Saturday, May 14, 2011

Ubuntu Studio Moving to XFCE

As outlined in my last post, the Ubuntu Studio team is currently establishing goals for Ubuntu Studio 11.10 Oneiric Ocelot and I believe we have some incredible ideas for improvement.  One notable decision is to use XFCE as Ubuntu Studio's desktop environment (DE).

I would like to take this time to explain the rationale for this particular consideration.

User Experience
As most reading this blog should be aware, Ubuntu has recently moved from using GNOME 2 (aka GNOME panel) to Unity as the default DE.

Unity represents a huge paradigm shift in work flow, usability, and user experience.  Stability is also another realistic consideration as Unity is a new technology.

Ultimately, the outcome was appreciable concern regarding how Unity would affect the typical Ubuntu Studio user’s work flow.

The consensus within the team is that Unity was not an optimal choice for Ubuntu Studio at this time.

Unfortunately, Ubuntu 11.10 will apparently not ship with the ‘classic’ GNOME desktop (i.e. GNOME panel), which is currently Ubuntu Studio's DE.  While not officially announced, I think this is the logical outcome given GNOME 3's release and there are many, many articles giving voice to the same conclusion.

The unfortunate yield is that the gnome-panel package within the Ubuntu repositories would soon likely begin to suffer bit rot given that upstream (GNOME Foundation) and Ubuntu would both decrease and ultimately stop support and maintenance.).

Therefore, it quickly became evident that staying with GNOME panel as our default DE did not represent a viable choice for project sustainability.  I found this !disturbing (haha @bkuhn).

We could not currently move 'forward' with vanilla Ubuntu, nor could we maintain the status quo.  We needed to find an alternate DE.

Enter XFCE
Several DE’s were discussed but XFCE was chosen because it offered appreciable advantages that other DE's could not.

Some advantages would be more immediately tangible to users. For example, XFCE represents a familiar desktop metaphor (@Fab thanks) for users and provides a more resource friendly environment than GNOME, KDE, or (I would expect) Unity.

Other advantages would more tangible to the Ubuntu Studio team (and perhaps to users later on). A large potential advantage is to develop working relations with the Xubuntu team, which could reduce the work load on the limited Ubuntu Studio team but also increase the actual development yield.

And ultimately, some benefits of this relationship might eventually include a graphical installer and a live image, both of which have been desired by users.

It appears that XFCE seems to provide the most familiar user experience without significantly changing user's work flows while also ensuring sustainability at this point.

Although, we have also taken this opportunity to explore updating the DE's user interface and are seriously considering leveraging the appreciable advantages of AWN.  Cory Kontros has done some wickedly cool stuff and I have been consistently using something similar to his prototype for several weeks with exceptionally good experiences.

I believe that Ubuntu Studio 11.10 Oneiric Ocelot should therefore be one of the most exciting and effective releases in some time given the proposed changes and updates that are planned.  I hope you feel so too.

Wednesday, May 4, 2011

Audacious Use of the Vernacular

It's been a long time since I last apologies.  Going out of town on business one week and getting sick the next got me out of the routine.  Now I intend to get back into said routine.

I am purposefully limiting time spent on this blog post so it will probably have a certain "shoot-from-the-hip" feel to it.

This week's post I want to discuss the recent release of Ubuntu Studio 11.04 Natty Narwhal, a Fun Fact, and glimpses of the future.  So let's be off....

Natty Knowledge
Since we have release Natty, I wanted to highlight and discuss certain elements.

Package Selection
The package selection underwent an overhaul based on developed work flows.  In a nutshell, we worked from a "top down" approach of asking what tasks our users would want to accomplish, then developing a viable tool chain to support those tasks.  This resulted in fewer packages being shipped (and an appreciably smaller image), but a more functional and effective release.

While I believe we have a strong understanding of the audio application selections, I would like to explore finding help to develop the video and graphical applications selections to the same strength.

