Re: Re: OpenGL to OpenGL ES change on arm64?

[Timo Jyrinki]

2016-06-06 5:23 GMT+03:00 张超 :
> Hi All，
> We have ARM64 devices used AMD Radeon  Series graphics cards。So，same as x86
> we really need a full OpenGL Qt support With ARM64 。

Hi. I noticed this e-mail went unanswered. All Radeon graphics (and
NVIDIA) run OpenGL ES applications fine, so it's not strictly a
requirement. It's the OpenGL ES GPUs that are the majority use case on
armhf/arm64 that do not support traditional OpenGL and its legacy
features all the way from 1991. OpenGL ES 3.1 does support most of the
advanced features of OpenGL 4.5, so applications should be possible to
be made OpenGL ES compatible regardless of how advanced features they
use.

In the long run, upstream does have a bug open to make the switch
between OpenGL and OpenGL ES run-time selectable, like it already is
on Windows.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Xenial startup disk creator and custom ISO

[Timo Jyrinki]

2016-05-06 21:42 GMT+03:00 Darren Landoll <darren.land...@gmail.com>:
> Are there any current issues with creating a custom Ubuntu Install CD
> (server install for my particular case) and using Startup Disk Creator in
> Ubuntu 16.04 to burn a custom *.iso file to a USB drive?
>
> It claims to finish successfully (does not take as long as I would expect),
> but does not result with a bootable USB drive.
>
> I create the *.iso based on instructions here:
> https://help.ubuntu.com/community/InstallCDCustomization

The instructions might be out of date, it'd be nice to update the wiki
page with latest verified/new steps, if someone successfully finishes
the battle of getting it working.

I have successfully created bootable (both when using USB Creator and
simple Disks tool copy which equals to 'dd') live CD remix of 16.04
LTS, where I switched from mkisofs to genisofs [1].

-Timo

[1] http://bazaar.launchpad.net/~timo-jyrinki/ubuntu-fi-remix/main/revision/31

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Default languages strategy for Ubuntu desktop CD

[Timo Jyrinki]

2016-04-21 10:59 GMT+03:00 Timo Jyrinki <timo.jyri...@gmail.com>:
> though, there may be new hurdles hurdles to overcome. I did a quick
> test run yesterday evening but somehow it seemed unbootable ("Failed
> to load ldlinux.c32"), I'll need to look into it during the weekend.

Ehem, I looked at it already now and it seems I had ran out of space
on the partition where I worked on the image...

Other than that, I switched to genisoimage and tested the resulting
modified image successfully by writing it with Disks tool on 14.04 LTS
to USB stick and booting both on a real UEFI + Secure Boot laptop and
testing installation in qemu in Legacy mode. Everything seems
functional and fine.

The only regression is that it seems the Legacy mode now defaults to
English despite "echo fi | sudo tee extract-cd/isolinux/lang". Tips
welcome there... however, the Legacy mode boots to the nice GUI
language selector unlike UEFI mode which requires selecting between
Live and Install modes immediately, so it's not that serious.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Default languages strategy for Ubuntu desktop CD

[Timo Jyrinki]

2015-12-01 10:06 GMT+02:00 Didier Roche :
> Some loco team used to have respin of the default localized ubuntu image.
> Yours truly used to have handle this quite regularly for the french loco
> team. I know that the italian and finnish locos used to do it as well (not
> sure about the current status for them). However, UEFI support made it
> complicated and so those efforts have been dropped some cycles ago as far as
> I know.

No, it was discussed at UOS [1] and not dropped [2], UEFI works, and I
added UEFI GRUB translations as well [3]. Not sure about 16.04 LTS yet
though, there may be new hurdles hurdles to overcome. I did a quick
test run yesterday evening but somehow it seemed unbootable ("Failed
to load ldlinux.c32"), I'll need to look into it during the weekend.
Tips welcome if you know what I encountered, and also feel free to
experiment with the "script" (copy-paste commands manually in reality)
from the bzr using any xenial image as a base. I'd welcome fellow
tinkerers.

[1] 
http://summit.ubuntu.com/uos-1411/meeting/22380/development-1411-iso-l10n-uefi/
[2] https://lists.ubuntu.com/archives/loco-contacts/2014-November/006895.html
[3] https://lists.ubuntu.com/archives/loco-contacts/2015-August/007009.html

> 1. Install full language support for those shipped on xenial image. It means
> that opening "language selector" won't request any additional package to
> install[1].

Thanks for adding these, my script was installing the selected English
packages manually for 14.04.x releases, since otherwise Ubuntu would
complain about them even with full Finnish support installed.

> full support for those 8 languages. This is of course a non negligeable
> image size increase. However, getting full language support for at least 8
> languages (and no additional download on install for those) seems like a
> huge increase in quality to me.

I agree!

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: qt5ct for Wily

[Timo Jyrinki]

2015-09-20 12:24 GMT+03:00 Ralf Mardorf :
> does somebody know a link that introduces how to build qt5 applications
> for Wily?

The closest might be http://pkg-kde.alioth.debian.org/packagingqtstuff.html

> [weremouse@moonstudio qt5ct-0.17]$ qtchooser -qt=5 -run-tool=qmake 
> PREFIX=/usr/local

Just use export QT_SELECT=5 (or export QT_SELECT := 5 in debian/rules)
and run qmake normally.

> main.cpp:29:38: fatal error: qpa/qplatformthemeplugin.h: No such file or 
> directory

That's in qtbase5-private-dev since it's a private header that normal
apps should not be using.

> I wonder why the include paths are qt4.

Possibly since you were only specifying the Qt version to use for the
single tool, and the default fallback is Qt 4. That is, Qt 4
applications don't need to know about different Qt versions, but Qt 5
applications need to for example set that environment variable.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: [ubuntu-studio-devel] Feature Spec Discussion: Desktop Agnostic

[Timo Jyrinki]

2015-09-01 1:15 GMT+03:00 Len Ovens :
> on all flavours... well maybe not Unity (actually I think I did, but
> disliked it so much, I gave up), Unity is a beautiful experience and
> entertaining, but does not make "work" as easy for me.

It depends on the user, but for me it has been the most suitable for
work. The first reason it gives the biggest amount of screen space to
the apps which I like, and the second reason is that the Super + 1-9
hotkeys are enough for me to quickly start + switch between key apps
dragged to the launcher. Super + F for recent files/file search and
Ctrl + Super + Arrows and Super + W/S for window manipulation complete
the set of what I usually use. Another flexibility is that it using
Compiz means it's pretty customizable (aside from the usual woes of
lack of Unity specific location settings), for example one could set
certain apps to always start to certain workspace etc.

But, it depends, like said. For some/many users it's seen as clunky UI
that's mouse driven but meant for touch screens from design
perspective. In my opinion that's only the surface.

-Timo

-- 
ubuntu-studio-devel mailing list
ubuntu-studio-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel

Re: [ubuntu-studio-devel] Wubi - nuke it from orbit

[Timo Jyrinki]

I'm sure almost nobody starts to consider copying an executable off
the installation media and running it that way, but I agree even that
functionality should be removed.

.Timo

2015-08-22 21:10 GMT+03:00  lukefro...@hushmail.com:
 The linked post or one of it's replies warns of bricking windows, if it is 
 moved
 off the disk, the exe file executed from the windows drive, and installer mode
 selected. I have no idea if that is true or not, but be sure to do any testing
 of this beyond the reboot greeter on an expendable install.

 On 8/22/2015 at 10:09 AM, Timo Jyrinki timo.jyri...@gmail.com wrote:

Didn't https://lists.ubuntu.com/archives/ubuntu-devel-discuss/2015-
August/015757.html
just tell wubi has had only a simple please reboot functionality
for
a long time, but causing confusion since it's still named wubi, and
all the documentation everywhere?

It's useful that there's a dialog popping up when a Windows user
inserts USB/DVD.

-Timo

2015-08-22 15:22 GMT+03:00 Kaj Ailomaa zeque...@mousike.me:
 I have only used wubi to install Ubuntu, and my experience with
that is
 not exactly great. The virtual hard drive is really, really
slow. I fear
 that even if Ubuntu Studio works with wubi, it will give the
user a bad
 impression.

 So, there's that too.

 I could give it a try on a Win8 machine during the coming week,
just to
 see how it works.

 --
 ubuntu-studio-devel mailing list
 ubuntu-studio-devel@lists.ubuntu.com
 Modify settings or unsubscribe at:
https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel

--
ubuntu-studio-devel mailing list
ubuntu-studio-devel@lists.ubuntu.com
Modify settings or unsubscribe at:
https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel


 --
 ubuntu-studio-devel mailing list
 ubuntu-studio-devel@lists.ubuntu.com
 Modify settings or unsubscribe at: 
 https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel

-- 
ubuntu-studio-devel mailing list
ubuntu-studio-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel

Re: ISO generator for LoCo teams - UEFI booting language setting

[Timo Jyrinki]

2015-08-22 0:10 GMT+03:00 Steve Langasek steve.langa...@ubuntu.com:
 I haven't looked in any detail at how changing the default language is done
 for live images, but what I can tell you is that any configuration changes
 that you can make via grub.cfg are allowed under SecureBoot.  So if you can
 figure out what your grub.cfg should look like for this, then it should be
 possible to do.

 Anything that refers to gfxboot is not relevant, because we don't use
 (BIOS-only) gfxboot under UEFI.

Right, they are completely separate. Thanks!

I now got all parts working, adding the following to my semiautomated script:

sed -i '6iloadfont /boot/grub/fonts/unicode.pf2' boot/grub/grub.cfg
sed -i '7iset locale_dir=$prefix/locale' boot/grub/grub.cfg
sed -i '8iset lang=fi_FI' boot/grub/grub.cfg
sed -i '9iinsmod gettext' boot/grub/grub.cfg
sed -i 's%splash%splash locale=fi_FI console-setup/layoutcode=fi%'
boot/grub/grub.cfg
sed -i 's/Try Ubuntu without installing/Kokeile Ubuntua asentamatta/'
boot/grub/grub.cfg
sed -i 's/Install Ubuntu/Asenna Ubuntu/' boot/grub/grub.cfg
sed -i 's/OEM install (for manufacturers)/OEM-asennus
(laitevalmistajille)/' boot/grub/grub.cfg
sed -i 's/Check disc for defects/Tarkista asennusmedian eheys/'
boot/grub/grub.cfg
mkdir -p boot/grub/locale/
mkdir -p boot/grub/fonts/
cp -a /boot/grub/locale/fi.mo boot/grub/locale/
cp -a /boot/grub/fonts/unicode.pf2 boot/grub/fonts/

So this includes the GRUB UI, the entries in GRUB (since they are
hardcoded) and the language + keyboard settings of the live/install
session. Naturally doing this before the md5sum.txt is recreated.

I think I'll put a 14.04.3a out or something, as UEFI is becoming more
and more popular.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: [ubuntu-studio-devel] Wubi - nuke it from orbit

[Timo Jyrinki]

Didn't 
https://lists.ubuntu.com/archives/ubuntu-devel-discuss/2015-August/015757.html
just tell wubi has had only a simple please reboot functionality for
a long time, but causing confusion since it's still named wubi, and
all the documentation everywhere?

It's useful that there's a dialog popping up when a Windows user
inserts USB/DVD.

-Timo

2015-08-22 15:22 GMT+03:00 Kaj Ailomaa zeque...@mousike.me:
 I have only used wubi to install Ubuntu, and my experience with that is
 not exactly great. The virtual hard drive is really, really slow. I fear
 that even if Ubuntu Studio works with wubi, it will give the user a bad
 impression.

 So, there's that too.

 I could give it a try on a Win8 machine during the coming week, just to
 see how it works.

 --
 ubuntu-studio-devel mailing list
 ubuntu-studio-devel@lists.ubuntu.com
 Modify settings or unsubscribe at: 
 https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel

-- 
ubuntu-studio-devel mailing list
ubuntu-studio-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel

Re: ISO generator for LoCo teams - UEFI booting language setting

[Timo Jyrinki]

2014-11-10 8:25 GMT+02:00 Timo Jyrinki timo.jyri...@gmail.com:
 The pseudo-script (really needs copy-pasting sections instead of
 executing) I've used to create it is at
 https://code.launchpad.net/ubuntu-fi-remix

All is still working for both UEFI and Legacy boot users with the
Ubuntu 14.04.3 LTS Finnish Remix I released during a weekend after the
release: http://ubuntu.trumpetti.atm.tut.fi/ubuntu-finnish-remix/14.04/

But I did have a persisting problem where I'd like to hear some input:

Is it possible to set the default language on UEFI boot, or is it
somehow restricted due to Secure Boot signing or such? Neither the
simple method I use (echo fi | sudo tee extract-cd/isolinux/lang) nor
the gfxboot hack mentioned at
https://help.ubuntu.com/community/LiveCDCustomization has an effect -
UEFI boots always in English.

Optimally the language would be set for the boot menu already, since
on UEFI the boot menu is always shown unlike on Legacy boot where
there's just the icon if no input is given. But if that's not
possible, maybe setting it before the desktop or installer starts?

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Qt 5.5.0 - bug fixing needed

[Timo Jyrinki]

2015-07-15 17:31 GMT+03:00 Timo Jyrinki timo.jyri...@gmail.com:
 Qt 5.5.0 is testable on wily from
 https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-012/
 - including the new Qt 3D tech preview.

It's still there and now fixed to build against GCC5 so brave ones can
again use wily + that PPA. At this point it's safe to say Qt 5.5 won't
be in 15.10 by the next week's Feature Freeze, and will need either an
FFe or target to wily+1. Currently I've not had much requests to get
5.5 in, but at least the Qt 3D would be very interesting to have.

