapid release thing. I just don’t want the changes to be often.

Maybe others really said they hate just the notifications, i can only say that’s not me.

Personally, my “irritation” over the notifications are a symptom, not a cause.
But I can only speak for myself.

Good luck Mozilla.
21
Alice said on October 4th, 2011 at 9:25 pm:

Update Fatigue? too bad.

because my addon not work , so i use Firefox 3.6.
22
Pingback from Nowinki Informatyczne » Blog Archive » Mozilla wprowadzi ciche aktualizacje w Firefoksie 10 » Nowinki Informatyczne

[...] Baker, prezes Mozilla Foundation napisała, że kiedyś jej firma starała się dokładnie informować użytkowników o zmianach w [...]
23
Pingback from Mozilla might do silent updating to Firefox 10 browser

[...] Mozilla did not say it was copying Chrome — it’s denied doing so with other features — but the chairman of the Mozilla Foundation, Mitchell Baker, acknowledged what she called “update fatigue.” [...]
24
Pingback from 1081009 | Firefox เตรียมแก้ปัญหาเรื่องความเข้ากันได้ของส่วนเสริม-วิธีการอัพเดต | ร้อยแปดพันเก้า.com

[...] ที่มา – Mozilla Future Releases, Lizard Wrangling [...]
25
Pingback from Las actualizaciones silenciosas vendrán de la mano de Firefox 10 - Taringa!

[...] | Mozilla Wiki, Mitchell Baker En Genbeta | La pregunta de la semana: ¿Cuál es el futuro del navegador Mozilla [...]
26
Pingback from Las actualizaciones silenciosas vendrán de la mano de Firefox 10 | Geeks21.com - Tecnología, Entretenimiento y Cultura

[...] | Mozilla Wiki, Mitchell Baker En Genbeta | La pregunta de la semana: ¿Cuál es el futuro del navegador Mozilla [...]
27
Pingback from Las actualizaciones silenciosas vendrán de la mano de Firefox 10 - La Isla Buscada

[...] | Mozilla Wiki, Mitchell Baker En Genbeta | La pregunta de la semana: ¿Cuál es el futuro del navegador Mozilla [...]
28
Pingback from Mozilla : des mises jour transparentes pour Firefox 10 | Allomonsite.com

[...] son blog personnel, Mitchell Baker, présidente la fondation, admet : « aujourd’hui les gens nous rise savoir – très clairement – que les [...]
29
Pingback from Firefox will zurück zu stillen Updates, 6-Wochen-Release-Zyklus ermüdet Nutzer » t3n News

[...] muss Mozilla sich etwas überlegen. Den Entwicklern ist ebenso wie der Chefin klar, dass der kurze Releasezyklus Maßnahmen erfordert. Nutzer wollen sich nicht alle sechs Wochen [...]
30
Pingback from Las actualizaciones silenciosas vendrán de la mano de Firefox 10 | EmaCorp News

[...] Vía | Mozilla Wiki, Mitchell Baker [...]
31
Chris said on October 5th, 2011 at 4:05 pm:

In response to comment 14:

Will the automatic updating require administrative permissions? On XP I am usually in a user account with limited permissions. From once a week to once a month, I log in to an account with Administrator permissions to perform maintenance. I noticed that the recent upgrade to FF7 (and FF7.01) only occurred when I logged in as administrator. Other than XP, I’m not sure what other OS’s are affected. It is currently possible that I could surf for almost a month with an insecure browser.

Very early version of Firefox (or Firebird or Phoenix?) included a change to the throbber if an update was available. Perhaps that should be brought back?

*********************

Hi Franadora,

Thanks for your comment.

The current implementation plans for silent updates will not require admin accounts (across XP/Vista/7) to update to the latest and most secure version of Firefox.

