Alon Swartz's picture

Canonical have graciously sponsored me to participate in the upcoming Ubuntu Developer Summit to be held in Orlando, Florida (starting 24th October). If you don't know what UDS is, see this post.

I'd like to get input and ideas from the TurnKey community on how to make Ubuntu rock even harder than it currently does.

I'll be participating mainly in the server and cloud tracks, but may expand my horizons and join in some other tracks. So if you have any ideas, even crazy ideas - let me know.

Forum: 

My 2 cents

* Avoid mono based applications

* Create 2 types of ISO s. one with necessary audio/video codecs and other without those codecs. In countries like India , where software patent is not applicable , they can use the ISO with codecs.

Alon Swartz's picture

For what it's worth, during installation of Ubuntu Maverick (10.10), the installer asks whether you want to install MP3 playback and Flash support - see this screenshot.

Oh..  i didn't tried Maverick .. It seems good :)

Are those packages residing in cd ? or those packages need to be downloaded from internet ?

For people without internet connection , the package need to be put inside the cd.

Alon Swartz's picture

I'm not sure. I assume they are in the package archive, so an internet connection is required (licensing restrictions?)

Jeremy Davis's picture

That sounds pretty cool!

Not sure if this is completely relevant, but: One thing that really jumps to mind for me is how Ubuntu is constantly moving forward and doesn't seem to spend a lot of time looking back. Now in many ways I think that's a good thing, Ubuntu should be constantly moving forward and trying to make the next release better than the last. But IMO when it comes to LTS releases more should be done to backport bug fixes. An example (which isn't probably that relevant to TKL) is the 'load balancing tick' bug in the Lucid kernel (excessive power usage/heat with some Intel CPUs - particularly noticable on laptops). It has been fixed (upstream) for the Maverick kernel but AFAIK the fix hasn't been backported for the Lucid kernel.

I think for a normal release only backporting security fixes is legitimate but I would like to see more bugfixes for LTS.

My 2c...

Alon Swartz's picture

Yes, it's relevant. I'm not sure about the specific issue you are referring to, but just the fact that it really jumps to mind is what is important, and something possible needs to be improved. Ubuntu do have a have a process called SRU (Stable Release Updates), which is what I think you are referring to. 

Stable release updates will, in general, only be issued in order to fix high-impact bugs. Examples of such bugs include:
  • Bugs which may, under realistic circumstances, directly cause a security vulnerability. These are done by the security team and are documented at SecurityTeam/UpdateProcedures.
  • Bugs which represent severe regressions from the previous release of Ubuntu. This includes packages which are totally unusable, like being uninstallable or crashing on startup.
  • Bugs which may, under realistic circumstances, directly cause a loss of user data
  • Bugs which do not fit under above categories, but (1) have an obviously safe patch and (2) affect an application rather than critical infrastructure packages (like X.org or the kernel).
  • For Long Term Support releases we regularly want to enable new hardware. Such changes are appropriate provided that we can ensure to not affect upgrades on existing hardware. For example, modaliases of newly introduced drivers must not overlap with previously shipped drivers.
  • New versions of commercial software in the Canonical partner archive.
  • FTBFS(Fails To Build From Source) can also be considered. Please note that in main the release process ensures that there are no binaries which are not built from a current source. Usually those bugs should only be SRUed in conjunction with another bug fix.

For new upstream versions of packages which provide new features, but don't fix critical bugs, a backport should be requested instead.

 

Do you think the beef you have with the 'load balancing tick' bug can be SRU'ed?

Jeremy Davis's picture

Lots of reading there. I've just had a quick look at the links you gave and re read the bug report re the 'load balancing tick'. Also out of interest it (if I've read it right) it looks like they may be backporting the Maverick kernel and modules for Lucid under the SRU (security bugs).

I really need to check the state of play with that bug. I've been following it for a while but there seem to have been some recent developments. Some are reporting improvments with the latest Lucid kernel update and some still reporting the issue occuring with the latest Maverick kernel.

On the Laptops I have and look after I ended up going back to Karmic because Lucid was radically reducing battery life and increasing heat. Not a good look when your trying to encourage management to consider open source!

Anyway thanks for the info. Having a read about SRU is good too. Some of the points raised (about the possibility of bug fixes opening cans of worms) are good points and worth consideration.

Alon Swartz's picture

Back in October 2008 I summarized my thoughts on what I refer to as Pymin. It was just after TurnKey's first public release, and we were exploring options for a web administration console (we decided to go with webmin).

The notes might be a little out of date, but the idea's are still valid. If you have feedback, I'd love to hear it, as I am considering proposing a session for it at UDS-N.

Adrian Moya's picture

I now ubuntu as it is targeted at the general public. I did my best trying to deploy a large network of ubuntu workstations (9.04) at a bank and failed. Lot's of things happened, from likewiseopen full of bugs to the people complaining because their 64MB excel archives took half an hour to load on calc. But the experience was great although we failed our objective (replace lots of win workstations). 

During that period of time, profiles for different users were a need. Sure there are some old documents on locking down a gnome desktop. And I also used Sabayon and other thing I don't remember, but everything was so buggy and alpha that I found a big hole in all the linux ecosystem. Beginning from trying to replace Active Directory which I found almost impossible. Of course you can deploy an openldap and do some authentication, but I'm talking about mid-sized enterprises which need to apply politics to the whole network (of at least 100 PCs). And I'm talking of the guy (in this case a girl) which knows how to do things in an AD but doesn't know what a console is. Something as simple as enforcing the wallpaper was a complex issue with the current tools (at the time, I haven't check the progress of these projects)

So if there were some steps to make an ubuntu enterprise workstation or something like that, or a set of tools to easily control authentication/authorization/profiles for different kind of users in an enterprise environment, that would be SOOOOOOOOO NICEEEEEEEEE. After all, ubuntu is for human beings eh? There are humans in the enterprise! 

P.S.: But keep it free :P

Jeremy Davis's picture

Very good points Adrian. I think with the right effort and tweaks the workplace is where Linux in general, but Ubuntu in particular could really shine. It is a huge market where Ubuntu could really capitalise I think. To some degree I think SOHO type business environments are catered for.

But with Samba3 really showing its age and Samba4 still in alpha, the complexity of LDAP (I really can't get my head around that!) and the buginess of Likewise (at least the open version anyway) its definately a serious headache for anyone. Its one of those things that I think if we could get some larger businesses (and/or governements) transitioning to Ubuntu/Linux then it would grow. Because IMO if you discount the gamer market and ignore the reality of the bugginess of a lot of 'enterprise' type packages there are very few environments that couldn't at least seriously consider the move to Linux/Ubuntu.

Add new comment