Bug fixing via https://bugs.launchpad.net/bugs/+bugs?field.tag=qt5.5
has been slow. New bugs are also being found and at the moment there's
a new blocker that prevents testing the PPA on a phone. I will keep
https://wiki.ubuntu.com/Touch/QtTesting updated regarding the status.
The PPA is usable in eg a chroot to fix any build problems.

Debian is also still in the middle of the GCC5 transition and I'd like
to always sync from Debian instead of temporarily forking the
packaging, even if I've now moved a lot of the Ubuntu packaging to
Debian git.

On the plus side, we're on track to delivering Qt Creator 3.5 RC1 and
the refactored Ubuntu plugin soon. A lot of the features in Ubuntu's
Qt Creator were upstreamed during the 3.5 cycle and now also the
Ubuntu specific plugin is more streamlined.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: pretty please can we kill the png optimizer?

[Timo Jyrinki]

2015-08-12 4:34 GMT+03:00 Matthias Klose d...@ubuntu.com:
 we don't ship cd images anymore, we are not limited to CD sizes, and as long 
 as
 we don't hit some 2GB limit, we shouldn't optimize for size.  This optimizer
 adds for some packages 100% build time, in rare occasions up to 2000%.  This 
 is
 not worth the savings.  If we want to optimize for size, this should be done 
 by
 test rebuilds and individual patches, not consuming scare buildd resources.

I'd support this, as the PNG repacking is hugely time consuming on
especially armhf Qt builds. If I recall correctly it's pretty near
that 100% in eg qtdeclarative, making it 2h instead of 1h, and very
slow also with for example qtbase amd64. And whenever doing a big
amount of no-change rebuilds, the build times add up.

In test builds I try to remember to use export NO_PNG_PKG_MANGLE := 1
in debian/rules.

It's not that the feature itself is bad, but optipng is just slow for
the gain it brings on average. Just being able to run it utilizing all
CPU:s would probably help, or maybe different parameters.

There are alternative tools in archives like pngquant, but that one is
lossy (funny for a PNG). Some are faster and do better compression,
but are not free software.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Qt 5.5.0 - bug fixing needed

[Timo Jyrinki]

Hi,

Qt 5.5.0 is testable on wily from
https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-012/
- including the new Qt 3D tech preview. It includes package rebuilds
needed for Ubuntu phones, Unity 8 desktop and Plasma 5 desktop, but I
have only tested the phone so far. On the desktop one would lose Qt
Creator since the old version does not build anymore and our new
version 3.5 git snapshot + Ubuntu plugin isn't ready yet.

I've filed bugs since I started with Qt 5.5.0 alpha in March, and more
now with the final 5.5:
https://bugs.launchpad.net/bugs/+bugs?field.tag=qt5.5 (10 bugs already
Fix Released not shown). Feel free to file more bugs, we don't want
regressions on the phone.

There are many bugs to be debugged before considering QA:ing and
landing Qt 5.5, so please help if your or your team's package is found
in the bug list.

There's also the big GCC 5 transition ongoing which must finish first,
and me + (z)Benjamin have Qt Creator 3.5 + Ubuntu plugin brewing which
I'd prefer to land separately first. Let's see if there's any chance
of getting Qt 5.5 in by the feature freeze - which is on August 20th.

Everything you need to know is at
https://wiki.ubuntu.com/Touch/QtTesting or linked from there. Note
especially not to upgrade if there are suspicious packages being
removed by dist-upgrade. Instead, ask me (or if I'm away, sil2100) on
IRC to update the PPA as needed.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Qt 5.5 early look

[Timo Jyrinki]

Hi,

I've spent a bit of time with Qt 5.5 alpha since its release in March.
Here's the first screenshot of Ubuntu on it:
https://plus.google.com/+TimoJyrinki/posts/Jw6Tx9QhC7S - thanks to the
FTBFS bug fixers of maliit-framework, qtmir, unity8.

More information at https://wiki.ubuntu.com/Touch/QtTesting. WARNING:
Not for end-users of any kind, only for fixing bugs, anything can
break and the packaging is not proper. KDE Plasma 5 or Ubuntu SDK (Qt
Creator) don't work yet. Unity 8 on phone or desktop should be
testable.

At this point it's too early to talk about shipping Qt 5.5, and I
don't decide that. But what this partial alpha enables is starting to
file and fix bugs earlier than ever:
https://bugs.launchpad.net/bugs/+bugs?field.tag=qt5.5 - I've filed ~20
so far.

Upstream is 1.5+ months behind the schedule
(https://wiki.qt.io/Qt-5.5-release), so even though this early work is
useful, it's not necessarily going to be a fun ride to the feature
freeze again.

Looking beyond 5.5, I think it'd be nice to stay flexible about the Qt
version. If Qt 5.6 targets November, slips to mid-December like 5.4
did, we get .0 about ready January/February assuming no big
regressions and that's quite late for an LTS. The alternative to 5.6
would be to ship something like Qt 5.5.3 instead of the .0/.1
releases. Qt upstream has considered giving one of their release an
LTS status. If that'd happen to be 5.5, maybe it would make sense
for Ubuntu LTS too. But 5.6 would be a possibility as well, especially
if upstream would target and deliver nearer to October.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: [ubuntu-studio-devel] Feature Spec Discussion: Desktop Agnostic

[Timo Jyrinki]

2015-04-27 3:34 GMT+03:00 Len Ovens l...@ovenwerks.net:
 The link above gives less info than needed. after doing some more
 research... I think it will depend on the dev. It is still possible to lean
 on the system libs and not include them. I expect packages from debian will
 continue to do so. I think ubuntu is trying to apeal to comercial
 develoment. Make it easy and acceptable for people to release blobs for
 money. I do not know how this works with GPL, with debs, all packages have a
 SRC package... I do not see how someone would DL a src package of this or in
 what form.

I don't think commercial development has anything to do with this,
Ubuntu Software Center already has had commercial apps/games on sale
for years, and the process in there supported everything needed for
such. Snappy is more about for example error-free upgrades (think full
Ubuntu version upgrade without any .deb unpacking, setting up,
configuring, fixing dependencies), smaller downloads, ability to
better offer real application updates to LTS releases like latest
LibreOffice (or Ardour etc for that matter..) since the current model
doesn't really support that. PPA:s also are not needed anymore with
Snappy.

There are some QA at:
http://www.webupd8.org/2015/04/ubuntu-desktop-to-eventually-switch-to.html

It's also possible to use normal Ubuntu via .deb based upgrades, even
if the default image would be Snappy based at some point. Since Ubuntu
still actually consists of Debian + Ubuntu's modifications, it's just
whether the .deb:s will be exposed to users or if the .deb:s are only
used by distro developers to build images.

None of the flavors are probably going to even try Snappy before after
16.04 LTS, but maybe afterwards it could be realistic if Ubuntu first
demonstrates what benefits users get from it?

 don't think anyone is ready to dl 2gig security updates every 2 days,

 Further reading says that won't happen, it is more zsync like, just
 replacing the parts of the packages that have changed.

You can basically see Snappy in action on the current Ubuntu phones,
or the former iteration of it (Click) - base image updates that are
pretty small and contain all the latest fixes to all kernel+libraries,
plus applications that don't break since the system image / defined
framework provides the compatibility needed by apps by definition.
This part is a bit like Android, since apps don't break as
uncontrollably.

-Timo

-- 
ubuntu-studio-devel mailing list
ubuntu-studio-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel

Re: [Ubuntu-phone] Qt 5.4 QVariant change and module news

[Timo Jyrinki]

On Wed, Feb 18, 2015 at 8:51 PM, Bogdan Cuza bogdan.c...@hotmail.com wrote:
 I have just received a vivid-proposed image and noticed that U1db no longer
 loads the defaults (which was probably a processed QVariantMap from JS). Did
 this change break it/could it break even more components?

Indeed this was a QVariant related problem and a fix is landing now.

There might be some other similar bugs lurking still, even though no
actual problems related to it have been spotted. I'll adjust the
landing test plan to include more non-required rebuilds in addition to
current 26 Qt private headers using packages. This would also have
been caught if eg u1db-qt was using autopkgtests to run its unit
tests. I think there were about 20 autopkgtest suites successfully run
against the new Qt at the time of landing. Increasing the amount of
autopkgtests would be useful, but so is catching these problems even
earlier, which is why the non-required rebuilds would help.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Qt 5.4 QVariant change and module news

[Timo Jyrinki]

Qt 5.4 is now in vivid, powering Ubuntu Phone starting with devel
image #102, Kubuntu's Plasma 5 and also the Unity 8 desktop preview.
No app changes are generally needed, but there is a change in QVariant
that may be relevant to C++ using apps.

== QVariant change in Qt 5.4 ==

QVariant has changed a bit in Qt 5.4, although in a way that fixing
its use does not affect negatively use in Qt 5.3 either.

Quote: ”When a JavaScript object/array is passed to C++ through a
QVariant, the engine no longer immediately converts the object
recursively into a QVariantMap or QVariantList but instead stores a
QJSValue in the QVariant. This prevents a loss of data when the JS
object contains non-primitive types such as function objects for
example. Code that expects the variant type to be exactly
QVariant::Map or QVariant::List may need to be adapted. Registered
conversion functions however ensure that code that merely calls
toMap() or toList() continues to work.” [QTBUG-40431]

Examples of fixes:
- Oxide: https://code.launchpad.net/~aacid/oxide/qt54_variantjs/+merge/248765
- Ubuntu UI Toolkit:
https://code.launchpad.net/~aacid/ubuntu-ui-toolkit/statesaver_54/+merge/247986

Another noted small difference:
”In Qt 5.4 an alias to a null variable is correctly null instead of
undefined like it was in Qt 5.3”

== Status of less supported Qt modules ==

It's useful to remember that not all of Qt we have in Ubuntu is
supported by the Qt Project:

- qtlocation: Upstream releases this as part of their releases, but
they emphasize that Qt Positioning is API stable while Qt Location
might change. Still, this should be pretty safe to use and will
probably be declared stable soon.
- qt3d: The original module has not been developed for two years. A
complete rewrite has been in development during that time. Qt 3D 2.0
will be available as a preview with Qt 5.5.
- qtsystems: Semi-active development. Qt 5.4 absorbed QStorageInfo to
be official part of qtbase, but it is a bit different and oddly does
not have QML support at this point. We include the SystemInfo module
on the phone images, but currently not the Publish Subscribe or
Service Framework modules from the same source.
- qtpim: Semi-active development, but can be considered supported
since our Renato is one of the upstream developers. Ubuntu uses the
Contacts/Organizer extensively in phone/scheduling apps.
- qtfeedback: Stable in the sense that there is no development
happening. It's also a very small module. If you are interested in the
module, consider taking an active role in the Qt Project where
contributions would surely be welcome.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: [ubuntu-release] Vivid Vervet (15.04) now open for business

[Timo Jyrinki]

2014-10-27 16:45 GMT+02:00 Jonathan Riddell j...@jriddell.org:
  - A switch from C to C.UTF-8 on the builders and the default
locale when none is selected in the installer.
  - A move to glibc 2.20 or 2.21, timing dependent.
  - Steady progress on the systemd transition, with a goal to
switch over in the first half of the cycle.

 Out of interest where is stuff like this decided these days?

Those have probably been discussed within the Ubuntu Foundations team
(on #ubuntu-devel?), but the plans can be discussed in more detail
during the Ubuntu Online Summit [1] and its blueprints, announced in
July [2]. The Ubuntu Development / Platform track includes such
topics.

[1] http://summit.ubuntu.com/uos-1411/
[2] https://lists.ubuntu.com/archives/community-announce/2014-July/20.html

 The release schedule for Vivid seems to be missing, anyone working on adding 
 it?  Shall I have a look at it?

Maybe you (release team) should have a meeting about it or a session
at UOS. A draft by a member would probably be helpful, especially
regarding derivatives' milestone release wishes.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Patch pilot report: 2014-09-25

[Timo Jyrinki]

2014-09-25 19:42 GMT+03:00 Robie Basak robie.ba...@ubuntu.com:
 I can't close the MPs. Can somebody do that, please?

The MP:s lack Abandoned status option, and removing the MP would
lose the discussion history over it. So I did what I've seen others
do, ie set as Merged to mean nothing to do anymore, even if it
wouldn't be really merged.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Patch pilot report: 2014-09-25

[Timo Jyrinki]

2014-09-29 10:27 GMT+03:00 Martin Pitt martin.p...@ubuntu.com:
 The MP:s lack Abandoned status option, and removing the MP would
 lose the discussion history over it. So I did what I've seen others
 do, ie set as Merged to mean nothing to do anymore, even if it
 wouldn't be really merged.

 What's wrong with Rejected?

At least for me only the following options are available:
- Work in progress
- Needs review
- Merged

For example at 
https://code.launchpad.net/~filip-sohajek/ubuntu/utopic/silversearcher-ag/fix/+merge/232369

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Patch pilot report: 2014-09-29

[Timo Jyrinki]

73 items at beginning, 51 at the end (but also thanks to at least two
other active pilots). My first pilot shift as MOTU.

Needs community testing:
* [SRU] lsdvd (invalid output for acidrip) http://pad.lv/1310049
  - Readied an upload fro 14.04 LTS and tested the change, but was
unable to get acidrip really functional so asking for community
testing.

Done:
* lp:~saltmakrell/ubuntu/trusty/kicad/missing-modules missing-modules
  - Sponsored into 14.04 LTS queue after testing that the errors are gone.
* Sync pushpin 1.0.0-3 (universe) from Debian unstable (main)
  - Bugfixes only, synced.
* Please sync grace 1:5.1.24-2 from Debian unstable main
  - Bugfixes and documentation updates only, synced.
* lp:~reversiblean/ubuntu/utopic/ubuntu-restricted-extras/bug-865379
  - Fixed a bit and uploaded. Only package description changes.
* lp:~reversiblean/ubuntu/utopic/videoporama/fix-for-1345565
  - Uploaded.

