Submitted by In a blog post today, Adobe’s Director of Open Source and Standards said: “we will be focusing on supporting partner implementations and will no longer be releasing our own versions of Adobe AIR and the AIR SDK for desktop Linux”. McAllister says that “way back in 1999” he’d predicted “a significant market for desktop Linux by 2005. Obviously I was wrong. So we, Adobe, also need to shift with the market.” Source code for AIR will be made available to partners so they can make their own Linux implementations if they so desire. Is there anyone in the audience who cares about no more AIR on Linux from Adobe? Anyone…?
The same can be said about desktop linux in general.
That hurts. Just about every dev at every software company I’ve been with uses linux as a desktop. I’m not saying it’s better or worse, I’m just saying a lot of people who know a lot about software use linux as their primary desktop.
As a foolish choice (because it was easy) we chose to make our company UI in air for linux. It works well and is remarkably stable after hundreds of hours fixing memory leaks but I think we are one of the only companies that produced a linux only air UI. In retrospect it was a dumb choice and we should have gone with something based on html5 (like webian or QT+webkit)
Edited 2011-06-16 23:38 UTC
The thing is that Linux is a “dev” choice when it comes to workstations and we are in the minority of computer users.
In terms of managing risk of a software project maybe you should have monitored the situation more carefully and transitioned or ported a prototype as time goes on … at least Adobe hasn’t completely left you out in the cold.
Personally I have always built stuff as a web app so as long as my customer has a browser they can use it … but not every application can be a web application.
I hope your company manages to transition to a new technology.
Edited 2011-06-16 23:46 UTC
We chose air because it was easy to build a nice browser to view other content
We are looking at adding some dev time to the mozilla chromeless project so it will be ready to use come time to drop AIR completely.
Using air wasn’t all bad because a lot of my code is just performing JS on loaded documents in an instance of mx:HTML (webkit.) It can be easily ported back to pure JS.
I like GOG but I hate the downloader (privs aside).
*nod* I ended up just tossing everything into DownThemAll! instead.
I actually contacted them a couple of months ago and they ARE planning to replace their AIR-based one with something else.
This Air download tool has already been replaced.
Oh cool. Well I take my statements back then
Now that they’ve dropped Air and Reader X (according to the linked article), wonder how much longer before they drop Flash on desktop Linux as well, instead of merely treating it like a second-class citizen?
If they do drop Flash, can’t say I blame them too much. Linux on the desktop/PC hasn’t really taken off like some pundits thought it would, and the uptick in web apps/smartphone/tablet usage probably isn’t going to help either.
As other commenters have said, this drop might well forecast them dropping Air altogether. The Linux spin is just them trying to save face at the expense of Linux (which is dirty, and people will remember it). Google has phased out Gears as well. All similar, distinct platforms with client-side storage are being phased out in favor of Web Storage, part of HTML5.
I wish they’d have just come out and say it. I don’t understand why their PR department thought it would be a good idea to take pot shots at Linux geeks instead. How’s that a smart thing?
I don’t think they’d dare.
First of all, it would send a very, very bad message: “Flash is not cross platform. We’re incapable of making it work properly on anything else but Windows, so we’re just giving up.”
Secondly, how do you kill Flash on desktop Linux but expect people to believe you’re maintaining it on Android? It doesn’t make a lot of sense. It can either be made to work properly on both, or on neither. And I don’t think Adobe wants to give up Android after being thrown out of iOS. It would pretty much relegate them to Windows 7 [Phone] only. And Microsoft is a quirky company to have to depend on for your last stand… like swimming with a shark while bleeding.
Speaking of which, everybody’s on Balmer’s case for not doing “smart” things with the pile of cash Microsoft’s sitting on. Well here’s an idea, how about buying Adobe… If they’re going to be a Windows-only software house anyway, they would fit right in.
Edited 2011-06-17 07:50 UTC
I really don’t care if they do. The flash spec is open source now after all. If they drop flash on Linux, it will just a be a fire under the butts of the people who are working on the FOSS flash implementations that they better switch to high gear.
