Slashdot Mirror
Mozilla Announces Enterprise User Working Group
Lennie sends this quote from an announcement at the Mozilla blog:
"Recently there has been a lot of discussion about enterprises and rapid releases. Online life is evolving faster than ever and it's imperative that Mozilla deliver improvements to the Web and to Firefox more quickly to reflect this. This has created challenges for IT departments that have to deliver lots of mission-critical applications through Firefox. Mozilla is fundamentally about people and we care about our users wherever they are. To this end, we are re-establishing a Mozilla Enterprise User Working Group as a place for enterprise developers, IT staff and Firefox developers to discuss the challenges, ideas and best practices for deploying Firefox in the enterprise."
Enterprise has never been (and I'll argue, shouldn't be) a focus of ours
is Asa Dotzler part of this workgroup?
In true Mozilla fashion, I'm sure that will mean "We'll pretend to listen while we continue to do whatever we want"
Your hair look like poop, Bob! - Wanker.
Mozilla lost this race long ago. Chrome has the tools I need (MSI packages, GPO templates, etc.) to deploy it properly throughout my organization. Mozilla can go back to making Super-Duper(tm) bookmarks or whatever.
Online life is evolving faster than ever
No, it's not evolving faster than ever. Everything works with IE7. All innovations beyond IE7 are just sugarcoating, most of them invisible on the deployed web. The slow players still decide which features are widely available. The other players are falling over their own feet trying to outrun each other and the users are getting annoyed by an ever changing environment that doesn't let them do their work, for no benefit at all. The browser is a tool, you tools!
Didn't they just say, in so many words, that Business wasn't their focus? Is this doublespeak, or have they forgotten already?
...tune in next week for the continuing drama.
1 (short ton / firkin) = 89.1432354 slugs / keg
Don't release a buggy browser with new features only half implemented, and/or poorly tested just for the sake of a bigger version number.
My advice to everyone? Don't use firefox anymore.
Our it department would be more than happy to roll out Firefox but the lack of msi and group policy support is just a plain no go for them in our field (banking it).
MSI packages and Group Policy templates are available here: http://www.frontmotion.com/FMFirefoxCE/download_fmfirefoxce.htm
This was the story in 2007 when they first tried this: New Mozilla working group aims to simplify enterprise Firefox deployment
1 (short ton / firkin) = 89.1432354 slugs / keg
I'll follow and contribute as much as I can, hoping that something changes, but having the cold expectation that nothing will. On the windows side, FF essentially needs three things:
1. MSI for deployment.
2. GPO management.
3. Mozilla branding and support for the above, so I can automatically update the browser.
That's the peanut butter and jelly for enterprise. I can get the first two from other people, why not you guys? Why it has taken this long to get to this point is beyond me. Seriously, the 'battles' between chrome, opera, and firefox are like watching soccer moms fight to the death over the last tickle me elmo at a Walmart when there's a toy store next door with aisles full of the same toy, cheaper. Seriously, do you guys want to keep scratching with each other over grandma's machine, or do you guys want people like me to push your product to 50 machines at once, and let 50 people *see and use* your browser, learn for themselves that it's better, and take it home with them?
There are some people that if they don't know, you can't tell 'em.
Does this finally mean that there will eventually be complete Active Directory integration or something similar of a sort? Having a centralized way to manage Firefox clients would be brilliant.
o stop supporting enterprise deployments (by rapid release, no bug fixes only)
o start an enterprise working group
o profit! (charge for support)
wtf is a faget anyway?
This hideously misspelled derivation of 'faggot' is primarily used by uneducated rednecks who fail to see the irony of calling someone a derogatory name but having no idea how to say the word.
1) Throw the MCSEs a bone: give them their MSIs and GPOs. Alternatively, bless FrontMotion's MSI and GPO projects as the "official" ways to get these things for businesses that need them.
2) From time to time (but no more frequently than once every two years), tag a release as Long-Term Support. This is exactly what it says on the tin: this release gets official support from Mozilla, including security fixes, until the next Long-Term Support release.
3) Support for a non-LTS release is not dropped until there have been at least two major releases since then. Under the current situation, that means FF5 support would not be dropped until the release of FF7, which in turn would not be dropped until the release of FF9.