(+ already last week
lp:~israeldahl/ubuntu/trusty/lmms/lmms_1.0.0+stable - in
trusty-proposed now and verification done)

+ some misc commenting and cleaning.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: What's the right way to get qmlscene to work out of the box?

[Timo Jyrinki]

2014-06-23 17:22 GMT+03:00 Sebastien Bacher seb...@ubuntu.com:
 I've discussed the topic a bit on IRC and the advices/replies include:
 - qmlscene is a dev tools and shouldn't be used by those applications
 - you should change the environment/export a variable to make qt5 the
 default (seems to be what ubuntu-touch does)
 - install qt5-default ... but that would bring some extra 50MB on the iso

The discussion has come up earlier, but we've stuck with qmlscene
because we haven't had big enough problems to get eg. SDK team to
write our own wrapper. For a brief moment we had 'qmlrunner' symlink
idea but that was backed out when the problem at hand was solved in
another way.

As mentioned in the Dmitry's codereview link, upstream thinks Qt
Creator is the only user tool and and the rest are development tools.
Qtchooser is meant for using the development tools, which don't want
automatic fallbacks. You don't want to accidentally use Qt 5's qmake
when you have selected to use Qt 4. KDE is having a similar problem
since they use qdbus. Even though Qt 4 is the default if no selection
has been made, an user might install qt5-default or set an environment
variable, and so KDE has resorted to workarounds too.

A direction worth going towards in Ubuntu at least would be to get rid
of hard-coded 'qmlscene' calls. If eg. ubuntu-app-launch would be the
only place that launches apps, and it would know how to launch
QML-only apps (file.qml) by using a launcher instead of calling an
executable directly, then we could more easily switch between
launchers. It doesn't solve eg. the SDK Gallery problem on the desktop
before we have that alternative launcher, but it's a necessary step to
consider switching away to our own wrapper.

Both 'qmlscene' and its successor 'qml' are around 400 lines of code,
so they should be simple to replace with another one. Unity8 already
has 'uqmlscene' modified from qmlscene in their tests, maybe that's a
good name :) There shouldn't be a problem in using two wrappers in
parallel for some time, although I'm currently running my mako with
'qml' (I overwrote 'qmlscene' with it) and there seem to be small
layout differences in eg. clock. So apparently the scene is set up in
a bit different way from qmlscene.

We could survive a bit longer by distro patching qtchooser (another
workaround), but I feel that eventually we'd like to have more control
on the QML app startups and have a simple binary location instead of
the workarounds? For example in case any startup time optimizations
would require hooking into it. There are some old requests too, I used
the 'qmlscene' tag:
https://bugs.launchpad.net/bugs/+bugs?field.tag=qmlscene

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: What's the right way to get qmlscene to work out of the box?

[Timo Jyrinki]

2014-08-13 9:03 GMT+03:00 Timo Jyrinki timo.jyri...@gmail.com:
 We could survive a bit longer by distro patching qtchooser (another
 workaround), but I feel that eventually we'd like to have more control
 on the QML app startups

For the Ubuntu specific new wrapper, discussion can continue on ubuntu-phone.

But we may apparently want a quicker solution for eg. Unity 8 on
desktop, so I'd not throw distro patching the qtchooser option away
yet. Let's move that discussion to eg. this merge proposal:
https://code.launchpad.net/~timo-jyrinki/kubuntu-packaging/qtchooser_qmlscene_fallback/+merge/230595

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Qt 5.3 landing update June 16

[Timo Jyrinki]

2014-06-18 16:46 GMT+03:00 Timo Jyrinki timo.jyri...@gmail.com:
 earliest time of landing would be later today in the more western
 timezones, or tomorrow morning if we get a green light from the people
 who are waking up now and testing/fixing things.

Qt 5.3.0 is now in utopic-proposed. Eight Qt modules auto-synced from
Debian were already waiting there for the base modules, and they
should start building shortly. Three more modules were made syncable
in the process of packaging 5.3.

Everything that was on the blocker list of the landing was resolved or
whitelisted today. For example the emulator worked fine after today's
additional fixes.

 - Gallery app header broken (artmello, t1mp/SDK team):
 https://bugs.launchpad.net/ubuntu/+source/gallery-app/+bug/1331382 -
 header works in all other apps

Gallery app new click package is not yet in the store. The Apps team
is still working on improving it and its autopilot tests.

 (- Upload the new music app (popey) - can be done after publishing the
 Qt 5.3 landing silo)

Music app on the other hand is in the store already and better than
ever, thanks to the awesome work by Victor, Nicholas, Andrew.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Qt 5.3 landing update June 16

[Timo Jyrinki]

2014-06-16 17:28 GMT+03:00 Timo Jyrinki timo.jyri...@gmail.com:
 The landing this week seems doable, let's do our best to fix the
 remaining blockers. Many bugs are being pinpointed to specific root
 problems, but we need to get rid of the remaining blockers.

Update for today below, almost everything fixed and tested. The
earliest time of landing would be later today in the more western
timezones, or tomorrow morning if we get a green light from the people
who are waking up now and testing/fixing things.

 - UITK toolbar empty (SDK team / t1mp)

Fixed by Tim, works now in all apps again.

 - Land new Oxide (chrisccoulson, oSoMon)

Fixed, works great.

 - Qt Creator 3.1.1 Ubuntu plugins (zbenjamin, me)

Fixed.

 - Swipe related UITK issues (SDK team / elopio)

Not Qt 5.3 specific.

 - Music app switch to mediascanner2 (Music app devs)

We've confirmed the mediascanner2 branch of music app works fine with
Qt 5.3 and it can a .click package be uploaded at any point  -
https://bugs.launchpad.net/music-app/+bug/1324105

 - Qt gles packages needed for emulator (rsalveti, me) -
 packaging/dependency fixes to qtbase-gles - ETA ?

Done, not tested yet.

 - Unity8 crashers (Unity 8 team) - memory corruption in Qt? This is
 the most worrisome blocker at the moment.

Fixed! A nice catch by Albert being upstreamed at
https://codereview.qt-project.org/#/c/87633/

** Remaining/new blockers **

- Gallery app header broken (artmello, t1mp/SDK team):
https://bugs.launchpad.net/ubuntu/+source/gallery-app/+bug/1331382 -
header works in all other apps

- Framework 14.10-dev2 to landing-005 PPA (lool)

- Qt base unit tests situation get ack or nack from QA (Chris Cagnon /
robotfuel): 
https://bugs.launchpad.net/ubuntu/+source/qtbase-opensource-src/+bug/1324000

- Confirm emulator working with the Qt gles packages (rsalveti)

(- Upload the new music app (popey) - can be done after publishing the
Qt 5.3 landing silo)

The best place with all the relevant links is probably still the
spreadsheet at 
https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0AjuCdq68GSyVdFI4QzNQdWpfME5aMEV2VXo0cUpOMkE#gid=20

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Qt 5.3 landing update June 16

[Timo Jyrinki]

The landing this week seems doable, let's do our best to fix the
remaining blockers. Many bugs are being pinpointed to specific root
problems, but we need to get rid of the remaining blockers.

Done:
- Calendar app AP failures are also on normal images - qt5.3 tag removed
- telephony-service fix confirmed to fix all address book app
problems, even though it still needs to be properly landed (boiko)
- Qt Creator plugins have functional branches
- popey ran an automated startup of all store apps, taking screenshots
and grabbing crashes
- As mentioned, several bugs combined to be duplicates of the root problems

Todo blockers (all open bugs http://is.gd/gZFEqm ):
- UITK toolbar empty (SDK team / t1mp) - Gallery, notes and dropping
letter AP failures are related to same root problem - ETA ?
- Land new Oxide (chrisccoulson, oSoMon) - required for truly
functional web browser with compositing enabled also on Qt 5.3 - ETA
this week, the sooner the better
- Qt Creator 3.1.1 Ubuntu plugins (zbenjamin, me) - small packaging
glitches - ETA tomorrow
- Swipe related UITK issues (SDK team / elopio) - Calculator,
messaging AP failures related to swipe have a common root problem, and
the specific swipe behavior changing. There's a related UITK fix being
released via landing-006 silo - ETA tomorrow
- Music app switch to mediascanner2 (Music app devs) - ...after
checking the new branch works as good as with Qt 5.2, since it's long
due and qtgrilo crashes on QT 5.3 - ETA this week, the sooner the
better
- Qt gles packages needed for emulator (rsalveti, me) -
packaging/dependency fixes to qtbase-gles - ETA ?
- Unity8 crashers (Unity 8 team) - memory corruption in Qt? This is
the most worrisome blocker at the moment.

Store apps testing:
Testing seems good so far with ~99% working with current knowledge.
All store apps have been started in an automated way by, with
screenshots and crashes grabbed. Additionally some manual testing is
being done all the time. The following six apps are known to have some
sort of problem with Qt 5.3, some of which may be related to us
missing the new Oxide (if webapp):
com.popey.nationalrail
com.ubuntu.developer.ken-vandine.pathwind_pathwind
com.ubuntu.developer.mzanetti.tagger_ubuntu-sso
com.ubuntu.developer.agdpsoftware.inkbunnyapp_inkbunnyapp_0.5.6
com.ubuntu.developer.rick-rickspencer3.franglish_Franglish_0.4
com.ubuntu.developer.gcollura.saucybacon_saucybacon_1.0.15

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Qt 5.3 landing update June 13

[Timo Jyrinki]

2014-06-13 20:20 GMT+03:00 Scott Kitterman ubu...@kitterman.com:
 Should I conclude from this that it will or will not get in on Monday as
 planned?

The June 6 update mentioned several devs are at the Qt contributor
summit next week, so the earliest we will land is still week of the
16th.

I'd say tomorrow is unlikely, but some day later in the week will
hopefully be the landing day. It does require getting fixes from many
teams, but there's nothing particularly scary in the bugs at the
moment. Crashers are bad as always, but so far all of the previous
ones have had a clear reason. There's possibly a need for a backported
fix or two from upstream.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Qt 5.3 landing update June 13

[Timo Jyrinki]

Good progress again. Looking at last week's actions: camera works in
addition to video playback, automated autopilot testing is now
functional, bugs for autopilot tests have been filed, store apps seem
quite fine initially, qtbase failing unit tests have been investigated
a bit, packaging and syncs with Debian are mostly done.

Still todo from last week's list: Qt gles emulator packages (I need
some help from Ricardo), framework bump (Pat, lool), bug fixes (see
below).

The landing silo [1] is now nearly ready, with some cleaning that
needs to be done at the time of landing as documented in the CI Train
[2] comments because utopic-proposed already has many Qt 5.3.0
packages from Debian that we will use instead. The emulator (Qt gles)
and Qt Creator plugin packages are still missing.

On the bugs front, we have 22 bugs open [3] out of 52 filed.
Telephony-service has now a fix that should fix the address book
problems. That would leave us is with:
- 1-3 autopilot test failures in UI Toolkit, gallery-app,
calendar-app, calculator-app
- Unity 8 crasher that is possibly an upstream bug in qtbase (first
backtrace just gotten)
- music-app (/qtgrilo) crashes on startup - probably the best would be
to make the anyway needed switch away from qtgrilo
- webbrowser-app is not working optimally since compositing is
disabled in Qt 5.3, but Oxide Qt already has a fix that should be
landing soon

Otherwise the landing - with current knowledge - is in quite good
shape. Only 1 store app is confirmed to have a problem on startup at
the moment, while all others should at least start and render (popey
will rerun this semi-automated test). More testing and bugs are
certainly welcome! Read the instructions page [4] for how to enable Qt
5.3. [5] has more of the todo list explained.

[1] https://launchpad.net/~ci-train-ppa-service/+archive/landing-005
[2] https://wiki.ubuntu.com/citrain
[3] http://is.gd/gZFEqm
[4] https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-beta2
[5] 
https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0AjuCdq68GSyVdFI4QzNQdWpfME5aMEV2VXo0cUpOMkE#gid=20

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Qt 5.3 landing update June 4

[Timo Jyrinki]

2014-06-04 17:52 GMT+03:00 Scott Kitterman ubu...@kitterman.com:
 The major thing this misses is merging from Debian.  There are quite good Qt5
 5.3.0 packages in Debian Experimental and what is uploaded to the archive
 should be based on that work.

The main modules qtbase and qtdeclarative were already merged with
Debian's Qt 5.3.0 packaging, including the moving of include files in
all Qt modules. Some more merging will be needed at least in case if
Debian changed something between 5.2.1 and 5.3.0. I'll do direct syncs
of the suitable modules to the PPA as well. Traditionally the most
meaningful packaging changes happen in base/declarative.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Point of reviews

[Timo Jyrinki]

2014-05-23 14:41 GMT+02:00 Scott Kitterman ubu...@kitterman.com:
 If you look at this merge proposal, it was disapproved with a suggestion that
 it was premature.  Despite that, it got released and into the archive anyway.

 So what's the point of review?

I'm not sure if you noticed the timeline, but it got released before
the reviews. Had I read negative reviews before I hit the publish
button in CI Train, I wouldn't have released it.

I didn't wait long with this trivial typo fix since I haven't been
expecting reviews (I noticed a change earlier this week when I was
preparing qtpim). I've largely worked alone on the Ubuntu side with
some awesome help from other developers working on Ubuntu Phone and
mitya57 regarding Qt 5 and the syncing with Debian.

Just let me know eg. on IRC if you want to start working on anything
related to Qt 5.3.0 packaging so that I can double-check everything I
have currently brewing is committed to some bzr branch. I first did a
quick but ugly PPA build
(https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-beta2) and
I'm now slowly working on a tests enabled, symbols updated versions in
parallel. That will also need to be readjusted later at minimum to
sync with Debian.