It still works on OS X. And you can bet that Adobe has no intention of dropping Flash on OS X.
I don’t know about that.
From what I hear the OS X version has issues similar to the Linux version. So they might get the boot from Jobs the same way they got booted from iOS. For example, do you think Flash will be allowed in the AppStore for OS X?
Secondly, Macs as desktop market share is not that much bigger than Linux. So similar reasons to the Linux drop may apply.
But we’re talking about something completely hypotethical. I’m not ready to believe Adobe would pull Flash from Linux. Not unless they’re downsizing and cutting all their non-Windows investments.
Given the amount of online content that requires Flash, I don’t think even Jobs would be bold enough to boot Flash off of Mac desktops. To do so would be fatal blow to the Mac desktop. There’s just way too much online content that requires Flash.
Sure it does. At this point, Android’s userbase is probably several orders of magnitude larger than that of “Desktop Linux.”
Not really. Adobe could choose to support/release Flash only for Android on certain ARM architectures.
Now there’s an idea…
Linux still has more users than AIR though.
Talk about damning with faint praise… Shit, what technology doesn’t have more users than AIR?
Blah. Some people pretend the desktop in general is dead, taken over by tablets and smartphones — but guess what? For productivity, the normal desktop Linux is far superior to and more complete than any of the tablets and smartphones on the market.
Sure, the iPad already has taken over Linux’s market share. So what? It still doesn’t come close as a productivity tool, and never will.
Why can’t something that is not only superior, but also free get any serious market share?
Because it’s not superior in the areas which matter for large-scale success in the desktop market, maybe.
You mean shoved down everyone throat whether they want it or not? You got me there.
Well, case in point : I might start envisioning putting Linux on my relatives’ computers once main desktop distros manufacturers stop breaking things as soon as they start to work.
Sure, I have fun with new features too. But that’s not the way to go when dealing with unskilled users.
As examples, one may think of…
-Distros brutally switching to the alpha and totally broken ath5k as soon as it’s around because they like the license better than madwifi’s.
-After years of careful dmix polish until linux audio bugs are finally a thing from the past, brutally declaring that pulseaudio is the future and putting it everywhere without making sure it work first.
-Same scenario with legacy bootsplash methods and plymouth. The latter still doesn’t work on a lot of hardware, even resulting in graphics bugs on boot in some implementations, but well it’s shiny isn’t it ?
-Almost everything which involves GPU-accelerated graphics. Oh, boy…
Christ, why don’t you just switch from Ubuntu instead of whining about all that?
/begin rant
Happened to me on Ubuntu and its derivatives, Fedora, Mandriva and SuSE. Most funny award goes to Mandriva, which included a messed up pulseaudio installation as well as the others but offered the option to easily disable the whole thing via GUI. How kind of them, but if they know it’s broken why do they keep shipping it instead of waiting before it’s ready ?
There’s a reason why I always install new distros on a separate HDD for testing first. I want to keep a working system at hand until I’ve fixed the next one. There’s ALWAYS something wrong somewhere. Sometimes I know how to fix it, sometimes not.
Every single major distro but Debian stable&testing or CentOS randomly breaks things from time to time because it’s cool to do so. And I have my issues with the latter too, as far as usability by unskilled users and packet freshness is concerned.
Only few obscure distros actually seem to care about the stability and usability of what they’re shipping at the same time. And when you go the way of obscure distros, you get obscure repositories too, which means that as soon as you want to do anything a bit fancy, you have to compile things by hand…
Oh, and I could have added to my list exceedingly early adoption of KDE 4 and GNOME 3 without leaving those time to mature first, as well.
I love the linux desktop for my personal use. But it’s just not for everyone yet.
/end rant
Edited 2011-06-17 12:08 UTC
-separate HDD +separate HDD partition, of course. Whole HDDs aren’t so convenient to deal with…
Edited 2011-06-17 12:30 UTC
I was under the impression that the “ath” drivers came from Atheros itself, whereas madwifi was a reverse engineered driver. IMHO, I’d rather have a beta from the manu. than a driver that may be missing key things (not saying madwifi did). Also, the manu. releasing a driver made me pick out only Atheros chipsets (this was before the other two big ones put out any good drivers)
Oh, I completely agree that pulseaudio was not, and possibly is still not, ready for prime time. It is a cool idea, but frankly I think it is too much for what it is used for in 99% of cases. sure edge cases need the advanced kitchen sink features, but most don’t.
