Distributions

WebOS reborn?

By Jonathan Corbet
December 12, 2011

On December 9, HP ended a long period of rumors and speculation with the announcement that it would release the code for its webOS platform under an open-source license. Very little information beyond that brief press release is available, so the net has duly responded with lots more speculation. To some, webOS is about to start a new and better life; to others, this announcement is the last gasp of a dying product. Never one to let a prime handwaving opportunity to pass unexploited, your editor has written some thoughts of his own.

In many ways, the mobile device market at the end of 2011 is in far better shape than many of us would have ever dared to hope for. Powerful handsets running Linux are ubiquitous, relatively cheap, and, in many cases, mostly open to hacking by their users. A great deal of creativity has been unleashed on both the hardware and software sides, to everybody's benefit. Linux has become the system for the bulk of these devices, and the companies that make them are getting better at contributing their work upstream. In many parts of the kernel, the old problem of missing hardware support has been replaced by the problem of dealing with the massive amounts of code contributed by manufacturers. That is, as they say, a high-quality problem; in many ways, life is good.

Naturally, things could be better for everybody involved. The bulk of those devices are running Android, which falls somewhat short of what many of us would like to see in an open-source project. The direction of the project is closely controlled by Google, source releases have been delayed and withheld (is it truly "open source" if one cannot get the source for the code running on one's device?), and some companies have better access to the source than others. Manufacturers have reason to dislike depending on Google for access, and they worry about being relegated to the commodity side of the business. As has been written here before, Android is a huge and valuable gift, but we can acknowledge that gift while still wishing for something a little better.

The dominant players in this market (handsets and tablets, mainly) are Apple's iOS and Android. The former is not available to other manufacturers, leaving them with a single choice for their operating software. In such a situation, there should certainly be room for another contender. Microsoft might yet fill that space with Windows 8 Mobile, but it does not have to be that way; Microsoft has always struggled in this market. Wouldn't it be nice if another Linux-based system could establish itself as a major mobile platform instead?

There is no shortage of alternatives in this area. Tizen announced itself as a sort of successor to MeeGo in September, but almost nothing has been heard from this project since. Various developers are trying to keep a MeeGo derivative alive as a community-driven project; the resulting "Nemo" project has made a few releases for the N900, but does not appear to be progressing quickly. The GNOME project has its eyes on this market with its "GNOME OS" concept based on GNOME 3; KDE's "Plasma Active" has very similar goals. Canonical, too, has ambitions in the mobile arena. Most of these projects have actively been looking for manufacturers to ship their software, but there are not a lot of high-profile results to point to thus far. Will webOS do better?

Those who have expressed pessimistic views certainly have their reasons for doing so. As noted above, Linux-based mobile platforms are not in especially short supply; webOS is a late addition to a crowded field. Given the time that has passed since HP abandoned its webOS plans and the rumors that went around, it seems certain that HP tried, unsuccessfully, to find a buyer for webOS before deciding to open-source it. If nobody wanted to own the system before, what are the chances that they will want to use an open-source version in the future?

That said, webOS is a system with a history of shipping in real products and with a core of enthusiastic users; the alternatives have neither of those. Given code, developers, users, and space in the market, it should be possible for a system like webOS to establish itself and prosper. Getting there, though, will depend on a number of things.

One of those, obviously, is the code itself. Is the quality of the code such that the community can pick it up and carry it forward without a huge amount of cleanup work? Will all of the code be released, or will it be necessary to find or create alternatives for pieces that have been withheld? And, crucially, how long will it take for the code to appear? Every day that passes between now and the code release will decrease the relevance of the whole exercise. If HP wants webOS to succeed, it needs to get the code out there quickly.

Then, there is the quality of HP's management of the project. The press release promises "good, transparent and inclusive governance to avoid fragmentation", which can mean almost anything. "Avoid fragmentation," alas, is often a euphemism for "maintain a firm grip on the project and where it can go." If, instead, HP were to create a structure that gave up some control and showed faith in the community it hopes to build, it could find itself with a crowd of enthusiastic contributors. That said, HP needs to remain at the forefront of that crowd for some time; it will be hard to convince others to contribute to webOS if HP stops doing so. Licensing, too, is a clear concern; some licenses are rather more attractive to contributors than others. HP has not yet said which license it will use, or whether copyright assignments will be required to contribute to the project.

Finally, the code is of limited interest without useful devices to run it on. Google has made a point of ensuring the existence of unlocked devices and making it easy for developers to get their hands on those devices. HP would be wise to emulate this example if it wants to developers to hack on - and improve - the code.

In summary: webOS has a real chance as an open-source project if HP manages things correctly and gets the code out there quickly. There is an existing code base, room in the market, a desire for alternatives, and a group of ready customers. That is far more than most projects have at their launch. The open-source version of webOS has a hard road ahead of it with many challenges to overcome but, with some luck and careful management, there is a real possibility for interesting things to happen.

Comments (51 posted)

Brief items

CentOS 6.1 released

The CentOS 6.1 release is now available; there have also been announcements for minimal, live CD, and live DVD variants. Information on 6.1 can be found in the release notes.

Comments (13 posted)

HP to open-source webOS

HP has announced that it will contribute webOS to the open source community. "HP will engage the open source community to help define the charter of the open source project under a set of operating principles: The goal of the project is to accelerate the open development of the webOS platform; HP will be an active participant and investor in the project; Good, transparent and inclusive governance to avoid fragmentation; Software will be provided as a pure open source project." Details beyond that are scarce at the moment.

Comments (14 posted)

Linux Mint switches Banshee's Amazon MP3 store referral code

In something of a reprise of the February incident where Canonical switched Banshee's Amazon MP3 store referral code so that it could collect the revenue (and share ~~30%~~ 25% of that with Banshee), Linux Mint has now done much the same thing. First reported by OMG! Ubuntu!, it has since been confirmed by Linux Mint lead Clement Lefebvre. So far, the revenue ($3.41) has been negligible, but he seems willing to negotiate a revenue share should that change: "Now, should we share the $3.41/month with Banshee? We could. With Ubuntu? Why not. They're both upstream to us and they're both important to us. If we agree with them on how to share, then it might happen, whether they keep control and share with us, or we keep control and share with them. What's for sure though, is that for this kind of revenue, not a lot of time is going to be spent in negotiations."

Comments (27 posted)

Distribution News

Debian GNU/Linux

Bits from the DPL for November 2011

In the November bits from the Debian Project Leader Stefano "Zack" Zacchiroli covers a Call for Help: press/publicity team, various interviews, so financial information, some legal advice from SPI lawyers, Debian's relationship with other organizations and a few other topics.

Full Story (comments: none)