I realize that long-term or even mid-term support is not sexy. Techies always want to live on the bleeding edge. But not every person or business is willing, or even able, to do that. They also need to be taken care of.
Please discuss the security additions I hope the security interest additions
user based add ons / extensions / plugins under their own profile directory is the #1 problem for deployments.
It is almost as bad as giving every user admin privileges on their machine.
Too bad we dumped your sorry asses a few weeks ago. -An enterprise user.
I'll believe in corporations having personhood when Texas executes one... - advocate_one
"have never had an add-on broken except when transitioning between major releases."
Thats fine and all when 'major releases' happen once a year or so. Nowdays its about once a fortnight or so.
Can you imagine Scotty and Spock having to upgrade the ships software every other episode (or Geordi and Data if you prefer TNG)
For myself I will probably continue using FF5.x for a while while I try out the competition. I certainly won't be upgrading Thunderbird .
"Enterprise". Hate that word! It feels like one of those [many] vague and esoteric business speak words that managerial-types throw around meeting rooms to try and look clever.
I thought I'd quickly check Wikipedia before posting this in-case there is some precise and useful meaning of the term and straight away I notice there is a section on the page titled "Definitions" not "Definition". The first sentence of that section is: "While there is no single, widely accepted list of enterprise software characteristics, this section is intended to summarize definitions from multiple sources"
So nice to be doing freelance programming now and not having to receive instructions from managers. I use the word "instructions" loosely.
I just don't get it. When Firefox was known as mozilla 0.3, they were doing this sort of crap. For several years now though, they've _mostly_ got the development and release models right. Now over the last year, they've totally gone off the rails.
FF4 is buggy, clunky, and has new UI elements that apparently came from the mind of a blind and retarded monkey--and in some cases, no way to turn them off! FF5 is...mostly identical. Same UI, and still buggy.
But hey--FF4 isn't supported anymore, and presumably FF5 won't be in another ten weeks or so when FF6 comes out. GREAT!
Here's a clue for this newly (re)formed enterprise group which, if followed, will be invaluable for the regular users as well: Quick trying to reinvent the fucking wheel, and spend more time rounding off the edges. Fix the bugs. Streamline the existing code. Quit adding random new features, quit throwing away versions that are a quarter-year old, quit moving stuff around for the sake of it. In short, quit making every browser update a training exercise.
"People who do stupid things with hazardous materials often die." -- Jim Davidson on alt.folklore.urban
Because people are exposed to Firefox at work and are thus encouraged to use it at home, for instance with Firefox Sync.
Whilst an msi would be better than an exe installer, at least Windows users get the latter. Mozilla has never provided a natively-packaged (.deb or .rpm) Firefox of any sort for Linux, never mind any official 64-bit builds. Yes, distros can and do provide both, but they aren't always updated timely, particularly if you're on an LTS release.
Mozilla should, if they're trying to catch any Linux admins, provide repos for the most common Linux distros (Ubuntu and Fedora to start, maybe others later) - it can't be too hard to automate the process? And to solve the "enterprise" angle, simply have an LTS repo that's updated infrequently and only for security/critical bug fixes. At the moment, I actually have a straightforward shell script to create RPMs of Firefox, Thunderbird and Seamonkey from the downloadable .tar.bz2's because apparently it's too tricky for the Mozilla dev to write such a script :-(
Did I not hear a projected release date for Windows 8 as the end of this year? Hmmm, I'm pretty sure NO ONE WANTS A NEW WINDOWS ENVIRONMENT.
Please excuse the caps, but putting out a new Windows environment when the vast majority of corporate environments - both large and small - are probably still fragmented between XP SP2, some Vista, and a good bit of Windows 7 is just plain fucking stupid. How many have upgraded all of their internal systems to work with 7? I've been in IT in various guises for nearly 20 years and I'm getting tired of shorter and shorter releases between new versions. It is getting to the point that even the "sheeple" are starting to notice and say "What the hell do I need a new version of Windows for? My computer works just fine." Plus, small businesses and even corporations are getting tired of shelling out cash for new releases in short time frames and many of these companies - especially small business - CANNOT afford the upgrade cycles.
I am not sure but I predict that this will be Windows XP vs Vista again, with Windows 8 being the major loser. Windows 7 works just fine, thank you very much.
Dream as if you'll live forever.