These two go together for me. But the second one was a kind of a jab. You are really going to blame the devs who (like madwifi above) are having to reverse engineer one of the most complicated pieces in a modern computer? And (like madwifi above) I still will pick the manu.’s driver over a reverse engineered one.
Now, the biggest problem with the pretty boots comes down to nvidia, ati, intel, and kernel devs not setting a standard system for video. While the kernel devs have added kvm, and intel and ati (I think) support it, nvidia refuses. This is also a problem with GPU acceleration, intel and Ati and nvidia all use completely different subsystems for it. I would hate to be a developer having to write GPU accelerated anything. You would have to rewrite your code four times just make sure you got everything.
The same reasons why tablets are taking off is the same reason Linux(read: ubuntu) should be strong. It does 90% of what everyone uses a computer for. the other 10%, for linux, comes down to support from Microsoft, and we all know that will be cold day in hell.
That’s a choice
Myself, the first thing I ask from my wi-fi driver is to work. By the time SuSE and Ubuntu started shipping it, ath5k didn’t, while madwifi did. So I’d have preferred it if they just sticked with the working driver until the new driver was ready.
SuSE’s situation I can understand better, though, because they couldn’t ship madwifi with the distro to begin with, so a broken driver which partially works on some hardware was still better than nothing. The problem was that ath5k’s presence actually complicated the installation of madwifi if you had the package at hand, for some weird reason which I don’t remember anymore.
No, but I’m totally going to blame them for supplying a default setup which relies on technology which is not reliable yet. If the HW accelerated drivers are not ready yet, they should stick with VESA imo. Working everywhere is more important than being fancy somewhere.
I have no big problem with politics preventing quick emergence of good new video drivers standards on Linux. This will probably get sorted out one day, and at that point it’ll still be time to get the shiny standard out of the door.
What I’m against is distributions relying as a default on something that’s not ready yet. If NVidia is a dumbass, so be it, but NVidia users shouldn’t be penalized for it. Just provide them with a nice VESA fallback. Instead, what we get is something which either flat out doesn’t work with some hardware which used to work in the past, or offers highly penalizing second-class citizen treatment. I cannot agree with that.
Heh
Again, linux software could be ready for the masses, in my opinion, if it weren’t for the “breakages or outdated” choice ruining the experience.
Edited 2011-06-17 12:25 UTC
It sounds like what you want is BSD, not Linux.
Perhaps
Though it’d probably also come with its own, new range of problems.
You have that backwards. The madwifi drivers were written by Atheros and first released to FreeBSD (one of the source committers for FreeBSD worked for Atheros at the time). The driver was open-source but included a binary HAL. The madwifi drivers were then picked up by Linux devs.
OpenBSD devs didn’t like the binary HAL, so they set out to reverse-engineer the driver and came up with the completely open-source ath5k driver.
Linux devs picked up the ath5k driver, removed all mentions of the BSD license, swapped the copyrights, and included it in the kernel tree, creating a huge scandal on the Internet. Once they were caught, they reverted the license and copyright changes.
Not sure who created the ath9k driver, but it eventually supplanted both ath5k and madwifi in all the OSS OSes.
madwifi worked beautifully, and is still the only Atheros driver that supports SuperG (108 Mbps). Ath5k in comparison missed functionality and stability for a long time. Ath9k is better, but is still missing all the SuperG features.
Edited 2011-06-20 14:46 UTC
One could do an article with all the happenings, situations and events that would lead to The Year of the Linux Desktop.
Think hard and you will come up with a pretty long list when something did or did not happen causing a united cheer amongst Linux users claiming victory.