The final Qt 5.3.0 landing should also be prepared by doing archive
quality uploads to a CI Train silo, so that it can be fully tested and
then published as a whole. As Ubuntu Phone is not just ramping up but
doing daily releases, it's important not to disturb this process. The
silos work neatly in this regard, since they also allow syncing
packages from Debian to the PPA from where the whole set of tested
components is then synced to archives.

 I'm starting to think Canonical's Qt5 stack should go in it's own namespace
 separate from the one used by Debian/Kubuntu as was discussed at the last
 vUDS.  I don't sense much interest in collaboration.

The Qt5 was originally put to under ~kubuntu-packagers even though it
was only used by Ubuntu so that it could be worked on in co-operation
in the long term more easily. Co-operation has in my opinion worked
nicely with anyone who has been willing to contribute to the packaging
work. Obviously with Ubuntu as the almost sole user of Qt 5 so far it
has been largely people working on Ubuntu Phone, but that's changing
now.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Upgrade iso image

[Timo Jyrinki]

2014-04-29 3:56 GMT+03:00 Phillip Susi ps...@ubuntu.com:

 With the removal of the dvd iso, I feel a void has been left in the
 upgrade process.  I understand the desire to avoid the burden of
 testing the bootable install dvd image, but wouldn't it be nice to
 replace it with an upgrade image that just contains packages you can
 use for an upgrade?


I'm not sure about the details, but I did notice that there's a new option
upgrade Ubuntu when booting from 14.04 LTS image on a machine that has
12.04 LTS. Maybe that function fills in the void?

Secondly, the new official image is already DVD image size wise, since
it's around 1GB while the old DVD image was 1.6GB (checked 12.04 LTS).
There is some size delta, but was it so that the old DVD image also
duplicated content because there was both the live squashfs image and deb
pool directory separately?

So it might be the new 1GB image works quite well for upgrades, at least to
a similar extent that the old DVD image did.

There'd be more use for the old old (until 11.04) DVD images that were
4GB+, ie something that contains more packages, but the DVD image of late
was quite close to the new CD image.

-Timo
-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Patch pilot report [2014-04-10]

[Timo Jyrinki]

This was quite on-and-off piloting turn, but I'd like to hilight LMMS
needing someone to look at it so it's worth sending this e-mail
(Ubuntu Studio folks would like to have it in very much):

https://bugs.launchpad.net/lmms/+bug/1291675 - LMMS 1.0, fixed various
packaging issues, checked upstream sources (git+bzr), built in a PPA
and got (even more) test results from Ubuntu Studio team, seems ready
for upload... except for that there are still two new lintian errors
(ships now shlibs file while there are no actual shared libs, and uses
png for icon instead of xpm).

https://bugs.launchpad.net/ubuntu/+source/ubuntustudio-lightdm-theme/+bug/1304214
- Got an UIFe ack, built in a PPA, tested locally and got sponsored

https://code.launchpad.net/~noskcaj/ubuntu/trusty/xfce4-xkb-plugin/lp-733563/+merge/214365
- built in a PPA but then again it's a feature so I believe it'd need
FFe instead.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

[ubuntu-studio-devel] Updated pitivi 0.93 for Ubuntu 14.04 LTS?

[Timo Jyrinki]

Hi,

The release team asked me to ask for Ubuntu Studio's comment to the
question of whether to upgrade pitivi for 14.04 LTS, at:

https://bugs.launchpad.net/ubuntu/+source/pitivi/+bug/1253009

Could someone comment there? I'm not Ubuntu Studio developer, but my
personal experience is that 0.15 is nowadays next to unusable while
the 0.93 I'm offering in a PPA as mentioned on the bug report is at
least usable, and also recommended by upstream over 0.15.

The only thing aside from sync needed to be done is that it'd need to
be moved from main to universe because of the new dependencies. But
the dependencies themselves are all there.

To give the PPA a whirl on 14.04:
---
sudo apt-add-repository ppa:timo-jyrinki/pitivi
sudo apt-get update
sudo apt-get install pitivi
---

-Timo

-- 
ubuntu-studio-devel mailing list
ubuntu-studio-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel

Re: Qt 5.2.1 final push

[Timo Jyrinki]

2014-02-27 8:32 GMT+02:00 Timo Jyrinki timo.jyri...@canonical.com:
 Qt 5.2.1 should be soon ready for landing to Ubuntu 14.04, the target
 being March 6th if the FFe is accepted. The previous update two weeks
 ago was at https://lists.launchpad.net/ubuntu-phone/msg06396.html

We didn't hit that target in the quest for perfect landing (as
discussed during vUDS as well...), but it's now done! We have Qt 5.2.1
in Ubuntu 14.04 LTS, thanks to the team effort of multitude of people
yesterday, throughout the night and timezones until today.

Please find Didier's summary of the next steps to soon have Ubuntu
images with Qt 5.2:
https://lists.launchpad.net/ubuntu-phone/msg06943.html

Thanks to all that helped me in preparing the final landing PPA of
~120 source packages over the last two weeks, which resulted in 8GB
of binary debs for six architectures. The next, albeit smaller 'CI
Train' landings are visible at https://wiki.ubuntu.com/citrain

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Qt 5.2.1 final push

[Timo Jyrinki]

Hi all Qt lovers,

Qt 5.2.1 should be soon ready for landing to Ubuntu 14.04, the target
being March 6th if the FFe is accepted. The previous update two weeks
ago was at https://lists.launchpad.net/ubuntu-phone/msg06396.html

The FFe is open at http://pad.lv/1278329. It has the latest status
updates in more detail. The landing is late because it was wanted it
wouldn't block other Ubuntu work for uncertain amount of time, and
because it was found that Qt 5.2.1 + additional patches upstreamed by
Ubuntu developers is the minimum that can be considered to be free
enough of regressions. Qt 5.2 saw a complete rewrite of the QML script
engine from Google's V8 to Qt's own V4 among else, so it's not a big
surprise there were hurdles since Ubuntu  Unity 8 are probably one of
the heaviest existing Qt5 users.

The biggest remaining blocker was the lack of functional audio/video
playback and camera. Fixes for both should be landing today. After
that the most important work is doing QA to ensure that there are no
such regressions that could be considered absolute blockers. App store
compatibility initially seems good, but it will need a closer
inspection.

Thanks to everyone involved in fixing the numerous issues in the stack
and apps, and welcome to join for the remaining testing / bug filing /
bug fixing at https://bugs.launchpad.net/bugs/+bugs?field.tag=qt5.2

For those able to join the push, upgrade instructions are available at
https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-beta2 Note!
You will need to use the ppa-purge tool eventually, since it will not
be possible to upgrade from the PPA to the eventual archive versions.

On the desktop, Qt Creator 3.0.1 and the Ubuntu SDK plugin for it are
now available.

-Timo

ps. just in: audio/video/camera all work! I'll be copying the fixes to
the main PPA after I've ensured upgrades still work smoothly

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Patch pilot report 2014-02-13

[Timo Jyrinki]

My first Patch Pilot report! I read docs and thought about how I'll approach it.

Done:
* metacity http://pad.lv/907640 alt-grave fix. Commented, built,
tested and did a packaging merge proposal out of the debdiff. Uploaded
by didrocks, thanks!
* xubuntu-docs http://pad.lv/1251332 14.04.0 release. Checked with
#xubuntu-devel that yes they'd like sponsoring support. Tested, made a
packaging merge proposal and also proposed a small pedantic change to
trunk. Uploaded by dholbach, thanks!
* atk1.0 http://pad.lv/1277460. Prepared and tweaked a branch from the
debdiff for an upload and built in a PPA, but I ran out of time for
giving it testing.

During my turn I also noticed I'm not alone with mitya57 handling evas
sync, cjwatson telepathy-logger sync and rbasak uploading nginx saucy
fix. Nice! Thanks to them too.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Qt5 plans for trusty

[Timo Jyrinki]

2014/1/22 Aeron Farax aeron2...@yahoo.ee:
 QT 5.0 is like beta in 5 series, apps that use QT5 and most of them move
 to v5.2

 It would be wise to include 5.2 for sake of LTS

In progress! Everyone hopes to move to Qt 5.2 as quickly as possible.
Incidentally I just posted an update from Ubuntu Touch point of view
at:

https://lists.launchpad.net/ubuntu-phone/msg06049.html

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: qreal change in Qt 5.2

[Timo Jyrinki]

2013/12/4 Dmitry Shachnev mity...@ubuntu.com:
 Also, IIRC, nothing was decided about the SONAME stuff (and I would really 
 want
 if we got synchronized with Debian on (not)doing the bump).

I'd go for not bumping SONAME and doing all rebuilds at once. Debian
should lead the way, and since they seem to do either this or not
switch from float on arm, this seems a lot better option for the
future. According to the bug report Fedora already switched without a
bump.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: On cross-compiling qml extensions, qt packaging needs changing

[Timo Jyrinki]

On Wed, Sep 4, 2013 at 1:34 PM, Dmitrijs Ledkovs
dmitrij.led...@ubuntu.com wrote:
 Even a more simple case of:
 $ apt-get install qtbase5-dev qtbase5-dev:armhf

 Doesn't install because of GL somewhere down the stack:
  qtbase5-dev : Depends: libgl1-mesa-dev but it is not going to be installed or
 libgl-dev
Depends: libglu1-mesa-dev but it is not going to be installed 
 or
 libglu-dev
 E: Unable to correct problems, you have h

It's actually worse, since we shouldn't have qtbase5-dev multi-arch:d
either, and Debian doesn't accept it. The contents of the dev packages
differ from architecture to another, so it's a policy violation. I
haven't removed the multi-arching of it yet since it wasn't added by
me, and I'm not sure if it was required for the arm64 support added at
the same time or just something nice to have.

Currently I don't see it possible to have all of Qt (or ubuntu-sdk) to
be multi-arch:d, or at least I wouldn't know how to go forward. I also
agree we shouldn't deviate from Debian. But if there's a policy
compliant way forward it should be good enough for them too.

So in the short term (Ubuntu 13.10) I think your 'ugly but works'
solution you've started upon is needed. In the longer term it's a goal
worth having to improving the multi-arching of Qt.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Call for Testing: Mir multi-monitor

[Timo Jyrinki]

2013/8/27 Robert Park robert.p...@canonical.com:
 I also had libhybris installed, and removing it fixed xmir.

Yep, I hit that as well 2.5 weeks ago when I started using xmir:
https://bugs.launchpad.net/unity-system-compositor/+bug/1210798

Aside from the lack of proper multi-monitor support and that bug I
haven't had any other problems or seen much difference. It has worked
as reliably and smoothly as before. I'm actually using external
monitor, but the Mir's default mirror mode has been enough while
waiting for this new support which I'll try out now.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Side-by-side installation of Qt5 and Qt4

[Timo Jyrinki]

Hi,

Yes, Qt4 and Qt5 are fully co-installable in Ubuntu 13.04 (and 12.04 LTS +
12.10 with the help of the PPA). You should install either qt4-default or
qt5-default package to select which tools are given you by default, ie.
when you type eg. qmake without any parameters. You can then switch between
the toolsets/libraries runtime with the QT_SELECT variable. This also
applies to packaging, so if you are doing a .deb package, you can either
add a build-dependency on qt5-default or qt4-default, or use export
QT_SELECT in debian/rules (and remember to have build dependency to
qtchooser in addition to other Qt dependencies). See 'man qtchooser' for
more information.

The only additional hindrance is, that upstream KDE hard-coded the Qt4
binary directory in PATH (see LP: #1176686 for more information), which
prepends even /usr/bin which has qtchooser et cetera. This temporary
measure to make sure they always get Qt4's qdbus will most certainly go
away later, but until then you need to check your PATH variable in your
development environment when using KDE.

-Timo



2013/5/14 The Spencers spencers1...@gmail.com

 Hello,

 I'm working on an application written using KDE 4.10 and Qt4.8. I'm mostly
 using KDE for the just UI, but since I am using Ubuntu and targeting Ubuntu
 (not KDE or Kubuntu) I am considering switching to Qt5 and the Ubuntu SDK
 once it becomes more stable and usable on the desktop, not just touch
 devices.

 Because of that, I'd like to try experimenting with Qt5 and the Ubuntu
 SDK. However, is it possible to install Qt5 in addition to Qt4 and the KDE
 development stuff? I assume it would be fine for running a pre-built
 application, and all the normal KDE apps and stuff would be fine, but will
 the combination work correctly when I try to build my KDE app? I'm asking
 this because I tried installing Unity Next after reading about it on
 omgubuntu.co.uk, and was unable to compile my KDE app due to it finding
 Qt5 instead of Qt4.

 While researching this, I came across a note on the canonical-qt5-edgers
 PPA [1] that said:

 1. Qt4 and Qt5 are co-installable with the migration to the new qtchooser
 tool. The previous Debian/Ubuntu approach relied on renaming binaries and
 using alternatives. When packaging against Qt5, set environment variable
 QT_SELECT=qt5 or include a package dependency to qt5-default in the
 packaging. See man qtchooser for more information.

 So if I set QT_SELECT to qt4 when I want to build my KDE app, and change
 it back to qt5 when tinkering with the Ubuntu SDK, will this be sufficent?

 [1] https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-proper

 Thank you,
 Michael Spencer

 --
 Ubuntu-devel-discuss mailing list
 Ubuntu-devel-discuss@lists.ubuntu.com
 Modify settings or unsubscribe at:
 https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss


-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: This missing kernel headers on our latest stable release madness...

[Timo Jyrinki]

2013/2/22 Scott Ritchie scottritc...@ubuntu.com:
 I've been absolutely flooded with informal reports over a period of several
 months now of 12.10 being still broken with regards to proprietary drivers.

 Reports like this are typical, especially after the influx of steam users:
 Installed ubuntu + proprietary amd drivers, got no unity at 800x600 on next
 reboot and uninstalled.