Live as if you'll die tomorrow.
~Anonymous~
Whilst an msi would be better than an exe installer, at least Windows users get the latter. Mozilla has never provided a natively-packaged (.deb or .rpm) Firefox of any sort for Linux, never mind any official 64-bit builds. Yes, distros can and do provide both, but they aren't always updated timely, particularly if you're on an LTS release.
This thinking is part of the problem from the 'Enterprise' point of view. In environments with highly defined workflows and resources - even HR or Accounting type depts. - standardizing on a web technology means investing in tweaking and massaging across all resources/processes vital to that environment.
Given that network segregation/gateway security/code signing technology goes hand in hand with 'Enterprise,' it is not surprising that many would be willing to slow progress in the name of stability.
The ideal Enterprise web platform would be delivered in source complemented by build script repositories, with only security updates merged into main for at least two years. Anything faster than that will simply contribute to network convulsions designed to keep stable technology separated from the modern Net at large... Making web tech the perpetual stinky goat in the room.
If you are already managing package distribution centrally, which is easy even for home users with a few readily available scripts, setting up a cross-compilation environment to generate the bins from one machine is all you need to add.
I haven't checked out the LFS cross-comp book, but that might be a good place to start. You can really end up with a cherry set-up provided you don't have to upgrade constantly or physically compile on every target architecture.
it is not surprising that many would be willing to slow progress in the name of stability.
Indeed. What many in the web development sub-industry don't seem to grasp is that progress that breaks existing stuff isn't progress, it's just random unmotivated thrashing around, aka, destruction. Progress means going forward, and that means adding features - not breaking existing ones.
In the software industry, we've somehow internalised a false idea, which is that all new development necessarily means changing the way we used to do things. But that's not actually true. If we did things right in the first place, and used extensible protocols, we shouldn't need to break anything; just add new stuff.
The unexamined implication of the "old is bad, all progress requires destruction" meme is that everything you are currently doing, you are doing wrong - because today's "new hotness" will always be tomorrow's "old and broken".
But it should be possible, at least in theory, for us to do things right the first time - or at least to know when we're doing them better than worse - and then stop changing it once we've got it right.
Conversely, if whenever we invent a technology, we have no way of telling if we're doing it right rather than wrong -- then sheesh, we shouldn't even be in the technology business, because we obviously don't know what we're doing, and we're going to just hurt ourselves.
The Latin alphabet, for instance, is around 2000 years old, and we're still using it, give and take a few tweaks. Is it bad because it's old? No. So why should a technology get outdated just because it's five or ten years old?
tl;dr: Quit breaking stuff, just get it done right, then stick with it. It's not broken because it's old, it's proven and trustworthy.
You are not a brain: http://books.google.com/books?id=2oV61CeDx-YC
How about fixing this bug from 2004?
Bug 235853 - [PAC] Defer proxy resolution for HTTP and HTTPS PAC to avoid blocking main thread during DNS resolution
We're on the edge of deciding to keep Firefox in our corporate setting, based on the same old chestnuts from years ago. Before people say "Pro MS" - just give me an installation that respects if a user has permissions or not (yes, FF will install to userdata\ if it can't write to program files); and a way to lock down or centrally manage add-ons. Without this, we run licensing risks (yes, add-ons can include the "free for home but paid for corporate"), data validity (yes, add-ons and version changes have caused interesting results within standards-tested applications (browser changed to a sandboxed approach and was a minor point release apparently - snuck past testing and caused problems). If users or PC's retrieve these updates in an unmanaged way, it's Ops who will get it in the neck because it wasn't tested to the business standards before it was released to the business.
I would prefer an MSI and GPO integration option (we actually use ZENWorks for deployment, but MSI/MST is always nice and standard), but truly, having the application respect some registry settings would be a nice start. Without it, it's too much of a risk ultimately to the business units that want to try and use it. Security pushing constant patching, and not enough time/money to keep up with the constant rounds of updates/testing = decision time.
Your environment may be different, you may have patching auto-enabled from the vendors directly with no intervention. You may never test any updates coming down, and only respond to faults if they arise. It's always an option. But some of us other ICT folk are not so lucky, and are accountable to business (and very public inquiry) if it all goes wrong. Try looking after 5000+ workstations and you'll see what I mean.
Murphy always catches up...