For example the long wait for Longhorn and the eventual Vista mess. Two (related) events that Linux faithful claimed would cause people and businesses to switch to Linux. Or Novell taking over SuSE, finally Linux would have global support and become a serious player. Even Bill Gates announcing he would like to get more involved in the technical bit caused some to claim this would be good for Linux growth. Cities switching to Linux (it has started) and switching back (oh well). Ubuntu giving away Linux cds, now it was too easy to get Linux and install it.
And related to Linux in a way, what about OpenOffice vs Microsoft Office? How many times did we hear “How can you compete with something that’s free?”.
Don’t even have to think that hard actually to remember several of the events that were touted as “Year of the Linux desktop” events:
1997 – First beta of KDE is released. Linux enthusiasts were claiming victory and feature parity with Windows. KDE was going to give Linux mass appeal on the desktop. KDE’s Web Site at the time was claiming, in h1 font “Is Linux ready for the desktop”? 1997 and the release of KDE was, I believe, the start of the “Year of the Linux desktop” movement.
1999 – Caldera OpenLinux 2.3 was released, touting a GUI installer and admin tools that were so easy to use, a monkey could supposedly have installed and administered Linux. Caldera received rave reviews, won several awards, and was touted as the Linux distribution that was truly going to bring Linux to the desktop masses. Of course, we all know how this fairy tale ended. The company that was once the darling of the Linux community, became the most hated company in Penguin Land after it aquried SCO’s UNIX properties, changed it’s name to SCO, and then turned like a rabid dog on the Linux community it had helped to build.
2004 – First version of Ubuntu is released. Again, Linux enthusiasts were claiming victory. Truly, this was finally the year of the Linux desktop. It had to be. Ubuntu made things even easier than Caldera had Surely the desktop masses would jump on the Linux bandwagon now.
2007 – The year of the Netbook. This was supposed to finally be an untapped resource where Linux could compete from the ground up. Linux enthusiasts were once again touting the virtues of Linux on netbooks and how Linux was going to dominate the netbook OS market. Of course, this never materialized either, and most consumers overwhelmingly chose Windows XP based netbooks.
Those are just the “Year of the Linux desktop” events that come to mind without even having to think too hard about it.
Edited 2011-06-17 15:22 UTC
It’s not superior in the right segments. I mean, saying desktop Linux is superior for productivity compared to iOS isn’t really relevant for the success of iOS, since its success is based on being great for consumer gadgets, pushing content — at a price! — to buyers. Desktop Linux certainly isn’t a good consumer OS, lacking as it is in support from the content industry.
My point was simply that if those gadgets are ready to compete with the traditional desktop, then so has Linux been for quite some time. Or, in other words: those over-hyped things of the future that supposedly will take over everything aren’t nearly ready for that, but that doesn’t seem to influence perception. And Linux, despite providing a very usable desktop and in many respects being far easier to set up, maintain and use than Windows 7 (even when Windows comes pre-installed), has never had that kind of hype.
Perception is hugely important in the market, but people are often wrong. VHS won instead of Beta, Windows instead of OS/2, the Germans are closing their nuclear plants, etc.
Luckily, Linux doesn’t depend on a huge market share in order to thrive. As long as it has a fair share of developers (it does), it will get new software and updates to the old.
One word: marketing.
Apple and Microsoft have over thirty years experience building and marketing systems and software for non-technical end users.
Apple sells an upscale urban lifestyle. Microsoft, solid middle class values. Ideological purity or political correctness is not part of the deal — and that in itself is liberating.
Because its superiority is and has always been bullshit.
In his book “After the software wars,” Keith Curtis says:
“For example, Intel claims to be a strong support of Linux, but is doing only a decent job in its support of Linux drivers. An Intel engineer told me at a Linux conference that their Linux efforts are just 1% of the
manpower that their Windows efforts receive. Doubling their Linux development team would cost less than .1% of their total R&D.”
Intel probably won’t invest more because they don’t see a big future in the Linux desktop. In other words, Linux isn’t gaining market share, because it has so little market share.
Tbh the platform hasn’t gained traction on the desktop in general (which the author admits in the story itself) …
However I know the BBC iPlayer for desktop uses it and apparently tweetdeck.