 The proximate cause is a combination of
 https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers-updates/+bug/1068341
 and https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1070427

I'm not 100% sure, but I believe this bug report
https://bugs.launchpad.net/ubuntu/+source/jockey/+bug/1123107 could be
the one that has the fix for all the bugs. The fix is now in
precise-proposed, with quantal being In Progress. CC:ing Alberto who
has been working on the jockey/ubuntu-drivers-common/nvidia-common
updates.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: This missing kernel headers on our latest stable release madness...

[Timo Jyrinki]

2013/2/22 Scott Ritchie scottritc...@ubuntu.com:
 I'm not sure what the underlying fix should be, but it is making me question
 if there's some sort of larger process issue here because we've managed to
 drop this on the floor for so long.

A good question. It might be that the proprietary drivers haven't been
pushed in QA/Testing teams high enough, compared to how much users use
them and how jockey suggests installing them. The problem is furthered
by the history of non-Ubuntu bugs/problems with at least fglrx.

Right now there has been the case of AMD dropping support for HD
2000-4000 series in the newer drivers, and only the newer drivers
support newer kernels. I'm not sure what's the exact situation for
12.04.2 (or 12.10) - do they suggest an uninstallable fglrx driver for
HD 2000 - 4000 users or not. I think the Modaliases of the package's
control file is used, so the question is whether that is properly
stripped of the now unsupported series. If anyone wants to tinker...

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: This missing kernel headers on our latest stable release madness...

[Timo Jyrinki]

2013/2/22 Scott Ritchie scottritc...@ubuntu.com:
 I've been absolutely flooded with informal reports over a period of several
 months now of 12.10 being still broken with regards to proprietary drivers.

 Reports like this are typical, especially after the influx of steam users:
 Installed ubuntu + proprietary amd drivers, got no unity at 800x600 on next
 reboot and uninstalled.

 The proximate cause is a combination of
 https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers-updates/+bug/1068341
 and https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1070427

I'm not 100% sure, but I believe this bug report
https://bugs.launchpad.net/ubuntu/+source/jockey/+bug/1123107 could be
the one that has the fix for all the bugs. The fix is now in
precise-proposed, with quantal being In Progress. CC:ing Alberto who
has been working on the jockey/ubuntu-drivers-common/nvidia-common
updates.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: This missing kernel headers on our latest stable release madness...

[Timo Jyrinki]

2013/2/22 Scott Ritchie scottritc...@ubuntu.com:
 I'm not sure what the underlying fix should be, but it is making me question
 if there's some sort of larger process issue here because we've managed to
 drop this on the floor for so long.

A good question. It might be that the proprietary drivers haven't been
pushed in QA/Testing teams high enough, compared to how much users use
them and how jockey suggests installing them. The problem is furthered
by the history of non-Ubuntu bugs/problems with at least fglrx.

Right now there has been the case of AMD dropping support for HD
2000-4000 series in the newer drivers, and only the newer drivers
support newer kernels. I'm not sure what's the exact situation for
12.04.2 (or 12.10) - do they suggest an uninstallable fglrx driver for
HD 2000 - 4000 users or not. I think the Modaliases of the package's
control file is used, so the question is whether that is properly
stripped of the now unsupported series. If anyone wants to tinker...

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: [Idea] PhoneGap support for Ubuntu

[Timo Jyrinki]

2012/11/6 James Haigh james.r.ha...@gmail.com:
 What? Could you please provide a link?

http://www.markshuttleworth.com/archives/1195 - look at the core of
Ubuntu and review it through a mobile lens. I think PhoneGap would
fit in the core by being essentially also a sort of build/packaging
tool, and having support for Ubuntu in more mobile world build tools
would be nice. Or if nothing else, at least in the 14.04 LTS context
PhoneGap could make sense if not yet during this cycle - but the work
could already begin IMHO.

 - and I really do hope Canonical get it out there before it goes the same
 way, or competitors get there first. I think Canonical should focus on
 Ubuntu for Android before revisiting the ambitions of Ubuntu Mobile.

I think the current plan sounds quite sensible, since improving eg.
memory footprint, battery life etc. will help also desktop users as
well as the potential Ubuntu for Android users.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: [Idea] PhoneGap support for Ubuntu

[Timo Jyrinki]

2012/11/6 Ma Xiaojun damage3...@gmail.com:
 This is inspired by the following post:
 https://mail.gnome.org/archives/desktop-devel-list/2012-November/msg6.html

 Since Ubuntu is heading towards Tablets and mobile phones now.
 It would be very nice if we also join the PhoneGap camp.

This occurred to me is as well when the new mobility focus was
announced a few weeks ago. If you or someone has interest to tinker
with the code, go ahead and add Ubuntu support :) There are indeed
many platforms which try to do the same as PhoneGap, so pick your
choice and add Ubuntu in front of developers' eyes!

In the end I guess it will mostly matter when some of those projects
really start to have developer mass, ie. real applications get
published to app stores from within the framework. In that phase it'd
be cool if a single button push would make it available also in Ubuntu
software center, right? I don't know if some high profile apps already
are published solely via eg. PhoneGap. Wikipedia app by Wikimedia
foundation and BBC Olympics look to be interesting.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: How to install Precise without getting screwed?

[Timo Jyrinki]

2012/4/11 Dane Mutters dmutt...@gmail.com:
 There's somewhat more to it than that.  The major issue (among many other
 issues) is that the new GUIs don't do the things that used to be available
 on the old one (Gnome 2).  Example: I can't add a good system monitor to
 Gnome 3 because the old gnome-system-monitor applet (being an applet at all,
 apparently) is incompatible with Gnome 3.

Not sure about the system monitor applet case (I'm not using one), but
there is at least indicator-multiload now. For me, usually using
development versions and not sticking with LTS, losing the weather
information was unfortunate at the time. Currently I'm personally not
missing anything anymore (although weather info should be a default
feature), and the Unity in 12.04 feels more productive than GNOME 2
ever did. Mostly because of super + (shift + ) numbers, super + (shift
+) alt + arrows and the Dash search features, plus the screen space
(even though I've 1600x900 display). The search features for accessing
recent documents and apps is much nicer than browsing through the
menus, at least after getting used to it.

It did take learning time to become not annoyed with Unity, although
now if I'd start with 12.04 the situation would be more welcome,
because it's that much more stable and faster. I'm not that much
against change, so the biggest irritation for me was all the bugs
previously. Also the simple thing of showing quick help when keeping
Super pressed down helps a lot in learning how to use Unity more
powerfully. I can understand the pain people have gone through if
using Unity since 11.04 (I didn't start to use it back then). I was
annoyed with the lack of application menu for a long time, but finally
nowadays using the search feels natural and fast, plus navigating
menus manually on 10.04 LTS machines feels clunky. Super + A is also
available for an access to a list, but I'd prefer it'd be expanded
without an extra click.

Still, it's not for everyone of course, and is a big change as a
sudden switch kind of thing. I also hadn't even realized the missing
graphical way of doing desktop launchers, so I learned something from
this thread as well. Obviously I haven't used desktop launchers for
anything, since I prefer using terminal anyway for such use cases that
I could possibly do custom launchers. And I have a big .ssh/config,
yes :)

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Drop Gwibber from default install

[Timo Jyrinki]

2012/3/14 Rodney Dawes rodney.da...@canonical.com:
 Can you please clarify what you mean by laggy here? And are
 there bugs open for what you mean by your use of the term?

The smooth scroll bug covers it mostly, but some specifics might use
separate bug reports. At least the following come to mind right away:

- You can see the photos/images being drawn, and they flicker from
time to time when the scrolling stops.
- Keeping arrow down pressed feels laggy, both because of the low
update rate, lack of smooth scrolling and visible lag of drawing the
images.
- With mouse wheel scroll, the initial responsiveness feel snappier
than with keyboard, but the application may easily become unresponsive
for 2-10 seconds on my machine at least (Core 2 Duo 2.5GHz)
- When you send a message, the lag when pressing Send seems to be gone
in 3.3.91, although there is still no feedback of the actual sending
until it's fetched from the remote server (user needs to just trust
that it's being sent)

Some are performance issues, some are simply about additional code
needed for giving immediate visible feedback/animation/transition on
all actions user may do. But in principle the truly smooth scrolling
covers already fixing so many things, that after that the rest will be
comparatively minor UI enhancements to reduce the perceived lagginess.

I now filed bug #955747 [1] for the feedback wishlist item.

-Timo

[1] https://bugs.launchpad.net/ubuntu/+source/gwibber/+bug/955747

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Drop Gwibber from default install

[Timo Jyrinki]

2012/3/12 Ken VanDine ken.vand...@canonical.com:
 For those that haven't tried the latest version, I really suggest
 getting 3.3.91 and taking that for a spin.  Most of the effort we've put
 into it recently have been quality, so no stunning changes but more
 reliability.  Duplicate detection and handling of the content.
 Scrolling and keyboard navigation has improved quite a bit, but we
 really need smooth scrolling.  I would say that would be at the top of
 the todo list.

Thanks for your efforts so far! I tried the 3.3.91 out. Sure it's the
good ol' laggy Gwibber, but at some point earlier it regressed in
usability for me so far that I stopped using it. Now it seems to be
back on track, and hopefully indeed the four bugs mentioned in the
original post will see themselves fixed at some point.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: MySQL's future in Debian and Ubuntu

[Timo Jyrinki]

2012/2/7 Clint Byrum cl...@ubuntu.com:
 So what should we, the Debian and Ubuntu MySQL maintainers and users,
 do about this?

This is really more for the heavy users to answer, but I haven't seen
any indication that MariaDB wouldn't be an LTS worthy replacement. It
has the majority of former MySQL core developers, corporate backing
and indeed an open development process. Essentially it could be
possible to say that it's more the new MySQL than LibreOffice is
the new OpenOffice.org? (given Apache/IBM backing for the old OOo,
but well at the same time ignoring that while Oracle has given up on
OOo, it's doing something with MySQL still).

Anyway, the pros sounds like outweighing cons, especially given the
problems in security support et cetera with MySQL.

But I hope you'll have a consensus within the Debian/Ubuntu
maintainers. For Ubuntu check up the possible memo notes from past
UDSes since this is not like it'd be a completely new topic.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Proposal to delay release of Precise Pangolin

[Timo Jyrinki]

2011/10/19 Dustin Kirkland kirkl...@ubuntu.com:
 The 12.04.1 (the first of the dot releases) is scheduled for 23
 August 2012, which is about 4 months after the 12.04 release (26 April
 2012).  The dot release is, in fact, a bug-fix and
 hardware-enablement only release cycle.  Realistically, some
 enterprise server and corporate desktop users won't upgrade until that
 first dot release.

Actually since Update Manager only starts offering 10.04 LTS - 12.04
LTS around the .1 time frame, most of the LTS users won't upgrade
until then. I think that's great and precisely like it should be. Most
of the current 11.10 users will upgrade to 12.04 earlier, and already
get a lot more polished release than the usual half-yearly. Then the
LTS users should get a really smooth experience in August with the
additional fixes.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Default Desktop Experience for 11.04

[Timo Jyrinki]

2011/4/8 Timo Jyrinki timo.jyri...@gmail.com:
 There are a lot of bugs and lack of features (and many have been fixed
 already as well) and the performance is quite bad in parts, but those
 are not as serious as a) crashers and potentially b) accessibility and
 lack of any help.

Just reflecting on the more recent posts, I'm using Unity on my work
machine, but I do have only 1280x1024 (or 1400x900, depending on if I
use internal or external display) resolution. I mostly run everything
full screen, and most of the time I don't use Unity to switch between
windows at least yet - I use either alt-tab (somewhat annoyingly slow)
or compiz's scale plugin which I have bound to lower right corner of
the screen (works pretty nicely for me).

As for what I say to others factor, I (and Ubuntu Finland website as
decided by us already around 10.10 time) continue recommending Ubuntu
10.04.2 LTS for everyone. I don't believe users should generally
install a non-LTS Ubuntu, even with the caveat of not the newest
hardware support. 18 months of security support and therefore need to
upgrade N number of times to get to the next LTS is too much for many,
since the upgrade is still something of a hassle at times, regressions
appear et cetera. Unity being still maturing is just one factor that
contributes to this, but I wouldn't have any problem recommending
12.04 LTS with Unity to everyone, since it's going to be ok already
in 11.04 (and fabulous effort / re-write since 10.10) it's a piece of
cake to believe it keeps improving. Not that I would have any problem
with gnome-shell either, it's becoming great nowadays as well.

I know that as a power user I'm from the more adjusts to the
environment part of scale. I don't need to keep doing the way I've
been doing before, and I usually stick to quite near the shipping
defaults. I did have focus follows mouse though, which I disabled
since it worked so poorly with Unity :P Of course I wouldn't keep
using Unity if it hadn't improved in the last month like it has, but
the application launching via Super key or Alt-F2 really starts to
work now, better than ever in GNOME 2. Still too laggy and does not
always just work, but most of the time it's neat.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Default Desktop Experience for 11.04

[Timo Jyrinki]

2011/4/8 Martin Pitt martin.p...@ubuntu.com:
 I couldn't have believed it even two months ago still, but today I
 feel the same. When I switch back to classic GNOME it feels inferior
 now; I'm particularly missing the super-fast keyboard
 shortcuts/search/navigation and bigger screen real estate.

I have started to like Unity a lot, at least on a though level and
also seeing in practice that it's really improving. The biggest issues
have been that unity has been crashing for me all the time. Today is
actually the first day that unity/compiz didn't crash within a minute
of logging in when alt-tabbing or something similar (then it usually
took longer time before it crashed the next time). Fingers crossed
that unity 3.8.4 is now actually more stable in real use - same was
said about 3.8.2. If it stays for a day of work without crashing,
that's a really good accomplishment compared to before.