Already we've reach out the to Libre Grahpics Magazine people to solicit some professional advice.  If you haven't heard of them then you should check them out.  They certainly "walk the walk" by using FOSS to create their work. Saudação!

Package Installation
We also gave users more control over which packages were to installed during the OS installation.  Namely we parsed the audio applications into two generalized camps; recording live instruments/sounds (recording) and sequencers/synths/MDID (generation).

A use case for the recording tasksel would be a band who wants to record.  They would be playing live instruments and recording them and vocals.  This is greatly oversimplified but it's just an example.

A use case for the generation tasksel would be a person who does not play any live instruments but rather creates music with sequencers and synths.  Again, simplification, but it serves as an example.

Plymouth Theme
Great strides were made to improve the Plymouth theme and the lion's share of the credit goes to Cory Kontros.

Last release (I think) we were caught short when Ubuntu moved away from xsplash and I hastily made a Plymouth theme based on the old xsplash theme.  While it looked acceptable, it performed poorly because the progress bar never completely finished filling in.

This time we have a nice spinning animation that does not connote progress, but rather simply action.

And it looks pretty darn good in my opinion.  Thank you Cory!

Ach.  This is one of several that we did not manage to complete in time.

Our team is small, with very limited skill sets and very little overlapping skill sets at that.  And it quite showed trying to get our website updated.  But I hope extremely high hopes for Ocelot.

Why?  I'll be happy to tell you why.

Jorge (Kokito) has joined our team and has been working quite nicely on a new website.  He has an beautifully artistic touch and quite sensible opinions.

Rather that start from scratch, he suggested we take the things that worked from the existing website and use them while replacing and improving the things that didn't work.  Brilliant!

As I mentioned before, I am entirely optimistic that we shall have a new website for the Ocelot release.

As a side note, we are also exploring using our own hosting for the website.  If anyone has suggestions or wants to help with this, please let me know.

Ubuntu Studio Controls
This is another goal that fell flat.

Basically, all functions that were currently included in the ubuntustudio-controls package are broken or no longer needed.  It was so borked and confusing that I pulled the package from the seeds for Natty.

Ailo and Paultag made some good progress during Natty but we didn't quite get it done.  But I expect it to happen for Ocelot though.

Okay, enough for now, moving on.

Ubuntu Studio Fun Facts
Ubuntu Studio Fun Fact #5
Hydrogen now comes with additional drum kits.

I wonder how many people realized before that Hydrogen even offered other drum kits.  Or did they just stick with the GM kit?

There are some really good drum kits available that sound amazing.  And they were in a package called hydrogen-drumkits.  But I think may people did not know about them.

Well, now they are included in Ubuntu Studio or whenever Hydrogen is installed.

Hydrogen also has the functionality to download kits from their website.  In some cases, some kits can't or simply aren't included in the package in the repository but still can be downloaded from the Hydrogen website.

Rock on!

Planning for Ocelot
We are discussing goals for Ocelot and we have lots of good stuff.

We've already mentioned updating the package selection, website update, and ubuntustudio-controls revamp.

We are also expected to get a -lowlatency kernel into the repositories as well.  And this is huge for our users.  Initial tests have shown that we can typically half latencies with this kernel compared to the -generic kernel.  Amazing!

In other kernel news, starting with 2.6.39 the kernel should be able to thread irq priorites without the use of the -rt patch.  This means that firewire users who suffer from irq conflicts can adjust priorities with the -generic kernel.  Outstanding!

We also have a plan to update our documentation.  This is pretty far reaching and may take some time to effect completely, but it will be worth it to our users, especially new users.

Lastly, I will mention ladish.

For those who are unfamiliar, ladish is the LADI Session Handler and a peer of JACK session.  The intent to to be able to restart all the applications with their settings for a particular project rather than open each application individually and then restore their settings.  It's a major undertaking and represents quite a potential.

We should see this for Ocelot as well.

If Nedko can deliver on his vision I think this will supersede JACK session.

Oh, and as a teaser I will mention that I didn't mention one thing.  But I'm not mentioning it yet ;)

Rock on.