Many FSF Priority Projects Still Not Progressing

Written by Michael Larabel in GNU on 22 April 2012 at 05:04 AM EDT. 73 Comments
GNU
Last October I wrote about the sad state of the Free Software Foundation's high priority projects... Most of the projects are basically not going anywhere. Many of them at the time were not really advancing in their goals, haven't had releases in a while, or coding hasn't even started. It's been more than a half-year and still there's no significant work towards clearing many of projects from the FSF list.

Most of the comments in regards to the October article that was critical of the Free Software Foundation's list agreed with me. Unfortunately, for most of these projects they still have a long ways to go before reaching their goals and some haven't even been touched in at least a half-year.

Here's a recap on some of the Free Software Foundation high-priority projects as of today.

The list of priority projects can be found at FSF.org.


Some of these high-priority projects are as active as this bumper car from Chernobyl...
[Via my time at Chernobyl.]

Gnash is one of the more successful FSF projects on this list, but it doesn't mean a lot. Gnash supports most of SWFv7 and most ActionScript 2.0, but beyond that is where its support for modern Flash is really lacking. The SWF file format is currently up to version 10. Doing a much better job at supporting modern Adobe Flash files used on the web is Lightspark, which can also fall-back to using Gnash for older Flash/SWF files. Lightspark had a new release earlier this month and is making steady progress in supporting more Flash files, but still has issues sometimes with even supporting YouTube videos. There's also no support yet for popular Flash games like Farmville.

There was the Gnash 0.8.10 release earlier htis year, but it was a rather mundane release with just Qt4 GUI support, UI fixes, an OpenVG renderer, touch-screen support, refactored input support, and some other overdue items. Having better support for Adobe Flash/SWF on Linux is especially important now seeing as Adobe will no longer be offering new Flash releases for Linux. Gnash/open-source-Flash certainly is very worthy of being a high priority project considering how popular the Adobe software remains, but it's unfortunate that these open-source alternatives are advancing at a snail's pace.

Another important FSF high priority project that has some success and at least is still being worked on frequently is Coreboot as an effort to have a free/open-source BIOS replacement. While Coreboot is actively being worked on and continues to support new motherboards/chipsets, it's seen limited commercial success or interest from tier-one vendors.

Coreboot continues to be supported on all new AMD hardware, but still even with AMD's official backing there aren't any tier-ones actually shipping it with their motherboards or laptops. The Coreboot folks at FOSDEM were a big disappointment with their "Coreboot laptop."

The success out of Coreboot recently is Google providing Sandy/Ivy Bridge support for Coreboot. Google's planning to begin shipping new Intel "Chromebooks" that will use Coreboot. Google likes Coreboot for the faster start-up time, among other benefits. With Google's weight behind Coreboot now and actually shipping products with this open-source BIOS/UEFI replacement, this is really exciting. However, Google Chrome OS mobile devices haven't exactly been flying off the shelves.

I remain quite hopeful for Coreboot as a project, but in the forseeable future I don't see it shipping on too many consumer products. There's also financial institutions, trading firms, and similar organizations that like Coreboot for its fast boot times and being open-source to verify greater security. Coreboot may have better luck here with garnering the interest of OEMs.

The list of motherboards known to work that can be flashed with Coreboot can be found on this Wiki page.

While Gnash and Coreboot are at least making some ground, a free software replacement for Skype really isn't reaching any critical mass. There are some open-source communication programs like Ekiga, Twinkle, Jitsi, but for widespread use on the Linux desktop, Skype's Linux client remains much more popular and feature-rich than these open-source alternatives. Until there is some really compelling multi-platform VoIP/communication program, this work item will remain on the Free Software Foundation's list for a long time.

Another high priority project that's actually an endorsement of several different projects is to create a free software video editor. There are some usable programs like PiTiVi, OpenShot, Cinelerra, and Novacut, and Kdevlive. However, I have yet to see one that offers an impressive feature-set and can be competitive with the professional non-linear video editing applications available for Windows and Linux. I've also yet to hear anyone actually switch to Linux on the basis of desiring one of these video editors.

There is some hope with Lightworks going open-source, but the Linux client has been delayed. In terms of the state of the current open-source video editors, OpenShot had a release in February (1.4.2), CinelerraCV 2.2 arrived in November, Novacut is still "coming soon", and Kdenlive 0.8.2.1 was released last December. They're making progress on the video editing front, but nothing rapidly.

A free Google Earth replacement is another long-standing item. The most promising potential replacement for Google Earth would be OpenStreetMap. Depending upon the geographic region, there can be a decent level of detail.