Chris Lee
Firefox Product Manager
32
Pingback from Gebruikers Firefox ‘update moe’ | | Web & Coding.nlWeb & Coding.nl

[...] updaten introduceren. Beschikbare updates worden dan automatisch gedownload en geïnstalleerd. Volgens Mozilla voorzitter Mitchell Baker zijn veel gebruikers “update [...]
33
Pingback from Mozilla coaxing old-era Firefox 3.6 users to upgrade | Christian Media Cross

[...] Monday, Mozilla Chair Mitchell Baker said in a blog post that other rapid-release course corrections are on the way, including a more silent upgrade process [...]
34
Pingback from Mozilla coaxing old-era Firefox 3.6 users to upgrade | Up2dateNews

[...] Monday, Mozilla Chair Mitchell Baker said in a blog post that other rapid-release course corrections are on the way, including a more silent upgrade process [...]
35
Pingback from Mozilla coaxing old-era Firefox 3.6 users to upgrade

[...] Monday, Mozilla Chair Mitchell Baker said in a blog post that other rapid-release course corrections are on the way, including a more silent upgrade process [...]
36
Mike said on October 6th, 2011 at 11:29 am:

In the past, news of support for silent updates would not significantly concern me for I felt I could reasonably count on:

a) Admins/users would insist on and maintain control over the update process so that computers within their network would only be updated after said updates were deemed safe, compatible, and appropriate.

b) Software developers would insist on and continue to maintain support for fine grained admin/user *control* over the updating process so that those who use their software could maintain safe computing practices.

However, it appears to me that more and more inexperienced admins carelessly allow software to pull updates into their corporate environment without so much as a sanity check of what changes are being made and when. Even worse, it appears to me that more and more inexperienced software developers feel it is appropriate for them to reduce admin/user control over the update process and the distinctly different steps of checking for an update to see what is available, downloading an update, and applying the update. Having seen some foolish and inappropriate changes to the Mozilla update mechanism(s), I’m concerned there will be more which further degrades admin/user control over their systems. It is about control, after all… not simply notification.
37
RobM said on October 7th, 2011 at 12:29 am:

I don’t suppose there’s any chance of you lot going back to *less* frequent browser updates that don’t update behind my back and which don’t install awfully unstable crap (hello there 7.0.1)?

This is a terrible decision on your part for those of us who like to maintain control of our computing environment, whether personal or professional.
38
Pingback from Mozilla idzie w ślady Google… będzie bezpieczniej, czy tylko wygodniej? « Spider's Web

[...] Mozilla obawia się, że dla części z nich może to być problem (Mitchell Baker, szefowa FM, mówi wprost, że boi się “zmęczenia [...]
39
Pingback from Mozilla coaxing old-era Firefox 3.6 users to upgrade | Partners In Sublime

[...] Monday, Mozilla Chair Mitchell Baker said in a blog post that other rapid-release course corrections are on the way, including a more silent upgrade process [...]
40
Pingback from Mozilla coaxing old-era Firefox 3.6 users to upgrade | Myfriendpal

[...] Monday, Mozilla Chair Mitchell Baker said in a blog post that other rapid-release course corrections are on the way, including a more silent upgrade process [...]
41
Pingback from Mozilla implementará actualizaciones silenciosas en Firefox

[...] información: Blog de Mitchell Baker (Mozilla Foundation) Tags: actualizaciones, Firefox, [...]
42
Gary said on October 7th, 2011 at 10:40 am:

Why not take a page from Microsoft and let the user decide if they want the automatic (silent) update or if they want to be notified and install the update at a later time?
43
Jesse Bees said on October 7th, 2011 at 11:01 am:

Re: comment 31

The current implementation plans for silent updates will not require admin accounts (across XP/Vista/7) to update to the latest and most secure version of Firefox.