I can’t imagine why anyone would use the iPlayer in air because it works quite nicely in a web browser via flash and the iPlayer website (which is the best thing ever from the BBC) … maybe the set-top box version works on AIR?
TBH I think they are focusing resources on the clients they do have and I think these clients mostly support Mac and Windows installations.
I have personally been interviewed by several companies that use Adobe Flex (silverlight competitor), several times but never has there been a company using Adobe Air or seen any jobs advertising for an Adobe Air dev …
Edited 2011-06-16 23:36 UTC
The reason I use the BBC iPlayer desktop app (which uses AIR), is downloads. I can download shows to my laptop while connected at home, and then watch them on the bus or train to work when I don’t have a connection / have a low-bandwidth 3G connection.
That does make sense.
The only time I ever even considered it was for GOG.com’s downloader as a convenience on the way to running the games in Wine (often just to extract the data files from the InnoSetup bundles so I could run them in native DOSBox/ScummVM) and I decided not to when I discovered that the installer absolutely, positively insisted on being given root privileges.
Aside from various things which seem to support Linux more as an “Oh, we support OSX and Windows, so Linux should work too”, everything I’ve seen targeting Linux has used Qt or Titanium SDK anyway… and Titanium SDK is essentially an open-source AIR-alike, so no. I won’t shed a tear for it.
Edited 2011-06-17 00:04 UTC
I don’t use any Air applications at all. In the past, I used one, but that is only because we had to use it at work. Can’t say I have missed that app, or Air in general. Flash may have carved out a nice niche at creating rich Web applications. But it doesn’t seem like consumers really cared about being able to run Flash apps on their desktop.
Wouldn’t surprise me if in the long run, dropping Air on Linux is the first step to dropping Air altogether. There might be almost no interest on Linux because of Linux’s limited desktop market penetration. But I suspect there’s not that much interest in Windows or Mac land either.
Not a big loss. Who uses AIR anyway?
Is there anyone who cares about AIR on any platform? I find this rather comical – some big wig at Adobe post on his blog (charts and all!) about how Linux desktop market share is stagnant so they are are now going to kill support for it…..in a product that no one uses anyway. Brilliant.
ps Adobe: You guys need to stop trumpeting the “100 Million Installations” silliness. Sure, LOTS of people installed AIR – some may even use it for something, but the vast majority ended up deleting it after the curiosity wore off. I’d be willing to bet the 100 million number is at least an order-of-magnitude bigger than the number of people who actually run AIR, even occasionally.
AIR == Silverlight == Dead End.
I’m sorry, but its true. All these rich-media “platforms” are just something novices use as a crutch until they figure out how to build HTML5 applications the right way…
Edited 2011-06-17 01:57 UTC
Also, some other Adobe products installed Air automatically a an “opt-out” extra payload for a long time. IIRC, the flash installer for Windows did this for a long time. Since most people just click “Ok” on installers without actually bothering to read what the installer wants to install, a lot of people who have never used Air for anything no doubt got Air installed on their systems just because Adobe was including it as a rider on installers for some of their other apps.
Edited 2011-06-17 02:40 UTC
Just think about how many Firefox installs there are at release time(about 2 million per day afaik). And Firefox is at 25-30% marketshare.
AIR is not really Silverlight … Adobe Flex is the direct equivalent. They both use a specific XML namespace for building the GUI.
Tbh I kinda like Silverlight. It is a nice bit of tech … but Flash is good enough and was already out there.
It depends ultimately what you are using them for.
There are still somethings that I expect HTML 5 and related tech can’t do yet.
One less attack vector/security hole to worry about.
Air was a failed experiment anyway.
AIR runs on Linux?
not anymore
(anybody who use outdated Adobe product is an idiot, so it is better to say it that way)
So, they’re dropping all Linux development? No? Not Reader? Then what’s he talking about?