One of the less detailed items by the Free Software Foundation is Help GNU/Linux distributions be committed to freedom. The item calls for supporting distributions like Trisquel and gNewSense because they are "distributing a complete GNU/Linux operating system that contains only free software. They are two of a list of high-quality distributions that modify Debian and Ubuntu to create a complete free operating system without any binary-only blobs or package trees that contain proprietary software." This item will likely remain on the FSF's priority list indefinitely; don't expect one of these niche distributions to suddenly unseat Ubuntu, Fedora, and other major distributions.

For catering to the scientific open-source fans, GNU Octave has been on the list as a free software replacement to Matlab. GNU Octave has seen some success and is still being actively developed (GNU Octave 3.6.1 was released in February) and they do hold an annual conference even (OctConf).

One of the odd items that remains on the High Priority Free Software Projects list is LibreDWG. The FSF views having a free software library to handle CAD format files is a big priority for open-source. LibreDWG hasn't seen a release yet and its code-base hasn't even been touched since January of 2011 (the Git repository). The LibreDWG mailing lists are also quiet. I'm not sure that an open-source CAD file library will lead to a boon in free software...

Reversible Debugging in GDB is also deemed a high priority. Here there is already reversible debugging support in place with GDB, they just want more love for it.

Free software drivers for network routers. The project the FSF mentions for this is OrangeMesh. Their web-site reads since 2010, "Due to other commitments, development on this project is currently on hold. Please contact us if you're interested in taking over project development." Like the need for an open-source library for CAD files, I'm not sure how many Linux users are losing sleep over not having open-source drivers for their network router.

Free software replacement for Oracle Forms. The only work this FSF item has is a discussion mailing list to replace Oracle Forms with free SQL databases. There's been a few mailing list messages in recent months lightly discussing this work, but nothing has yet to materialize.

Automatic transcription is another sought after piece of software by the Free Software Foundation. Like the Oracle Forms replacement, this initiative too doesn't have any actual work committed to it under the FSF umbrella. Rather than a discussion mailing list, this imaginary project instead has a Wiki. This Wiki hasn't been touched since last year.

The biggest farce of all, and the last on the list, is the PowerVR drivers. This has been a high-priority FSF project for more than one year yet there's virtually no work towards this goal. Having an open-source PowerVR driver with OpenGL/GLES acceleration would be huge, and many Linux consumers would be incredibly excited, but this FSF project has just been sitting in the water.

At last check, the "developer" behind this project was waiting for money before starting any work on creating an open-source PowerVR driver. He also isn't involved with the upstream Mesa/Gallium3D or X.Org development communities. The goals include writing a Gallium3D PowerVR driver, OpenGL 3.0 support, LLVM leveraging, and a DirectX accelerator for use under Wine. There's reportedly been some reverse-engineering of some opcodes from the start, but there's nothing to show for this work. Its Wiki page also hasn't been touched since last year.

Meanwhile, other developers have single-handedly taken to other hardware and done clean-room reverse-engineering work for other embedded graphics processors. There is the promising Lima project for a reverse-engineered ARM Mali driver, the new Freedreno driver for Qualcomm's Snapdragon, and a Tegra 2 DRM driver, among other smaller open-source driver initiatives.

By now any of the more well known open-source Linux graphics driver developers could have had a reverse-engineered, working open-source PowerVR driver. The issue is that most of the well known open-source graphics developers aren't allowed to touch PowerVR. With PowerVR SGX graphics being so widespread, most of these developers are tainted with NDAs and other restrictions from their employers. Luc Verhaegen had to do the ARM Mali driver because he's PowerVR-tainted from work he did for Nokia, there's many other open-source developers tainted by their time at Nokia, Rob Clark went to reverse-engineer his competitor's hardware (the Snapdragon driver) because Texas Instruments uses PowerVR on OMAP, and Intel is also currently using PowerVR in their SoCs (fortunately, that's changing in a great way).

The Free Software Foundation does continue to solicit donations for "High Priority Free Software Projects", but there's no word on how this money is being spent or how much the foundation has actually accumulated.

Another issue I have with the list is that these high-priority items are mostly about replacing existing functionality offered by closed-source software rather than trying to drive original innovations to make open-source software superior to the proprietary competition. Having real innovations on top would be more likely to attract new users of open-source software because of being a better product than just a rehashed version of something that's long been available as a proprietary product.

The only completed high-priority projects from the Free Software Foundation have been an unarchiver for the RARv3 file compression format, GNU PDF, and CiviCRM as a open-source fundraising software system.
Related News
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.

Popular News This Week