Besides fixing crashers I really would see need for more accessibility
support and help. I don't know how to access eg. indicators or system
menu from keyboard, which is quite essential for me even without
disabilities, but for people with disabilities I believe the
accessibility in general is relatively poor at the moment. gnome-shell
already has a lot of a11y stuff integrated in 3.0 (considering it's
the first stable release), and Ubuntu with accessibility as one of the
core Ubuntu philosophy items should have as well. Of course, by 12.04
LTS at least.

There are a lot of bugs and lack of features (and many have been fixed
already as well) and the performance is quite bad in parts, but those
are not as serious as a) crashers and potentially b) accessibility and
lack of any help.

For 11.10, probably something should be done about the logging in
time, with is terrible at least with a traditional spinning, encrypted
disk, compared to normal Gnome. Weirdly sometimes I saw a pretty fast
logging in even after reboot, but normally it's 30s+ from gdm to
desktop. Something is seriously churning the hard disk with seeks,
possibly something that only occurs with specific conditions.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Default Desktop Experience for 11.04

[Timo Jyrinki]

2011/4/8 Neil Jagdish Patel neil.pa...@canonical.com:
 3.8.4 should be much, much more stable, especially if you're on a 64-bit
 system. The entire team is concentrated on crashers and I think we'll
 have a very stable Unity by hard-freeze.

Sounds good, and yes I've 64-bit which explains a bit.

 F10 - Opens first available menu in Panel and then left/right arrows let
 you move between all the menus on the panel
 Alt+F1 - Focuses the launcher and allows you to keyboard-navigate the
 icons and also navigate the Quicklists.
 Super - Opens the Dash

Yippee! Super and Alt+F2 I had realized (and Alt+F2 was great to have
when it appeared and I noticed it from changelog), F10 and Alt+F1 were
news to me and they seem to work great at the first sight (except when
in calendar, where it seems there is zero way of getting out with
keyboard only - I will check if there is a bug for that).

The keyboard usability just got a huge boost for me. This illustrates
though somewhat the problem of finding these out and the lack of any
built-in Help.

-Timo

-- 
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

Re: Booting and login - why are users not logged in automatically?

[Timo Jyrinki]

2010/3/24 Evan eapa...@gmail.com:
 For security we definitely want users to enter their passwords by default.

 However, given this default you raise a good point that ureadahead
 should be optimized for this option, not for auto-login.

Note though that Ubuntu Netbook defaults to auto-login, and netbook
boot speed has been the primary goal with boot speed optimizations
AFAIK (although everyone benefits of course, a lot).

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: String break in ubuntone-client without freeze exception

[Timo Jyrinki]

2010/3/15 Rodney Dawes rodney.da...@canonical.com:
 which there are translations, and doesn't really explain which
 strings you feel broke this freeze. Can you list them please, so
 we can determine which ones exactly you think broke the freeze?

 I'll leave you in the capable hands of Elliot if there are any
 ones that do violate the freeze.

It's not as this specific case would be a problem anymore since I
basically took care of the string freeze exception announcement, it's
mainly for the future and all other developers as well. Ubuntu
translators and documentation community have been battling with all
kind of package uploads after string freeze release by release, and I
simply took an opportunity to point out an example so that more
developers would become aware of what the UI freeze means for
translations and docs team, and basically the whole Ubuntu experience.

But just as a note, after 1.1.3 the following new or changed (some
ever so slightly) strings were added in 1.1.4, resulting in those
being untranslated for all languages (until teams catch up, if/when
they notice it) and documentation being potentially out of date:

The devices connected to with your personal cloud network are listed below
LOCAL MACHINE
Stop sharing on Ubuntu One...
Stop sharing this folder on Ubuntu One
Stop synchronizing this folder with Ubuntu One.
Synchronize this folder with Ubuntu One.
Copy Ubuntu One public URL
Copy the Ubuntu One public URL for this file to the clipboard.
Stop publishing via Ubuntu One
No longer share this file with everyone.
Publish via Ubuntu One
Make this file available to anyone.

1.1.3 was btw also released after the UI freeze, but that was close enough.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

String break in ubuntone-client without freeze exception

[Timo Jyrinki]

Hi Rodney,

You recently uploaded a new version of ubuntuone-client source
package. Please note that Ubuntu has been in UserInterfaceFreeze since
March 4th, requiring freeze exception including a bug report and a
note to relevant mailing lists. See more information at
https://wiki.ubuntu.com/FreezeExceptionProcess#UserInterfaceFreeze%20Exceptions
. I see that there has been an ongoing feature freeze exception, which
probably should have been user interface freeze exception at the same
time.

cc:ing ubuntu-devel-discuss just to remind everyone of how to handle
the user interface freeze exceptions. Let's just try to keep the
processes in mind, since they are there for a reason. The translators
and documentation people are mainly interested in getting a proper
notification on time to make sure documentation changes and new
translations for both UI and the docs are done as completely as
possible.

The new UI strings are available at
https://translations.launchpad.net/ubuntu/lucid/+source/ubuntuone-client

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Including usb-modeswitch in default installation?

[Timo Jyrinki]

2010/2/25 Peteris Krisjanis pec...@gmail.com:
 By the way, reading all materials, I just thought - usb-modeswitch now
 has nice udev wrapper - why not ship it by default? Why it isn't in
 main? Is there any objections to this? Patents, legal issues (I can't
 think of any, but still)?

I think this is where come back to the beginning of this discussion :)
Ubuntu core developers would like to follow upstream, and it's
currently not yet clear enough what's the usptream direction (or what
upstream we are actually talking about). Josh from usb-modeswitch is
making the point that modem-modeswitch upstream has abandoned its
development already, and that (at least some) kernel developers would
like to see some of the things done in userspace like usb-modeswitch
is doing. So I guess a positive round of discussion on the
linux-hotplug mailing list Scott now pointed out would be what's
needed to get some backing up to what Josh is thinking, and then
Ubuntu could follow the direction most parties seem to agree with.

At least for Ubuntu 10.10, but also if the current situation is not
really wisely supportable from LTS point of view, freeze exception for
even 10.04 LTS. There is only very little time, though.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Including usb-modeswitch in default installation?

[Timo Jyrinki]

2010/2/24 Peteris Krisjanis pec...@gmail.com:
 I fully agree with you Scott, however I am not sure which one of us
 should start discussion there. Would be nice for you Ubuntu guys to
 start pushing this idea, because you have some backing.

 Anyway, I'm subscribing to that list and if no one will volunteer, I
 will try to initialize discussion.

It looks to me Josh from usb-modeswitch has the technical capability
of explaining why usb-modeswitch is the way forward, and has also
implied that modem-modeswitch of udev has already been abandoned by
the upstream. Unfortunately he has just started his vacation (see the
usb-modeswitch forum), so there is not going to be immediate backup
from him. If you think you can express his ideas and references [1]
from his postings well enough, you can start the discussion on
linux-hotplug. Just try to be as polite as possible, since I don't
know what we are touching here :) But so far it seems that Red Hat's
Dan William has abandoned modem-modeswitch, and if also Josh's
argument about it cannot be solved correctly in kernel space only is
accepted, then usb-modeswitch is the natural way to go forward -
possibly by integrating it to udev.

[1] http://www.draisberghof.de/usb_modeswitch/bb/viewtopic.php?p=1906#1906

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Including usb-modeswitch in default installation?

[Timo Jyrinki]

2010/2/13 Timo Jyrinki timo.jyri...@gmail.com:
 I think the main interesting thing would be that what are the
 usb-modeswitch developers' motivations and could they with some help
 from the kernel guys be directed to contribute more to the kernel.

In addition to previous concerns in this thread addressed in
https://lists.ubuntu.com/archives/ubuntu-devel-discuss/2010-February/010691.html
, it seems the upstream is convinced that the usb-modeswitch is
needed, and kernel quirks are not either enough or the way to go:

http://www.draisberghof.de/usb_modeswitch/bb/viewtopic.php?t=322 (scroll down)

I think the udev package's modem-modeswitch point is anyway very
relevant, if it's obsolete and somewhat broken. It's currently
included in Ubuntu's udev package.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Including usb-modeswitch in default installation?

[Timo Jyrinki]

2010/2/9 Timo Jyrinki timo.jyri...@gmail.com:
 If I read those correctly, they work neither with or without
 usb-modeswitch, hence adding it would not decrease any usability?

I poked around on IRC and summarized with my non-existent journalistic
skills the situation in my blog:
http://losca.blogspot.com/2010/02/ubuntu-1004-and-3g-modems-usb.html

I think the main interesting thing would be that what are the
usb-modeswitch developers' motivations and could they with some help
from the kernel guys be directed to contribute more to the kernel. Or
possibly a new group of people (or one person) who work between
usb-modeswitch and kernel people. And of course there is also the
question if solving the problem in userspace is actually a wise thing,
something that's debatable.

On a larger scale, I think it could be worth pondering what Ubuntu
could do in situations like this. I think this is one case where the
users understandably would accept any non-optimal technical solution
if it just seemingly works for them. But if we start having separate,
constantly SRU:d packages for 3G modem support, DKMS modules for eg.
DVB and video cam devices etc., is it something of a mess that Ubuntu
(or Canonical offering official support) doesn't want to wander to? On
the other hand, anything in the backport modules etc. doesn't reach
the average user other than via possible friends/forum visits.

In other words, self-updating hardware support is something that would
be nice for LTS at some point, since the only reason I sometimes
recommend non-LTS to random users is that it's more probable the newer
non-LTS version supports newer hardware.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Including usb-modeswitch in default installation?

[Timo Jyrinki]

Hi,

a) Is anyone aware of some earlier discussions on the topic wrt Ubuntu
and b) Would everyone agree that it would make a lot of sense?

3G broadband usage is incredibly popular nowadays at least around
here, and usb-modeswitch is a must have for a majority (?) of new 3G
modems these days. It's a very simple addition of udev rules, and
requires nothing else besides having it installed - the 3G modem will
show up in Network Manager.

Any objections or observations for a MIR (main inclusion request) of
usb-modeswitch and inclusion on the default installation of Ubuntu
10.04 LTS? Binary package size is 30kB + 10kB (data).

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Minutes from the Technical Board meeting, 2009-09-22

[Timo Jyrinki]

2009/9/22 Jordan Mantha laserj...@ubuntu.com:
   * Based on feedback all around, it was agreed that it is still early
     for sun-java6 to be dropped.  However, it was not yet clear if
     responsibility for the package should be changed.
   * ACTION: kees to drive sun-java6 email thread and get resolution on
     responsibility
 Would it help if ubuntu-restricted-extras was icedtea6-plugin |
 sun-java6-plugin instead of sun-java6-plugin | icedtea6-plugin ? I'm
 guessing a whole lot of sun-java6 installations are from people, like
 me, who just install ubuntu-restricted-extras and don't think about
 specifically what Java VM is on my machine.

Then again, what would be there restricted anymore? Maybe Java should
simply be removed from ubuntu-restricted-extras altogether, and
OpenJDK would be installed automatically (or automatically as needed,
similar to I guess the extra OpenOffice.org components)? At least when
it's seen suitable, ie. no big regressions.

And further about restricted-extras, is it truly that the MS-replacing
fonts from liberation and wine projects couldn't be installed by
default nowadays, getting rid of the MS ttf font installer also as it
would not be needed for majority of cases? Not that I'd have ever
installed it myself ever, but apparently some/many tend to install it.
But is it nowadays more of a habit than a real need?

After that restricted-extras would be simply limited to
patent-encumbered media formats and adobe's flash plugin, which I'd
prefer to offering so much possibly-(soon?-)not-really-relevant other
restricted stuff for the meta package installers.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Regression in getting fully translated Ubuntu installation - what to do for karmic?

[Timo Jyrinki]

Hi,

CC:ing ubuntu-translators just to get interested people on-board, no
need to continue CC:ing.

Please look at https://bugs.launchpad.net/language-selector/+bug/434173
- do you think I'm missing something, can the report be refined? The
new way of handling supporting translations, writing aids etc. moves
them from package dependencies to language-selector. However, it means
that new installations of Ubuntu do not get the full language support
at install time, even when connected to Internet. My previous guidance
for those reading an installation guide has been to enable Internet
connection before starting the installer to get eg. Finnish support.

https://wiki.ubuntu.com/language-selector-karmic describes the user
should get a notification after installation. First of all, I tested
it and it seems non-functional at the moment (a bug on its own).
Secondly, I do strongly think it's not enough - for example 50% of the
users could easily not be interested at the one-time popup, forgetting
about it - but they will run into a bad Ubuntu experience once they
are in a situation of lacking a spell-checker for example, or reading
English documentation.

Also, it does not seem Ubuntu tools currently are very error-proof in
case of lacking an Internet connection - they will simply fail and no
new notifications (or guiding to connect to Internet) are given.

To match Ubuntu 9.04 functionality, I think Ubuntu installer should
call language-selector to install full language support at the
installation time, if the Internet connection is available. This would
result in a similar behavior to 9.04, with the pop-up being there to
hopefully help those who didn't enable the Internet connection (there
is no guidance for it when starting from the Live CD).

Besides am I correct or could you refine the bug report, I'm
interested in against which packages the bug should be filed? Is it a
a) bug against language-selector (to have some silent --fix-missing
to be used by the installer), b) bug against installer (launch
language-selector, silent or not), c) bug against update-manager (I
would like to have more than just one pop-up about installing full
support - for example, update-manager could completely automatically
download full support together with the first updates for a single
time), d) bug report against The Whole World (language-selector +
lang-pack-o-matic changes + ...) since the problem appeared in the
first place? If you know the answer or part of it, please go ahead and
refine the bug.

I'm also interested if these were thought about during creation of the
Karmic language-selector blueprint? The popup notification is not
enough IMHO, and it wasn't enough in 9.04 either. But especially as
the situation is now worse in 9.10 than it was in 9.04, can we bring
it to 9.04 level somehow? After that, we might think how to actually
make it Just Work for lucid (I'm thinking about guiding the user by
hand to connect to Internet etc).