Chris Lee
Firefox Product Manager

—————————-

What a shame. This is the primary reason I can’t bring myself to use Chrome. I have to suppose that Firefox will get around admin privileges the same way that Chrome does: by installing itself in a “data” directory rather than one designated by the OS for programs.
44
Pingback from Mozilla Plans to Silently Update Future Firefox Releases | wpgrabber

[...] new silent update policy, which Mitchell Baker, Chair of the Mozilla Foundation, announced via her blog, will not take effect until Firefox 10 which will arrive in early [...]
45
Pingback from Mozilla Plans to Silently Update Future Firefox Releases « News Hub Today

[...] new silent update policy, which Mitchell Baker, Chair of the Mozilla Foundation, announced via her blog, will not take effect until Firefox 10 which will arrive in early [...]
46
Rich Smith said on October 7th, 2011 at 11:21 am:

Great, now I’m going to have to blackhole mozilla.org at my router and then manually open up a route every time I’m ready to update. Why is it so hard to figure out that the people who want Chrome will use chrome, not wait for Firefox to imitate it sufficiently?
47
Rich Smith said on October 7th, 2011 at 11:24 am:

RE: COMMENT 42

> Why not take a page from Microsoft and let the user decide if they want the
> automatic (silent) update or if they want to be notified and install the update
> at a later time?

Because you are not the demographic mozilla.org is aiming for. They have made the corporate decision to abandon the able users in favor of the inexperienced masses. When the able users start switching their friends’ and family’s browsers to something else that treats them with a little more respect, perhaps mozilla.org will start copying that browser.
48
Pingback from Mozilla coaxing old-era Firefox 3.6 users to upgrade | Social Media and Tech Tips - Cuhea - Your Social Media Guide

[...] Monday, Mozilla Chair Mitchell Baker said in a blog post that other rapid-release course corrections are on the way, including a more silent upgrade process [...]
49
Pingback from Web Development articles, tutorials, help » Blog Archive » Mozilla Plans to Silently Update Future Firefox Releases

[...] new silent update policy, which Mitchell Baker, Chair of the Mozilla Foundation, announced via her blog, will not take effect until Firefox 10, which will arrive in early [...]
50
Pingback from Mozilla Plans to Silently Update Future Firefox Releases | t3knoDorKs

[...] new wordless refurbish policy, that Mitchell Baker, Chair of a Mozilla Foundation, announced around her blog, will not take outcome until Firefox 10, that will arrive in early [...]
51
Pingback from Mozilla aims to add silent updating to Firefox 10 « Browsers « { DEV }

[...] Mozilla did not say it was copying Chrome — it’s denied doing so with other features — but the chairman of the Mozilla Foundation, Mitchell Baker, acknowledged what she called “update fatigue.” [...]
52
Pingback from Mozilla Plans to Silently Update Future Firefox Releases | Premium Shopping Store

[...] new silent update policy, which Mitchell Baker, Chair of the Mozilla Foundation, announced via her blog, will not take effect until Firefox 10, which will arrive in early [...]
53
Pingback from Firefox 10: Integrazione di aggiornamenti silenziosi » FratelloGeek

[...] 2012.Certo, è bello aggiornare le versioni, cosi puoi ottenere maggiori funzioni.Ma che noia star sempre li a installare l’update!Ti piacerebbe avere un browser che scarica a installa [...]
54
Pingback from סיכום שונות לשבוע החולף | הבלוג של רומ"ח

[...] דיווח בבלוגים של Firefox, כתוצאה מהשחרור המהיר של גרסאות, דפדפן Firefox יאפשר עדכון [...]
55
Caspy7 said on October 8th, 2011 at 1:15 pm:

Brings up a point about update size.
Short.
Worth a read if you haven’t already.
http://www.webmonkey.com/2011/10/mozilla-plans-to-silently-update-future-firefox-releases/
56
Pingback from | Web Design Florida Organization

[...] new silent update policy, which Mitchell Baker, Chair of the Mozilla Foundation, announced via her blog, will not take effect until Firefox 10, which will arrive in early [...]
57
Pingback from Mozilla Plans to Silently Update Future Firefox Releases | JLD Express Shopping

[...] new silent update policy, which Mitchell Baker, Chair of the Mozilla Foundation, announced via her blog, will not take effect until Firefox 10, which will arrive in early [...]
58
Pingback from Mozilla Plans to Silently Update Future Firefox Releases | Dive For Smiles

[...] new silent update policy, which Mitchell Baker, Chair of the Mozilla Foundation, announced via her blog, will not take effect until Firefox 10, which will arrive in early [...]
59
HT said on October 10th, 2011 at 3:29 am:

Hello
Many useful sites are blocked in a country in middle-east and these days almost most V.P.Ns not work inside the country.
Please create a new project and let them to find a way to connect to the free world.
Thank You
60
gm said on October 12th, 2011 at 5:59 pm:

.
We like Firefox!
We like the latest version FF8.
We welcome your new (rapid?) quick releases.