Actually, last I read they were not porting Reader X to Linux.
from the X FAQ:
“Adobe Reader supports multiple platforms, including Windows®, Mac OS, and mobile operating systems. For detailed information, see the tech specs page. Previous versions of Reader also support Linux and Solaris operating systems.”
http://www.adobe.com/products/reader/faq.html
They might as well drop Reader. I mean do any Linux user’s actually use Adobe Reader? Given all of the FOSS alternatives? I can’t remember the last time I have installed Reader on Linux.
Although, Reader is a QT app. In theory, it shouldn’t cost them much to support Linux right? Should be basically just a recompile?
Edited 2011-06-17 13:18 UTC
Wow, it is? Last time I used it it was still a motif app.
It’s been QT for awhile now. At least on Windows and Mac. I’m pretty sure the latest Linux versions are QT as well. But yeah, my point exactly. The last time you used Reader on Linux was so long ago that it was still Motif based. Just goes to show, that if Reader on Linux goes away, no one will care. No one even uses it since we are all using FOSS alternatives on Linux.
Edited 2011-06-17 15:17 UTC
Yeah, it was probably, what, six or seven years ago and then only to install it and realize how much it sucked ass and immediately uninstall it.
I’ve used it in the past to fill out my tax forms electronically.
If they drop reader, that could pose a problem. Okular’s form filling was simply not up to par last time I tried it. Mebbe it will be better next time. (I can always hope.)
Wait: how come Air won’t run on linux anymore while it runs on Android?
While Android technically is Linux because it has a Linux kernel, it doesn’t have much in common with what people usually mean when they say “Linux”.
I used it for some obscure Flickr photo browser/ripper.
Feels sad that they let it go – just not as sad as seeing Reader X for Linux go.
That is the fate of closed-source proprietary non-standard based software.
One of desktop Linux OS most overlooked role is “poor man’s Mac OS X” when it comes to Audio/Video production. Windows Vista/7 is completely utterly unusable for these use cases. Hardware audio acceleration anyone?
—
RimZi
Oddly while I was placement, one of the guys who worked there was a musician that produces music for many well known games and he used Vista on a MacBook Air.
Edited 2011-06-20 06:14 UTC
My experience of Air on Linux (Ubuntu 10.04 / 10.10 / 10.04) has been tale of continuous crashing.
Meh is all I have to say about Adobe Air on Windows or Linux. Whether it be the constant crashing of BBC iPlayer on Linux or the inability to download purchased games on Windows using the now obsolete GOG Air download tool.
Adobe Air is a poor quality product.
Edited 2011-06-17 19:46 UTC
The norwegian company Aspiro has a spotify-like media streamer called Wimp, that uses AIR to support “all” platforms (At least win+mac+linux). This is a good service, so it’s too bad it’s probably going down the drain Linux-wise because of this.
Why on earth they implemented it with AIR is another matter.
Same for the german clone of Spotify called Simfy.
And Spotify or any other good alternatives aren’t available here.
I didn’t realize this hadn’t been announced yet. It was only a few nights ago that I decided to see what Linux options Pandora has since I wanted to try it on my Atom nettop running Meego. It turns out that their Desktop app was written for AIR. It was the AIR system requirements page that mentioned their deciding to focus on mobile devices, etc. Then a couple days later, the blog post. I’ve heard so little about AIR since its launch that it didn’t even occur to me that this might be newsworthy. Heh.
Browser: Mozilla/5.0 (X11; Linux i686) AppleWebKit/534.30 (KHTML, like Gecko) Ubuntu/10.04 Chromium/12.0.742.91 Chrome/12.0.742.91 Safari/534.30
Kindof offtopic, but Pithos on Linux is great for Pandora.
I didn’t even use it once, and I could care less.
I have used Air and could not care less. Why could you care less?
“a significant market for desktop Linux by 2005. Obviously I was wrong.”
Though the significance of AIR is rather low, pulling the support of any application from any operating system is akin to recommending to all users that they not use that operating system. That may not be the intended communication, but that is the perceived communication by the general consumer market.
The quoted statement above is notable because it is companies like Adobe who have a direct impact on what operating systems end users use. If CS6 was released for GNU/Linux tomorrow with Windows and Mac OS X versions following no sooner then four months, GNU/Linux usage would start increasing immediately.