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Thanks to you all, 9.04 will be a great release in all languages

[Timo Jyrinki]

Hi,

This is just a small ”thank you” and also ”just look at that” kind of
post. The thing that should be looked at is:

https://wiki.ubuntu.com/TranslatingUbuntu/JauntyTranslationIssues

Virtually all I18N issues that were a) never fixed in eg. intrepid (or
earlier) b) found out as new (having existed also before) have been
fixed during jaunty. Developers have been extremely helpful in making
the ”Every computer user should be able to use their software in the
language of their choice” a reality. Also the general knowledge of
I18N issues and how to do it right has increased.

In addition to those mentioned on the page, there have been
improvements elsewhere as well. Debian package description
translations have now a working two-direction solution for merging
(proposing) translations between Debian and Ubuntu, and for the first
time anyway all the DDTP scripts are top-notch so that all description
translations from both Debian and Ubuntu are shown as they should be.
The short package descriptions shown in eg. gnome-app-install are also
fully translatable now for the first time, though also intrepid had
the possibility for majority of the data there (before that, nothing
was possible to translate regarding the short descriptions). These
affect a lot anyone considering installing a package, which now do not
necessarily see foreign language at all (of course it's useful if the
actual program is also localized after the installation ;)).

There are still places where more or better I18N would be needed. I
would guess some notifications might still be lacking I18N, or
translations are hard-coded. There are maybe a bit too many (not so
easy to solve) hacks where proper translations need manual
maintainer intervention. KDE (or Kubuntu, or Ubuntu build system) has
some obvious I18N problems in places. Hard drive check (and other boot
stuff, except live-cd boot menu) messages are untranslated. The
Live-CD simply cannot, ever, include all translations so for most
languages the live session is purely English except for the installer.
Rosetta, upstreams, Ubuntu-specific changes... well, has been
discussed forever and will need even more of that. Etc.

But still, the progress jaunty has made is truly awesome, since
basically everything that has perhaps bothered the casual Ubuntu
default installation user has been fixed. It's only that when you
approach perfectness, even the little flaws, in corners, show up
better. All in all Ubuntu 9.04 will be I think the most translated and
translatable free software desktop distribution there is.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Reason for removing animation from Gnome login?

[Timo Jyrinki]

2009/3/25 Marius Gedminas mar...@pov.lt:
 I'm wondering why readahead doesn't load all those thousands of small
 files needed for desktop startup during boot, before I try to log in.
 I'm guessing that it just doesn't know the particular ones that I need -
 everyone's GNOME settings are different, applets are different,
 wallpapers are different, themes are different.

Right, but also it wouldn't help since simply seeking to one thousand
files with an average seek time of 15ms (seek + rotational latency)
would mean 15 seconds of pure seeking, not to say anything about
actually transferring data (or doing something about the data on the
cpu). And I'd like grub - desktop loaded to be eg. 10s.

Readahead is already doing good things, and it should be loading the
files it does in an order that reduces the seek time (it can fetch
things as they are ordered on the disk), but in the end the amount of
files and seeking should be anyway reduced AFAIK. But optimization
during 9.04 has concentrated on a lot of other subjects which prevent
fast boot anyway, so it's also a matter of the whole picture - if
there is anything halting the boot, it doesn't help if one part is
completely optimized.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Reason for removing animation from Gnome login?

[Timo Jyrinki]

2009/3/23 Marius Gedminas mar...@pov.lt:
 Personally, for me the desktop startup is (or feels) I/O-bound, with the
 panel applets showing up one by one with agonizing pauses in between and
 the disk running at full throttle.

Yes, it is. In jaunty, the current warm start (everything in cache,
ie. login, logout and then login again) time is ca. 5 seconds, with
compiz, something that would be actually ok for a login time in
general instead of the current status.

And the I/O problem comes from the hundreds/thousands of small files
that are inefficiently read. The amount of transferred data is not
that much that it would take more than 2-4s to read on modern even
laptop hard drives, if the data would be sequentially available in a
one big chunk.

2009/3/23 Evan eapa...@gmail.com:
 My guess is that it's a bug in Compiz, coincidentally introduced in the same
 cycle as compiz-by-default. When I disable desktop effects (running
 intrepid) my login speeds up by a good 15 seconds. Does anybody else notice
 this?

This has been fixed in the newer libcompizconfig in jaunty.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: translating the categories in the totem BBC plugin or not?

[Timo Jyrinki]

2009/3/6 Sebastien Bacher seb...@ubuntu.com:
 The change add the list of categories to the translatable strings and I
 would gather some opinion on whether people think that's a good idea or
 not knowing that the media content available is in english.

I'd think it's enough that the actual titles are in English. To the
extent that there is also video content, people can enjoy sights and
sounds even if they do not understand the speech, and it's nice to
have categories in an understandable language for random surfing.

I cannot seem to be able to connect to the server at the moment, but
in the case there are no videos and won't be much in the future
either, I'd say it's also ok to remove the I18N of the categories from
the patch since usually there is not much enjoyment in audio-only in
foreign language. I seem to recall I was unable to find video content
something like half a year ago easily from there.

But in either case, because of the titles being in English, I don't
see much reason to worry about being mislead. Also like Adi just
wrote, it can be left for translator teams to decide, which means
keeping the L10n possibility.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Go-OOO.org?

[Timo Jyrinki]

2008/12/30 Joe Terranova joeterran...@gmail.com:
 a) Open Office (and its derivatives I assume) is a bear to package.
 Transition packages between releases open up more points of failure.
 Will Go-Ooo.org last, or die in 2 months?

Is anyone btw familiar how much of go-oo.org is Novell's business
competition to Sun, since it was initiated (?) there and has active
developers hired by Novell, and how much is actual community? Ie. was
it Novell/SUSE first with the hopes (very realistic, given the
problems in approachability of Sun's project) that will grow into more
true community project, or was it go-oo.org community first adopted by
Novell for sponsoring? Anyway, Novell's involvement with go-oo.org is
quite largely hidden. Just interested in the history.

Also are there any more (wrt Sun's version) potential risks in the
code in go-oo.org codebase related to Microsoft-Novell deal and the
covenant not to sue which has been mentioned to be risky business
even though Microsoft states in their FAQ that OSP applies to GPL
software _but_ that they also state that it's up to interpretation of
GPL what it means, and of course their preferred interpretation would
be that GPL would not be entirely libre (more at
http://www.softwarefreedom.org/resources/2008/osp-gpl.html). The main
problem I guess is that it's covenant not to sue only for office
document specification implementations, but what happens if code from
OOo will get used somewhere else under the GPL, and it's using some
OSP-related patent but for some other purpose? And of course my
thought about primary MS aims is that they'd like to make open source
non-commercial even though any non-commercialism is against open
source / free software principles, by using patent licenses as the way
to cripple down open source software (first yes you can use but in
case of major commercial use hey you're using our patents).
Hopefully that won't ever succeed, though on the other hand it's
already succeeding to an extent with all the unfounded, broad claims
about Linux kernel etc.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Ubuntu 8.10 significantly slower than previous versions

[Timo Jyrinki]

I wonder how this discussion is able to drift so much away from the
actual subject on both ubuntu-devel and ubuntu-devel-discuss. Many
people do not want to believe results or just point out one or two of
them are meaningless (like NVIDIA graphics performance with closed
drivers is not that interesting).

2008/11/6 mr [EMAIL PROTECTED]:
 According to the recent benchmarking article by Phoronix, the previous two
 releases of Ubuntu are significantly slower than Feisty Fawn. In some cases
 this can be seen as up to 50% performance drop with certain desktop tasks.

Yep. I found three important points I list below, sound encoding
(video might be because of changed default parameters, dunno), SQLite,
compiling.

1. See eg. the page:
http://www.phoronix.com/scan.php?page=articleitem=ubuntu_bench_2008num=4
It takes about double time to encode mp3:s, ogg:s or flac:s In Ubuntu
8.04/8.10 vs. earlier. Fedora seems to have been affected by the
problem all the time:
http://www.phoronix.com/scan.php?page=articleitem=fedora_test_2008num=2

2. SQLite is over two times slower in Ubuntu 8.10 vs. earlier:
http://www.phoronix.com/scan.php?page=articleitem=ubuntu_bench_2008num=5

3. Compiling has been almost two times slower in Ubuntu than Fedora
after Ubuntu 7.04:
http://www.phoronix.com/scan.php?page=articleitem=fedora_test_2008num=2

The 1. and 3. is what I'd worry about. 1. in general and 3. because
Fedora is almost two times faster.

(from later in the thread)
 The default scheduler is optimised for general desktop usage, where you
 have a large number of simultaneously running applications, applets,
 etc. and each one needs to be responsive.

Please, everybody, do not take this as granted. Question it, test it,
feel it etc. With the default CFQ, I could not do about _anything_
when I did eg. svn update or Firefox churned through its enormous
databases on a laptop hard drive. After changing to elevator=deadline
these cases work _much_ smoother without visible regressions
elsewhere.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: OpenOffice 3 and Firefox 3.1 in Intrepid?

[Timo Jyrinki]

2008/9/5 Chris Cheney [EMAIL PROTECTED]:
 released on Sept 8 which puts the final release around Oct 20. That
 doesn't leave enough time to make it even marginally stable, since that
 would be only 3 days before the Intrepid release candidate.

Note that there are also dependencies like openoffice.org-voikko which
would need upgrading to a new version too if OOo 3 would be put into
intrepid.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Boot-time improvements

[Timo Jyrinki]

2008/9/11 Krzysztof Lichota [EMAIL PROTECTED]:
 As an author of Prefetch, I cannot agree that it would not fix seeks ;)
 Part of my implementation, not enabled by default as it is highly
 experimental, is ext3 defragmenter which puts all files for prefetch
 in one place on disk, so the requests to read them can be merged into
 big streaming reads.

I stand happily corrected :) Hopefully you will get some help in
testing the defragmenter and getting that part stable in time, too.
The hard part in the whole prefetch is probably it being kernel-based,
but the diff [1] seemed quite non-invasive, ie. only a few calls in
the existing kernel code and then some new calls.

[1] 
http://code.google.com/p/prefetch/source/browse/#svn/trunk/kernel-patches/2.6.22

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Boot-time improvements

[Timo Jyrinki]

2008/9/12 Chris Jones [EMAIL PROTECTED]:
 the mailing list here. So, as a test, I timed my own system how long it
 takes to boot. From GRUB boot to login screen, it was 36.72 secs. And
 that is on a Celeron D with 512MB DDR RAM and booting from an old IDE
 hard drive. I'd imagine that a more recent dual core setup with more ram
 and a more recent sata hard drive would have better results. But my mere
 36 secs is certainly nothing to rant about.

One problem is that the dual core setups do not help that much because
the bottle-neck is in hard drives which haven't gotten speedier at all
in the last 15 years regarding seek times, and even transfer speeds
are relatively slowly increasing compared to other aspects. Especially
dual-core laptops are largely as slow to boot as desktop computers
from 2-5 years ago, because of the slower hard disks.

The other problem is that logging into GNOME often takes the same time
it takes from GRUB to GDM.

The aim would be IMHO something like 15-20s from GRUB to GNOME desktop
on a modern _laptop_ computer, with smooth transitions.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Boot-time improvements

[Timo Jyrinki]

2008/9/12 Oliver Grawert [EMAIL PROTECTED]:
 You have 25-35 seconds from GRUB till desktop appearing (using autologin)?

 i have managed 22 already with a not to drastic set of modifications (19
 after grub [1] plus the three second grub timeout) with the ten seconds
 my BIOS takes that even stays below 35 for the complete boot process
 from hitting power button to see gdm.

That's not till desktop appearing (using autologin). People often
speak about different things when saying boot, but regardless of the
term used the interest here is when the desktop is usable after
powering up the computer. It can be measured by temporarily activating
auto-login in gdm setup to bypass gdm. It can even double that figure,
and it's the figure which we are interested in. One cannot do much
work with gdm.

Like I said, I'd say the aim should be around 15-20s ultimately on
modern (but non-SSD) laptops for GRUB-to-desktop, since we cannot
affect BIOS-to-GRUB at the moment. Though that cannot be reached
without fully using prefetch or having GNOME login optimized otherwise
very heavily.

 i think we still have to do our homework first and get the existing stuff 
 into a
 reasonable shape

I agree that every optimization counts, even if it's small (but a lot
of small fixes make for a bigger change).

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Boot-time improvements

[Timo Jyrinki]

2008/9/10 Przemysław Kulczycki [EMAIL PROTECTED]:
 Maybe we could achieve it using Prefetch?
 https://wiki.ubuntu.com/DesktopTeam/Specs/Prefetch
 https://wiki.ubuntu.com/AutomaticBootAndApplicationPrefetchingSpec

Prefetch would be nice, and definitely would improve the user
experience. In the end, it would however not fix the problem with
seeking to thousands of different files / positions. The only way to
fix that problem is to go with sequential, big files, either by
precisely reordering stuff on the disk in an area that can be made
100% sequential, or by copying files to a cache of sort, being a
single sequential file which contains everything needed and is updated
when needed.

There are so many seeks done from GRUB to GNOME desktop that the other
methods can only bring the time closer to the time spend on doing just
the seeks, which is still probably 20s or more. Of course all good
methods are worth implementing, since eg. 25s from GRUB to desktop
would be of course an improvement...

-Timo
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Boot-time improvements

[Timo Jyrinki]

2008/9/9 Jonathan Carter (highvoltage) [EMAIL PROTECTED]:
 2. Remove the flashes and brown screens when starting Gnome

 Personally, my biggest annoyance as a user is the brown screen that
 flashes between GDM and when Gnome starts. It would be much nicer if gdm
 displayed the greeter for half a second longer, and when Gnome has the
 desktop wallpaper ready, it would do some nice GL transition to the
 actual desktop environment. Where there has to be transitions, it would
 be good to keep it as smooth as possible.