The little issue – though:
- Looks like you are not the same rapid & quick
enough for updating relevant Add-on’s, and other
vital functions.

That’s why the customers’ frustration.
Don’t you think so?

But Hey Guys! Keep trying!
.
61
Mj said on October 12th, 2011 at 10:32 pm:

Yes! This is an infinitely good thing.

However, it seems that there is still one thing that hasn’t been addressed: the downloads window. It’s always been annoying being the way it is. Will the next release have a fix on it already? Sure there are tons of add-ons to fix it up but still…
62
Bob Trower said on October 19th, 2011 at 10:15 pm:

After biting my tongue for some time, I just entered a comment against your original post about this:

http://blog.lizardwrangler.com/2011/08/25/rapid-release-process/#comment-19995

In this latest update, you seem to have missed the point of most of the criticism. You have addressed some of the symptoms, but you have not really addressed the problems. Until you address the actual problems, you will lose people like myself and the user communities that follow us.

Freeze the rollover of major version numbers at 8 and change 9/10/11/12 … to 8.09/8.10/8.11/8.12 …. Just do it. Your versions are totally messed up and the fact that you don’t get that is an indication that something has gone horribly wrong. Give yourself and the rest of the community time to regroup, understand the nature of the problem and address it reasonably. Maintain compatibility at version 8 until you have code for a truly major ‘version 9′. Hopefully that will not be too much before 2013 and between now and then you can figure some way to re-brand so you can go back to a ‘version 1.00′ via a version 0.90 in place of 9.0.

Versions, amongst other things, signify the nature of the changes they contain. They help to organize your release schedule around Major architectural changes, Feature Releases, Bug Fixes and builds. Things that will take a long time and create big changes go into major releases. Things that make incremental improvements and add features go into minor releases. Things that must be released quickly to fix bugs or plug security holes go into ‘release level’ releases. There are a number of different conventions for version numbers, but they have some things in common that form a sort of ‘language’ that developers, build managers, testers, API partners, etc all understand. The capricious move to abandon this ‘language’ breaks all sorts of things for zero gain.

Who at Mozilla decided that software development was sufficiently deterministic that you could predict a major release would be available at a particular calendar date six weeks away? You can’t even realistically expect to do that with a bug fix release. You have showstopping memory leaks going on years now, for goodness sake. If you can’t nail down these crucial basics, how can you hope to invent a new product every six weeks? The current release regime is willfully clueless. There is no way to defend it and the fact that you continue to cling to its defense is troubling. If I understand you correctly, you have a notion that the version numbers no longer have their usual meaning and are merely indicative of a release with no significance beyond that. That is hardly a defense. Version numbers serve a purpose and that is why you see them all over the place in operating systems, libraries, commercial products, specifications, documents, etc. Ceasing to assign version numbers rationally is a mistake. You are destroying a critical element of the system. It is still a mistake, even if you did it on purpose.

The people who release most of the major software used in the world routinely shift schedules by months. You do not even understand how to use version numbers, but feel confident you can beat the rest of the world in making major releases on a tight schedule. The notion that you can make major releases of a mission critical product to millions of users every six weeks is simply ridiculous on its face.

Anyone involved in this set of decisions with less than a bare minimum of ten years hard-core hands on development experience as designer/developer/build manager, etc is suspect. If you have some, you should think about taking them out of the major decision process. It is highly doubtful they have the skill and experience necessary to make such decisions. This latest misstep on the heels of hundreds of comments pointing out your mistakes is an indication that you likely will not understand this in time to save the project. If you can’t, get out of the way for goodness sake.

