Why is Unity 8 development taking so long?
Canonical has included a preview version of the Unity 8 desktop in Ubuntu 16.10. But that has not stopped some Linux users from wondering why Unity 8 still hasn’t been finished.
The topic came up in a recent post on the Linux subreddit, and folks there shared their thoughts about why Unity 8 still hasn’t been released in final form.
Linuxode: “Canonical has a lot of talented developers and a pool of resources. Not to mention the widespread popularity of Ubuntu in the Linux world - which should bring even more contributors from the outside. Even then, Unity 8 seems to be developing at a snail’s pace.
What are the constraints holding back Canonical?”
Blackout24: “New display server + protocol + shell + input server + all the patches they have to maintain for their own solutions that nobody else uses + the other dozen Canonical projects. Way too much for a small company.”
Rbrownsuse: “To put it short and sweetly - because Canonical are doing it alone.
And not only Unity 8, but Mir, the display server it requires.
These are two massive engineering exercises.
Such an effort would typically require either a massive engineering department (at the scale of Microsoft, Oracle, you know something stupidly huge - As a 500+ employee company this is something which Canonical doesn’t have)
The logical alternative to that would be to work together with other companies. This would be in keeping with the typical Open Source way, and is the model that Red Hat, SUSE, Collabora and other FOSS companies follow to put their solutions together in an affordable, sustainable way.
But this is a model which Canonical do not seem to follow easily - They always seem willing to consume from the Open Source ecosystem, but when it comes to contributing back, they seem comfortable only doing it within their own bubble.
From my perspective that really hurts the potential pace of progress for solutions like Mir and Unity 8 compared to alternatives like Wayland & GNOME, which have a more collective push behind them.”
Uoou: “I think one of the reasons they struggle to find collaborators for their projects is their CLA.”
Rbrownsuse: “That is certainly the reason why I will probably never contribute to anything of Canonicals, yes.
It is my very strong personal opinion that I will never sign a CLAs that that grants the license holder the right to re-license my work under a different license than the one I chose for my work.
I’m lucky enough to have an employer that has similar views on such matters, does not encumber any of it’s own projects with CLAs and typically avoids contributing to projects with CLAs and therefore it’s unlikely I’ll ever need to work on a CLA encumbered project as part of my work life either. :)”
Boomboomsubban: “They’re trying to replace and improve upon thirty years of X11 development on their own.”
Uoou: “I broadly agree but I think trust comes into it. I have no problem with the FSF’s CLA since I know exactly why they require it and trust them to stick to that (which isn’t hard since it’s their raison d’être).
Canonical’s CLA is a problem since I don’t quite know why they want it (or, worse, perhaps I do).”
Sherba800: “Who said that canonical has a lot of talented developers? Havent noticed that yet. Canonical team should vote for supporting Wayland instead of for developing Mir as it worked in case of systemd.”
Mordiken: “Lots of users doesn’t imply lots of contributors. Specially for Canonical, which have managed to alienate a substantial portion of their power user base through a mix of community mismanagement, bad decisions, and successful FUD campaigns by their competitors (and respective fanbase).”
Bkor: “Everyone pretty much agreed on Wayland, including Canonical. They then decided otherwise and started Mir. After 6+ months they finally told other people about their decision. They chose to go at it alone and ignore the contributions of others.”
Ventomareiro: “The fact that Mir is released under the GPLv3 and the copyright is assigned to Canonical, has led to speculation that Canonical want to open a business line providing Mir to mobile manufacturers under a different license.
(Not that I see anything wrong with that, it’s their code after all).”
DistroWatch reviews Ubuntu 16.10
Ubuntu 16.10 has been out for a little while now, and reviews are still coming in of the latest version of Ubuntu. DistroWatch has a review of Ubuntu 16.10, and found it to be a “solid, polished distribution.”
Joshua Allen Holm reports for DistroWatch:
At first glance, little has changed in Ubuntu 16.10. It looks almost exactly like every other recent release of Ubuntu and the included applications are the same ones one would expect to see. There is a newer Linux kernel, version 4.8, and Firefox, Thunderbird, LibreOffice, and the rest of the applications one expects to find are present and newer than what Ubuntu 16.04 LTS shipped with. Because all of my computers have Intel graphics, I cannot personally test to see if the updated packages in 16.10 fix or improve the issue with AMD graphics that are present in Ubuntu 16.04 LTS.
The biggest change comes from the update to using GNOME 3.20 applications. The new version of Ubuntu Software, which is GNOME Software 3.20 rebranded, provides a much snappier experience for installing software. The switch to the 3.20 release of GNOME Files (a.k.a. Nautilus) also brings many changes, many of which are far more noticeable than the improvements to GNOME Software. Files now uses the single global menu featured in modern GNOME applications.
Unity 8 has a lot of potential. I enjoyed trying it out, and I do hope that Unity 8 is ready for the next LTS release of Ubuntu because it does have a lot to offer. However, the developer preview included in Ubuntu 16.10 is so far from being ready that I almost suspect that the only reason it was included by default in this release was so there would actually be a desktop-focused new feature in the release announcement.
Ubuntu 16.10 is a solid, polished, usable Linux distribution. However, there is very little reason to recommend it over the previous 16.04 LTS release. There are a few tweaks and some slightly newer software packages, but nothing world shattering. The only compelling reason to upgrade is if Ubuntu 16.10 fixes an issue you were having with Ubuntu 16.04 LTS. That said, there are no major issues with Ubuntu 16.10, so if you are the kind of person who always wants to have the latest packages possible, go ahead and upgrade.
9 Android apps for Halloween
The spookiest holiday of the year is almost upon us, and what better way to celebrate it than to grab 9 must-have Android apps for Halloween.
Derek Walter reports for Greenbot:
You’ve probably decorated your house with cobwebs, spiders, skeletons, and carved some pumpkins.
But what about your phone? Yes, that ever-present companion of yours needs some attention as well when it comes to getting ready for All Hallow’s Eve. Your smartphone can keep track of where the kids are, zombify yourself, and keep the party going with some dark and delightful tunes.
Track & Treat by Glympse (free)
Mickey’s Spooky Night ($1.99)
Halloween Photo Stickers (free)
Google Play Music (free, $10 monthly subscription)
Funny Halloween Party 2 (free)
Halloween Watch Face ($0.99)
Dead Yourself (free)
Halloween Live Wallpaper (free)
Hidden Object - Happy Haunts (free)
Did you miss a roundup? Check the Eye On Open home page to get caught up with the latest news about open source and Linux.