The hurdles to an open source webOS | webOS Nation

The hurdles to an open source webOS

by Derek Kessler Sat, 10 Dec 2011 9:48 pm EST

HP has committed to going open source with all of webOS, and they'll be kicking off the process in the near future (likely early next year) with the open sourcing of the Enyo application framework. The OS itself will take more time, and HP currently plans to release it to the open source community in segments (there will be a roadmap, we've been told).

The reason for the delay is actually quite simple, but its implications could be quite broad for the open source version of webOS. The operating system as it stands today contains a number of parts that have been licensed from other parties, and HP can't go and distribute somebody else's work into the open source realm. What HP has to do is "scrub" webOS, to go through it and pick out all of the proprietary bits that they don't own and replace them with already available open source code, create their own code and distribute it as open source, or - if not vital to the operation of webOS - distribute the open source project without it.

Certainly there are plenty of under-the-hood parts that HP is going to have to replace, but there's one very obvious user-facing set of code that's going to have to be pared down, and there's not much HP can do to replace it: Synergy. Right now with a webOS device you can use Synergy to tie your photos, contacts, calendar, phone, tasks, video, documents, and messaging into AIM, YouTube, Facebook, Google, LinkedIn, Exchange, Skype, Photobucket, DropBox, and Yahoo!. All of those are closed source systems to which HP does not possess the rights to distribute as open source. So it's likely that the eventual open source release of webOS will not come with any Synergy services built-in.

But what it will likely come with is support for Synergy Connectors. This system was introduced with webOS 2.0 but as far as we're aware hasn't seen any public utilization. Synergy Connectors allow third parties to build their own Synergy plug-ins to integrate their services into the webOS experience on a user opt-in basis. It was a novel idea, but one that hasn't seen any traction. And open-source version of webOS would come without any Synergy Connectors built-in, but we can see a way that HP could release the current set of Google, Facebook, Skype, etc as individual closed source Synergy Connectors, enabling even webOS open source users to add their services of choice to the OS. Heck, they could even provide the option on start-up of the open source webOS to download and install user-selected Synergy Connectors.

Another stumbling block to HP going open source with webOS will be the cloud services behind it. Namely, backups and the App Catalog, plus the over-the-ar distribution of updates to webOS. HP has stated their long-term multi-year commitment to webOS as an open source project, understanding that it could take years before it gains any traction as a viable alternative to, well, anything. Part of that commitment will have to come in the form of these cloud services. But will that be a commitment to the HP-distributed version of webOS, or for the open source version? Will there be a difference? Could HP conceivably divorce the cloud services from webOS and offer backup, updates, the App Catalog, and everything else as 'third party' closed-source add-ons? Or can those go open source too?

That's the big question right now that nobody's really asking: what will an open source webOS look like? We've touched above on some of the changes that will have to be made to webOS to enable the open source future, but there are still plenty of questions left to be answered. Will HP's distribution of webOS be solely as an open source project, developed entirely in the public space with it open to code input by anybody at any time? What sort of governance structure will guide this sort system? HP can be there to be 'in charge', but to what degree?

Or will they go the Google Android route, with full versions developed behind closed doors, distributed to partners with closed source elements, and then eventually released as an open source OS, free of the proprietary bits? Then whoever wants to play with it can do so, but they'll have to sit and wait for the next version release, never knowing exactly what HP will spit out.

Open sourcing webOS could be the one thing that actually saves the OS. But it all depends on what open source webOS is going to look like. There are valid arguments for both the completely open open source model and the Android model, but we don't think that either would be a good solution for webOS. Is there a middle road to be forged, just as going open source was the middle road between making new devices and just killing webOS? Only time will tell.