Whether you continue with them or not, the version regime and the release schedule are clearly dead. However, the entire project does not have to die with them. You can go back to a rational release regime and make things right.

One of the things that just makes me grind my teeth is that you appear to be attempting to copy Google Chrome’s success *in spite of its mistakes* by copying its mistakes. Its constant stealthy updates are an unnecessary risk and makes it unsuitable for some purposes. Its versioning system is messed up. In its current incarnation, versions are not really used as normal versions, but I expect that they *will* have a watershed update that changes things at some point. They will probably call it Chrome 2. Whatever the case, empirically, Chrome rarely or never breaks on update, while FireFox invariably breaks on every update. I cannot remember the last time I even noticed Chrome was updating while it seems that FireFox involves some sort of interruption every single time I go to use it.

As I mentioned in my earlier post, FireFox is no longer my default browser. Getting me back is still relatively easy now, but it will not be once I cease using it altogether and remove it from my systems.

Although you are addressing some of the symptomatic fallout of your errors, the errors themselves persist and they will continue to do damage until you address the errors directly.

One of the indicators that you have lost the guidance of seasoned developers is that you continue to have unacceptable stability problems. I have had to create a separate Virtual Machine to house only FireFox and Thunderbird because they are so unstable they can take down the operating system. They routinely exhaust nearly a gigabyte of memory each and about half the time FireFox or Thunderbird exits either by crashing or being forced closed to restore memory. As someone who has personally developed a real-time multi-user system by hand in less than 16,000 bytes, I find the use of 1,000,000,000 bytes by a single application excessive.

Given that you still have some severe show-stopping issues to deal with, it is doubly puzzling that you are wasting the energy of both yourselves and the community addressing a non-problem and creating chaos and confusion by insisting on frequent, disruptive, compatibility breaking updates that render version numbering meaningless. Rather than messing around with rapid turn-around to satisfy your rampant ‘feature-itis’, you should turn your attention to the more important business at hand. The massive footprints of FireFox and Thunderbird are downright bizarre.

As before, I realize this is harsh and I apologize for the tone. However, this really is poised to become a tragic waste of years of hard work by thousands of individuals.

I am not sure who is in charge at Mozilla, but whoever they are, they have had plenty of time to respond to feedback and effect genuine solutions to the actual problems rather than the current damage control exercise. I do not think it unreasonable to ask that the people in charge show some humility and respect for the project and its many stakeholders by honestly ‘fessing up to their mistakes, addressing the source of the problems rather than the symptoms and correcting them before it is too late.
63
Pingback from Firefox Keeping Rapid Release, But Adding Silent Update Option For Irritated Users | Lose Your Fear

[...] is changing with their web browser before it changes,” Mozilla Chair Mitchell Baker said in a blog post yesterday. “ Our position was to err on the side of user notification. Today people are telling us [...]
64
Miroslav said on November 10th, 2011 at 2:15 am:

Rapid Release Process – that’s very nice way to do software development, yes it is! But does it has to mean releasing a major version every 1-2 months? Nooo, unless you are not familiar with contemporary software development or you do want to be kicked out of business. Actually this is what you are really trying to do, is it not? You are trying to ruin the Mozilla browser because you are a secret agent working for IE or Chrome or whatever?! This rapid major version release is bringing Firefox down and it is doing it fast!

Now I am just going to disable the automatic update and if in several months I see that I have to update ti version 10 or 11 I will uninstall this software.

Farewell Firefox!
65
Tom Sullivan said on November 12th, 2011 at 2:22 am:

Silent updates may be fine for some people, but I **strongly** recommend giving users the chance to opt out to a notification & permission to proceed method instead. There are significant security issues if Firefox or Thunderbird update in a very insecure environment such as an airport. There is also the issue if a user is on dialup – a dialup user needs all the bandwidth possible and would prefer to update overnight, etc.

As always, the more control the user has – if he wants to exercise it – the better.
66
Jeff Willener said on December 2nd, 2011 at 5:27 pm:

I support Mr. Trower; as a fellow seasoned developer and company official I have an appreciation for his points. I have been a supporter of FireFox for a very long time. Please know, I want to be part of the solution and not a contributor to the problem. My hopes are to inspire reflection and direction.

