Open at Adobe

There is all sorts of news today on the coming demise of MeeGo, the open source mobile system under the auspices of the Linux Foundation. Well, at least on the new general computing devices like phones and tablets.

On so many levels, this, if true, is not good news for anyone who likes their operating environments to be open, as in open development.

MeeGo is truly an open environment. Based on the Linux kernel, it managed to have significant activity in the discussion groups. But if it fails, it will be in a twofold cycle called lack of apps.

Apps require platforms to build on and to deliver on. 

Building is easy. While problematic, it is possible to build on a simulated device. (In the olden days we did this a lot with cross compiled software for “other” hardware. Intel and MeeGo certainly were supportive of the community.

But MeeGo (and Intel) never seem to have enough traction to drive significant devices into the market to create a demand for apps. The lack of devices in the hands of developers continually hampers the potential for serendipitous break thrus in either MeeGo itself or the potential  killer app. (I did have hopes for the Myriad Alien Dalvik application, which purports to run android apps on none android devices. But again, no MeeGo device, no way to test this)

The counter of this can be seen in the development of Linux itself. Linux was built for x386 systems. Such systems were common, relatively cheap (keep in mind we are discussing the earlier 1990′s here) and thus the ability to run Linux on your own hardware was assured. Granted you might have to tweak something, write your own driver, but that was the thrill of being on Linux. (And btw, thank Microsoft for enforcing a common hardware environment.)

Now imagine if Linux had been released for a SPARC platform, or MIPS, or PowerPC. It may have ended up successful, but probably not as fast.

So maybe that is the problem.  MeeGo just hasn’t had time to grow up.

UPDATE: Samsung rumors on MeeGO

Has the meaning of how we use “open” become diluted? 

Way back in the day, when arguments about “Free Software” were around “free as in speech” or “free as in beer”., a new phrase was coined to avoid the issues around free. And thus Open Source was born.

Ever since then we’ve been seeing more activities attached to open. Off the top of my head I can think of:

• Open Source
• Open Standards
• Open Specifications
• Open Core
• Open Ecosystem
• Open API
• Open Governance
• Open Data
• Open Code
• Open Development
• Free/Libre Open Source Software
• Open Community
• Open Roadmaps
• Open Surface (my new personal pet peeve. Why would we need this?)

According to a LinuxInsider article, Adobe is whiplashing the world in resurrecting Flash Player on Linux.

Actually, while it went quiet for a while, it was never dead. In fact, the website pointed out that it would be back, once Adobe got caught up with the various bits needed for 64b support.

Adobe launched the 64b Labs preview in July.  An update went live in August.

And so far, the response from the Linux community has demonstrated their interest.

As of last week, Linux downloads were in second place, behind Windows. This is great for those of us within Adobe working to extend Linux support.  Without obvious indications that there is a market and demand, it is hard to agitate for continued support.

Perhaps the confusion is that Adobe AIR for Linux reached EOL a couple of months ago, thus causing all this angst.

Way back in 1999 I predicted a significant market for desktop Linux by 2005.  (I was targeting better than Mac OS type numbers, in the 10-15% range.)  It was clear back then that Linux had found a substantial adoption as a server OS, and it seemed only time before the desktop adoption rivaled at least “the other desktop”.

Obviously I was wrong.

This is the current market share for Linux from Netmarketshare, who provide great insight into global market share of web technologies. While its never quite safe to read into statistics, I believe there is a correlation between shares of browser usage and desktop operating systems.  It’s likely these days that desktop users are browsing the internet; it’s also likely that such activity is not reflective of server operating systems.

Now, let’s take a look at two other OS shares:

Obviously the trend line in both of these cases is climbing. Adobe has certainly noticed it, our customers and partners have noticed it, and I’d bet that if you write or sell software, you’ve noticed it. The market is shifting to a mobile client, increasingly focused on the delivery of rich experiences and applications that travel with us via phone or tablet.

So we, Adobe, also need to shift with the market.  We need to focus our resources on the rapidly expanding markets like Android. we need to focus our resources on efforts around emerging web technologies as they become open standards.  We are beginning to focus on delviering more open capabilities in technologies that will expand those markets, like jQuery (see jQuery blog and Adobe Edge newsletter) and CSS Regions (news on cnet and on Adobe Labs).

So, with Desktop Linux, we see a basically flat growth curve hovering around 1%. And since the release of AIR, we’ve seen only a 0.5% download share for desktop Linux.

For Android and IOS we see substantial growth in share, and see predictions that indicate that in the Mobile OS market, the Android share could be 46%, with iOS at 16% (IDC March 2011).

The official announcement can be found here: http://blogs.adobe.com/flashplayer/2011/06/adobe-air-and-linux-increasing-distribution-on-devices.html

So, with the 2.7 release of AIR, we made a decision to prioritize our resources towards a Linux porting kit for AIR, which our Open Screen Project partners can use to complete implementations of AIR for Linux-based platforms. As such, 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.

But, to quote Jeffrey Hammond of Forrester, “There are already 100M Linux devices, it just so happens they’re running Android.”

This move will allow us to focus our platforms on the future of Linux clients – on mobile devices. Improving performance, expanding capabilities, cross-device compatibility, stability are things we think are necessary to drive the new Linux market.

And the new Linux client is where we and our customers and partners are focused.

There are a number of useful sources of information on the web that can help make a decision for a company to adopt open source (either directly as in to use, or in extension as in to enhance).  There are a number that help lead through the release of technology through an open source license. I’ve created a number of such lists for companies over the years and while the primary questions always seem to focus on legal issues, the range of questions is much broader.

Usually they focus on the self publication of such technology.

However, there is also the practice of moving technology to an outside organization. It seems that the list of such organization/foundations is constantly growing, and that each offers certain focus or advantages. Apache, Eclipse, Mozilla, Linux Foundation all immediately come to mind when I think of such foundations that provide active efforts and governance for multiple projects, either unrelated or semi-related.  I’m sure that you have your favorite as well; feel free to tell me about them in the comments.

But if there is a process for deciding the release the code in the first place, there should be a similar list that targets outside organization contributions. I spent some time thinking of the things I’d look for and came with the following ten. And I expect there are potentially dozens more that could dig ever deeper into this question.

Note these are in no particular order.

• Does the target organization have a solid record of delivering on new technologies, particularly new contributions?
• Does the organization offer a start up process for new projects?
• Do they have a consistent and acceptable set of governance policies and procedures?
• Is there clear indication that the organization itself is stable and well funded?
• Do we already have experience with the organization?
• Does the organization have expertise in projects that relates to our offering?
• Does the organization itself offer projects or technologies that compete with our offering?
• Is the organization itself open and consensus driven?  How does one (individual/company) join?
• Does the organization provide the capability for the project to live on should the donator stop active contributions?
• Does the organization require assignment of copyright or other intellectual property to the organization?

I like to hear your ideas around choosing what foundations make sense for projects. Leave me a comment with your question to add to the list.

davemc