The statement above is interesting because if every company waits for every other company to support something there is obviously no progress. Why was he wrong about his predication regarding the growth of endpoint GNU/Linux usage from 1999 to 2005? Because growth of an operating system is directly proportional to the growth of industry centric applications being primarily created for that operating system.
If Quickbooks, Photoshop, Autocad, Bloomberg Terminal, and other industry centric applications were primarily developed for GNU/Linux, the increase in usage would be immediate.
You may be asking, “Why does the operating system used by end users matter to application developers?”. There could be many answers, but lets try these quick points:
– Receive more information easier about the operating system for which you are developing. Instead of battling with unplanned OS changes, trying to pry info from the OS vendor, and generally working as a subordinate of the OS vendors, develop for on operating system that fully works for YOU while they stay in the background. The bottom line is your bottom line development cost decreases over night.
– If your users use better operating systems such as GNU/Linux, then their IT maintenance cost is immediately reduced. Currently the majority of most companies’ and end users’ IT expense is maintenance such as reinstallations due to instabilities and security vulnerabilities and other issues sourced by inferior operating systems. When they update to modern OSs, those expenses vanish. They now can direct that saved IT budget to more frequent applications upgrades and thus application developers have more funding for R&D thus more funding for innovation and every body benefits from better applications. In short, any money a user spends on maintaining their operating system is money they can’t spend on applications. It is therefore in the best interest of every application developer (especially companies such as Adobe) to both enable migration to better operating systems and be extremely vocal in recommending that users do update to better operating systems such as GNU/Linux.
Adobe identified the problem, but failed with the solution. Problem: there are not enough endpoint GNU/Linux systems. Solution: increase GNU/Linux development priority; release primary applications first for GNU/Linux maximizing the flexibility and power such an operating system provides and only later release scaled down versions for legacy operating systems such as Mac OS X and Windows.
Instead, companies that release primary applications only for legacy operating systems keep users trapped using those operating systems; the application developer such as Adobe is not doing the user any favors by requiring they continue to use those systems.
To me, it’s actually akin to recommending that users not use the technology that is being pulled. But that is something I’ve been evangelizing for a long time now. If you are a Web application developer, especially one trying to sell SAAS to other businesses, you are a fool if you develop the client side of your application using Flash or Flex (which of course, requires Flash). The reason is simple. Sooner or later, you are going to run into a potential customer who’s corporate IT security policy forbids Flash on their workstations. And when you do, that’s going to cost you a potential customer.
It’s far better to stick with AJAX and powerful Javascript libraries that ensure your application will run on any browser without requiring a plugin that potential customers may or may not allow on their systems.
Also, with both Firefox and Chrome adding “desktop application” modes to their browsers, you can do the same thing using Javascript that you could do with Air.
Edited 2011-06-19 22:38 UTC
Assuming for the sake of argument that say a Flash solution is the best technically for a given client you advocate not using it because some future hypothetical client might object. I don’t agree with this philosophy.
If you are building software specifically for a given client, and that client is paying you to develop the software for them, that’s a different story. The client owns the rights to the software you develop then. In that case, if the client allows flash, and it’s a good fit. Then go ahead and use it.
I’m talking something where you develop a more generalized solution to a business problem that you might want to try to sell to 50, 100, or more customers. Think of it this way, if Google Apps, and GMail were written in Flash instead of Javascript, that would be a deal breaker for some potential business customers that don’t allow Flash.
I’m talking software more along those types of lines. You may not have any customers when you develop the software. You plan to sell it to them as a service after you have developed it. In that case, it’s foolish to use flash / flex. Because not all businesses allow flash. So some future customers might reject your software because it requires flash.
Edited 2011-06-20 21:47 UTC
In Poland the government issued AIR application for yearly taxes declaration (about 1mln people used it). I successfully run it on Linux.
Maybe Adobe feels linux exists in a vacuum.
Seems they think one must open Windows to increase AIRflow…
I certainly care !
But do anyone care if I care ?? 😀