With that said, what drew myself and perhaps others to this product was, in short, the reliability, stability and the community.

-Reliability, the product was predictable. As a user we could depend upon the various working parts integrating nicely as a whole. Furthermore, FF seemed to be the one browser which would render pages to the W3C standard most accurately. Bulk management was trivial with near transparent impact.

-Stability, the product very rarely exhibited operational issues. System footprint was small and crashes or errors were non-existent. The application was responsive and fluid. Most importantly, add-on developers had time to perfect their product; they too complimented this point.

-Community, there were a large number of people who believed in this product. This community was built upon users who wanted to use FF and developers who “volunteered” their time and money, as individuals or institutions, to reach into this user base. Excluding those who wanted to profit from your success there were those who just wanted to be a part of something better. And being “better” is not a coincidence, it’s a choice. A terrific explanation, one which many have seen but never hurts to be reminded, is a small clip by Simon Sinek (http://www.youtube.com/watch?v=qp0HIF3SfI4).

My first challenge to Mozilla is to rediscover why your products were created in the first place. And this is beyond the fact to profit from but rather deeper to why competitors were so poor and why yours was so much better. This is what people followed.

My second challenge to Mozilla is to identify release cycle implications. My goal here isn’t to challenge your version control implementations but rather how and why you are using them. This is a tough one; one that cannot be answered on a cocktail napkin. This is a two step challenge.

First, as you already know, you’re audience (ecosystem) is not only those which click buttons and view pages but also those who develop against it. Of course the scale tips heavier to one side than the other but there is a dependency between the two for maintaining their relationship. Part of your job is to ensure the marriage between them never fail all amongst staying ahead of competition for feature and functionality. Remember, part of your product is the unacknowledged 3rd party development of others. Think iPhone!! 10 reasons to use the iPhone turned into 10k over night and Job’s didn’t have to spend a dime. Without that it’s just a phone.

The second step; one which has been mention frequently is the maintenance and implementation practices of your products by the user. This very much relates to step one so we cannot completely ignore it. To start, let me thank you for what has been done thus far. However study your users in regards to deployment cost, deployment time and deployment schedule. I’m sure you already do and have, but do more. You have a very popular product and you must consider man hours and money related to each release bestowed upon the user. There is a cost which is directly related for each release even if it’s one user. Can you imagine that each release might produce a global cost of $10k, $100k or perhaps $1M. Of which does not have a justifiable or even tangible ROI; so we are now back to “why” the community has chosen your products. Not to mention money lost by incompatibilities. This point completely excludes versioning and testing of add-on’s and extension’s which too has a direct monetary impact for 3rd party development. Most development or IT organizations regardless of industry or practice when they see an increment in the major revision number it then implies (again money and man hours) for compatibility testing and integration testing. The last thought of course would be deployment schedule for organizations. Prior to the 6-week strategy, organizations could wait for a major revisions to stabilize before implementation. Unfortunately this is not the case now.

The last challenge to Mozilla is probably a private challenge but the most important one of all. I’d like for you to study your (Mozilla) public image. Perception is global; reality is local. This means the most logical and best ideas are only as good as their implementation. By that no user will ever understand your “perfect” idea but rather only react to or judge its implementation. With no doubt you, your organization and your products are judged every day. Judgement, is typically short-term, which comes and goes whether its good or bad. No one can hide from it, either as individuals or institutions. However it’s what we do with judgement that matters most and as a result holds a long term effect. So I ask you to consider how do users perceive Mozilla with regular major revisions? Is it irresponsible, unorganized, under educated, careless or is it responsible, organized, educated and methodical. I ask because the answer to this, again, holds a long term impact, good or bad. There was a reason why Mozilla dominated the market at version 3.x while everyone else eased up the release ladder. This year alone competitor versions were closed in upon. In 2014, will IE be lagging behind at v11 whilst FF at v24.0.0 and add-on/extension developers behind racing to update their code just to v18.x.x? Holding steady with minimal versioning implied stability regardless if it was or was not.

These challenges might seem small, irrelevant and caddy. But sometimes we as people, or organizations, need to remember who we are, where we’ve been and what we know. A little “deep” reflection never hurt anyone.

My final request is to please provide access to all major version numbers for download. [Hint - if not already, might want to monitor downloads, updates and deployments. Numbers speak loudly.] With this strategy, the user base must have a mechanism to roll back.

Sincerely, thank you for opening discussion on your blog. I was a believer in Mozilla and what it stood for. I will support your organization as long as we share what we believe.

Only my best…
Jeff
67
Pingback from IE to Start Automatic Upgrades across Windows XP, Windows Vista, and Windows 7 - Windows Community Forum

[...] [...]
68
Pingback from About Damn Time: Microsoft Will Silently Upgrade Everyone To Latest Version Of IE | TechCrunch

[...] updates to the browser is now par for the course in modern browsers like Firefox and Chrome. Although Firefox’s move to silent update was a more recent addition, Google has [...]
69
Pingback from About Damn Time: Microsoft Will Silently Upgrade Everyone To Latest Version Of IE | Diseño web económico

[...] updates to the browser is now par for the course in modern browsers like Firefox and Chrome, as Microsoft points out. Although Firefox’s move to silent update was a more [...]
70
Pingback from About Damn Time: Microsoft Will Silently Upgrade Everyone To Latest Version Of IE | Hastings News | Hastings Local News

[...] updates to the browser is now par for the course in modern browsers like Firefox and Chrome, as Microsoft points out. Although Firefox’s move to silently update its browser [...]
71
Pingback from About Damn Time: Microsoft Will Silently Upgrade Everyone To Latest Version Of IE | Shorewood News

[...] updates to the browser is now par for the course in modern browsers like Firefox and Chrome, as Microsoft points out. Although Firefox’s move to silently update its browser [...]
72
Pingback from About Damn Time: Microsoft Will Silently Upgrade Everyone To Latest Version Of IE | Bitmag

[...] updates to the browser is now par for the course in modern browsers like Firefox and Chrome, as Microsoft points out. Although Firefox’s move to silently update its browser [...]
73
Pingback from About Damn Time: Microsoft Will Silently Upgrade Everyone To Latest Version Of IE | Startup Help

[...] updates to the browser is now par for the course in modern browsers like Firefox and Chrome, as Microsoft points out. Although Firefox’s move to silently update its browser [...]
74
Pingback from About Damn Time: Microsoft Will Silently Upgrade Everyone To Latest Version Of IE | Krantenkoppen Tech

[...] updates to the browser is now par for the course in modern browsers like Firefox and Chrome, as Microsoft points out. Although Firefox’s move to silently update its browser was a [...]
75
Tim said on December 15th, 2011 at 2:54 pm:

Sorry but I’m afraid this is just too little too late. IE is becoming better and chrome is great, so I’ve had to drop support for firefox from my application.
76
Pingback from About Damn Time: Microsoft Will Silently Upgrade Everyone To Latest Version Of IE « dejavuu

[...] updates to the browser is now par for the course in modern browsers like Firefox andChrome, as Microsoft points out. Although Firefox’s move to silently update its browser [...]
77
Pingback from About Damn Time: Microsoft Will Silently Upgrade Everyone To Latest Version Of IE | TechXpress

[...] updates to the browser is now par for the course in modern browsers like Firefox and Chrome, as Microsoft points out. Although Firefox’s move to silently update its browser [...]
78
Pingback from 【脆弱解決か】Microsoft、来年からIE自動アップデートと発表!　これでIE6問題解決か!? | みるくビズ

[...] Microsoft自身も認めるとおり、自動アップデートはFirefoxやChromeなど現代の多くのブラウザで標準になっている。もっともFirefoxのアップデート自動化は比較的最近のこと。自動アップデートの先鞭をつけたのはGoogleで、Chromeブラウザの一つのセールスポイントになった。言うまでもないが、ブラウザを最新版にアップデートすることはセキュリティーを飛躍的に高める効果がある。Microsoftは親切にもこの点を実証する独自の調査へのリンクを提供している。それ（Microsoft Security Intelligence Report vol 11）によれば、2011年上半期に発生したセキュリティー侵害事件のうち、ゼロデイ攻撃（セキュリティ上の脆弱性が発見された際、問題の存在が公表され、ソフトウェアのベンダー等によって対策が取られる前に行われる攻撃）によるものは1%以下だったという。 [...]
79
Pingback from Microsoft to start Automatic Updates for IE in Windows XP, Vista and 7 | Gobble D Geek

[...] customers can receive IE9 and future versions of Internet Explorer seamlessly without any “update fatigue” [...]
80
Pingback from Microsoft to automatically upgrade all Windows users to latest version of Internet explorer

[...] customers can receive IE9 and future versions of Internet Explorer seamlessly without any “update fatigue” [...]
81
Pingback from IE to Start Automatic Upgrades across Windows XP, Windows Vista, and Windows 7OKGrafix | OKGrafix

[...] can receive IE9 and future versions of Internet Explorer seamlessly without any “update fatigue” [...]
82
Pingback from IE to Start Automatic Upgrades across Windows XP, Windows Vista, and Windows 7 : News Sluice

[...] customers can receive IE9 and future versions of Internet Explorer seamlessly without any “update fatigue” [...]
83
Chris said on December 17th, 2011 at 1:39 pm:

You lost me on 3 – Update Fatigue.

I personally dont know anyone who wants away with release notes and been given a prompt to update. Silent updates would be a disaster for me.

why didnt you post a link to where you claim people are asking for silent updates?

Silent updates is fine for things like virus defenitions, its not fine for anything that changes how software works, needs a restart of software, and may cause things to break.
84
Pingback from About damn Time!! Upgrading IE

[...] [...]
85
Pingback from Kverker IE6

[...] Ryan Gavin viser i den sammenheng til et innlegg skrevet av styrelederen av Mozilla Foundation, Mitchelle Baker, som beskriver dette som [...]
86
Pingback from Las actualizaciones silenciosas vendrán de la mano de Firefox 10 | TecnoApps.net

[...] | Mozilla Wiki, Mitchell BakerEn Tecno Apps | La pregunta de la semana: ¿Cuál es el futuro del navegador Mozilla Firefox? [...]
87
Pharmd601 said on April 6th, 2012 at 4:56 pm:

Hello! bgfbdff interesting bgfbdff site! I’m really like it! Very, very bgfbdff good!
88
James said on June 6th, 2012 at 7:18 am:

i use ff for about two years. the 3.6 is stable and supports all my addons as adblock, noscript, better privacy… with every new update the addons melting away. I dont use ff because of ff I use it because of the addons! loosing addons with every update is annoying!
AND who got the highest release number? me! my penis is v20.0 …
89
viagra said on June 26th, 2012 at 4:04 am:

Hello!
, , , ,
90
Pharmc138 said on October 29th, 2012 at 5:57 pm:

Hello! edckbbd interesting edckbbd site! I’m really like it! Very, very edckbbd good!
91
Jenni said on December 24th, 2012 at 6:15 pm:

These are basically brief phrase loans that do deliver you meet your fiscal challenges inside a preferred specific time
period. Thinking about time limitations these loans are particularly made
above an obligation cost-free platform. As such, these
are kept fully cost-free from credential checksums.

Troubles such as defaults, arrears, bankruptcy, CCJs and even
IVAs are not regarded as here. Further, there are also no collaterals
associated with these loans. There is minimal paper operate necessary on the part of borrower.

There are also no hidden or additional documentation or
faxing essential here. Applying for these loans is also exceptionally
hassle-free. Many people just need filling an via the internet
form and as soon as this gets authorized money is received inside 24 hours time frame.

These loans are in general offered under convenient terms and conditions.
The general basic applicant criteria here is that
they must be a UK resident and of 18 years of age.