I think this is an excellent point that the GDM would show logging
in for as long as desktop wallpaper can be switched to.

 It seems that with the current process, it's close to as fast as it can
 be at the moment. Readahead avoids unnecessary disk seeks.

No, really, I think there is room for at least 5x improvement in disk
seeks. The disk usage is currently terrible, and it is what is is
because of the wish not to do too much optimization work so that
there is no risk of breaking any complex setup of programs. The
optimization of disk reads should be possible to do, but it would
require a large amount of careful planning and testing and modifying
different software, working together with the upstream. And some
things like reordering/copying needed files on the disk to eg. a big
sequential file just seems too ugly from technical point of view to
some people, even though there are simply no alternatives to bringing
the same kind of benefits.

The amount of information GNOME reads from disk is probably under
100MB, which modern hard drives read in one second. However, there are
thousands of seeks, and each hundred of those consumes another second
of seeks. The current login time from GDM to GNOME desktop is
something like 20-40s depending on the disk speed, while if you have
the information from disk already in memory (logout and login again),
it might be as low as 3s (excluding compiz which seems very slow).

There also often seems to be new bottle-necks when old ones are fixed,
eg. I'm unsure why intrepid GRUB-to-GDM seems so slow while earlier
GDM-to-GNOME was the most annoying. Also GNOME 2.24 would seem to have
(for me) something preventing the 3s startup I was seeing in hardy
times with all things cached, even when compiz is disabled.

 If something like Usplash continues to be used, I think it's important
 that it transitions well into the next phase. Not sure if that's even
 possible when using X.

There have been blueprints on the transitions before, so it's probably
very well known that it is wanted at some point. But it needs the new
features from kernel etc. to truly fulfill it. Fedora is indeed
thinking about the stuff already with their Fedora 10, from GRUB menu
to the GNOME/KDE desktop, and I hope they manage to do it well.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Did we really release 8.04?

[Timo Jyrinki]

2008/7/11 Matthew Paul Thomas [EMAIL PROTECTED]:
 That doesn't require a script. For example, bugs targeted to
 intrepid-alpha-2, sorted by number of duplicates:
 https://bugs.edge.launchpad.net/ubuntu/+bugs?orderby=-number_of_duplicatesfield.milestone%3Alist=1320

Cool, didn't know that! Added to https://wiki.ubuntu.com/Bugs/Stats.
Looks like that's also nowadays in the Advanced Search dialog of
Launchpad, which is great. There also
https://bugs.launchpad.net/ubuntu/+bugs?orderby=-message_count

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Did we really release 8.04?

[Timo Jyrinki]

2008/7/7 Bryce Harrington [EMAIL PROTECTED]:
 I think the single most needed feature in Launchpad regarding this
 would be the possibility for voting, as done in bugzilla.

 I don't know if I necessarily agree with this.  Voting may feel good
 from a user perspective and certainly couldn't hurt, but we already have
 good ways of measuring user interest in particular bugs.

Right, it may be it's more about giving users a good feeling, but we
should probably be doing that, too... It does work for some projects
like Wine and OpenOffice.org as a measurement of a popularity of a
problem.

 For instance,
 Brian Murray captures bugs with many subscribers (which I link to from
 http://people.ubuntu.com/~bryce/Xorg/status_current.html).

I linked to Brian's directory from https://wiki.ubuntu.com/Bugs/Stats
now, since those kind of services are valuable if they are just easily
found.

 With voting, I would worry that we'd see all the usual problems with
 gaming voting systems.  Someone might file a stub bug Ubuntu crashes
 randomly with no details, post something like plz vote this bug if u
 want ubuntu to never crash again, and accumulate +10,000 votes to it.

Yes, it might not work, too :) My hope would be that those bugs would
be in a minority and going through eg. 50 most voted bugs would reveal
some important bugs that wouldn't have been revealed otherwise, since
the people hovering around that specific bug didn't know how to use
Launchpad to propose it for a milestone or contact QA.

Granted, getting sorted by comments could help in the same thing as
voting, but voting could decrease noise in the bug reports.

 I'm not sure if it is even possible to sort bugs by the amount of
 duplicates?
 Yes it is; in fact I have a script I'm going to make available soonish
 which lets you do this with milestoned bugs.

Great. That's another thing that should actually be part of Launchpad,
but a script would help too. At least a portion of developers / QA
team needs to know an easy way to sort by duplicates, comments and
subscribers, so that the bugs are noticed in time.

 We need more people doing this.

Yes, we need more triagers, and like you said earlier we would need
more developers. And I think all the guides about how to join QA /
BugSquad / MOTU etc. are on the right track about this, but still we
could/should attract even more people.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Did we really release 8.04?

[Timo Jyrinki]

2008/7/7 Emmet Hikory [EMAIL PROTECTED]:
  While we all tend to be busy much of the time, perhaps there are ways
 that we can improve the view of bugs in need of attention, or
 otherwise help understand which bugs are likely to be perceived as
 painful to users at release time.

I think the single most needed feature in Launchpad regarding this
would be the possibility for voting, as done in bugzilla. There have
often been many very important bugs that no developer ever sees, and
it's a matter of a bit of luck if someone actually knows how to get
the bug proper attention. Some of those bugs are recognizable by the
month-after-month comments questioning if any developer sees it, but
the comments as such do nothing to raise the visibility of the bug as
such. Vote would do it, and also would get some quantitative measure
about if it's just a loud person or two or a really significant amount
of people.

I'm not sure if it is even possible to sort bugs by the amount of
duplicates? That's another measurement, though less certain since it
already requires some capable Launchpad user to have browsed through
the issue. Out of the ordinary users able to file a bug and use
Launchpad, only a few really are interested enough to do actual
triaging/searching/marking.

The current (core) developers are very largely busy enough without the
extra work of (really) going through all the hundreds of bug reports.
Bug triaging should help them, but there are not enough triagers
either, which also need special skills and privileges. Voting can be
done by everyone and would catch some possibly much-needed bugs to get
the attention they need (at least from the triagers). Of course some
of them will be false alarms (whyyy my fglrx is so broken), but not
too many I hope.

Another thing I'm not touching is how to convert the massive Ubuntu
user base into contributing developers :) That includes of course the
people who like to blog instead of filing bug reports / creating
patches.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Banshee by default in Intrepid

[Timo Jyrinki]

2008/6/9 Mackenzie Morgan [EMAIL PROTECTED]:
 I was a Banshee user for about 9 months or so.  That ended a year ago.
 Unless it's changed,  Banshee only recognizes songs you have imported.  It
 does not monitor directories for new songs, and especially not recursively.

Yep, and for this very reason (and other similar big problems) I
cannot understand the choice of F-Spot either, except that there is
less competition for it (no Rhytmbox-like equal or better, and
stable choice). F-Spot and Banshee are both would be nice, but
basically quite far from being really usable unless the user starts to
use them from scratch and/or never uses any other programs to handle
photos or music. Which is kind of what free software should not be
about.

Additionally, just for it to be noted even though Banshee is not going
to be default anytime soon anyway, please do not even think of
replacing Rhythmbox with a player that does not have the Jamendo and
Magnatune integration it it. And same goes for the DAAP/DLAN/UPnP
features.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: i18n in Ubuntu 8.04 LTS - great work! ( current status)

[Timo Jyrinki]

2008/4/20 (``-_-´´) -- Fernando [EMAIL PROTECTED]:
  I'll have to look for our bug number and report back, but as you can see, 
 this is a major showstopper, because any Portuguese Ubuntu User will be 
 presented with and English GUI of OOo.
  Is this a question of CD space?

Hi. Unfortunately the Internet connection _has_ to be enabled during
installation for complete language support to be enabled. There's no
space on the CDs for practically any (full) language support, so they
have to be downloaded. If the Internet connection is enabled during
installation, full support should get downloaded (and it works for me
and has worked in the past for all people, too). You with Portuguese
at least have some translations on the CD, most other languages do not
have any besides the installation program itself.

I have a long term wish that the fact that non-English people need the
Internet connection enabled would be recognized to the extent that
people would really be guided to click on the Network Manager icon (or
just presented with a dialog Please select your Internet connection
in the installer which would probably be better) to enable eg. WLAN
connection before the installation. Wired connections should
theoretically be automatically enabled, though.

I meant this generic problem with the larger issue with language
support I mentioned in the text you quoted.

Currently I'm just including it clearly in the Ubuntu Finland's
installation guide (http://wiki.ubuntu-fi.org/Asentaminen_8.04) that
the Internet connection should be enabled before starting the
installation.

The problem is not helped by the problems of Language Selector before
software sources are refreshed after installation, and even when they
are - https://bugs.launchpad.net/ubuntu/+source/language-selector/+bug/135752
(I just added notes regarding hardy installation without Internet
connection enabled).

Anyway, I'm very happy how well stuff generally works when the
Internet connection is available and enabled during installation.

-Timo
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: i18n in Ubuntu 8.04 LTS - great work! ( current status)

[Timo Jyrinki]

2008/4/21 Martin Pitt [EMAIL PROTECTED]:
  In cases where that's not possible, I think the current approach of
  having the language selector install the missing pieces is better.

Yes, without any Internet connection available during install,
post-install should have enough guiding for the language selector
usage.

  Since a few days ago this should work even better:
...
  Testing and feedback heavily appreaciated!

I noted in my bug [1] that I just installed using a image containing
that new version, and didn't notice any notes at any point of time.
When should the note come up?

Also other feedback in that bug report, ie. language support
downloading for the default language is not suggested even when
Internet connection is available post-install and repositories
(manually) refreshed, if the installation was done without Internet
connection.

[1] https://bugs.launchpad.net/ubuntu/+source/language-selector/+bug/135752

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Unneeded System Tools menu

[Timo Jyrinki]

2008/3/31, Matthew East [EMAIL PROTECTED]:
  A better solution in my opinion would be to move the
  Applications - System Tools submenu to a System - Tools submenu.

I agree that the current setup is very poor, for the reasons already presented.

If the System - Tools is unfeasible at this point, and if the System
Monitor is not wanted to put back under System menu (under a submenu
or as a separate item), it could fit in the Applications-Accessories
well enough. Print jobs tool is already there, which is a bit
similar...

That would leave only hwtest-gtk, which is strictly an admin tool
(asks for sudo rights) so it could maybe fit under the System menu. I
actually liked its place inside the HAL Device Manager when such
program was included, now I'm not sure where it should be...

By the hwtest-gtk is currently completely untranslated and
untranslatable:
https://bugs.launchpad.net/ubuntu/+source/hwtest/+bug/202447, which
adds to the need of it being not too visible unless the bug is fixed.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: What happened to IcedTea in Hardy?

[Timo Jyrinki]

2008/3/23, Thilo Six [EMAIL PROTECTED]:
 https://bugs.launchpad.net/ubuntu/+source/icedtea-java7/+bug/203636

I don't think that answers the last question, since the current
openjdk-6-* packages do not include Java browser plugin. So is there a
specific openjdk-6 browser plugin coming, or is there something else
that should be used instead?

icedtea-gcjwebplugin seems to work, but it's at least funnily named
now. It is however at least suggested by openjdk-6-jre, though many
may miss it.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Remaining i18n/localization issues in Ubuntu 8.04 LTS, please fix your packages and use non-English yourself

[Timo Jyrinki]

Hi, some i18n nagging from me again, yay!

First of all, I'd hope that developers that are non-native English
speakers or otherwise handle other languages fluently, would use that
language in their desktop environment eg. for the rest of hardy cycle
at least occasionally. Most non-developers use their own language
anyway, but I think a large percentage of developers tend to use
English because of their long computer usage background. Having
developers using non-English would contribute to testing i18n in
hardy, and help translation volunteers like me. Most translators are
not software developers, so they don't understand what's causing
problems, so developers should take care of proper i18n (translators
should also file more bugs though!).

If you see English in your default installation of Ubuntu in a
well-translated language, it's always a i18n bug in the package or
some translations being not fetched from Rosetta or not available at
all in Rosetta. Ubuntu-specific changes often render packages not
completely translated.

Packages below are the ones causing problems for people not using
English at the moment in hardy, to my knowledge. Please add if you
know more highly visible i18n problems.

- hwtest-gtk: not translatable in Rosetta, lacks .desktop entry's
Ubuntu tag for its translation too.
https://bugs.launchpad.net/ubuntu/+source/hwtest/+bug/202447

- firefox: menu entry translations lost since firefox 3.
https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/192333

- firefox: completely untranslated also otherwise, though on a
positive note ubufox translations are finally accepted in and no doubt
language support packages are being made for the final release.

- brasero: lacks Ubuntu tag in .desktop file preventing menu item
translation. https://bugs.launchpad.net/ubuntu/+source/brasero/+bug/200532

- gnome-control-center: patch not yet landed that makes the new
randr-1.2 resolution tool translatable - until it has, I'm not certain
if it fixes all the issues or only some. discussed on ubuntu-x mailing
list, I provided a patch that should be a good start at least (the
strings in .glade file still need additional tweaking to not add new
strings, though new strings can also be translated if the new version
of control-center is uploaded fast enough)

- network-manager-applet: does not use X-Ubuntu-Gettext-Domain like
also some others.
https://bugs.launchpad.net/ubuntu/+source/network-manager-applet/+bug/201328
(also solved if the both, duplicate menu items are removed...)

- transmission: not translatable in Rosetta, even though in main.
https://bugs.launchpad.net/rosetta/+bug/201334

- ubiquity: partitioning window still lacks i18n love.
https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/144741

- ubiquity: language support installation silently fails if no network
connection (should guide a person to enable network and maybe
otherwise increase information available).
https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/81683

- example-content: Examples directory / link does not have a
translation solution. https://bugs.launchpad.net/baltix/+bug/45489

(in default installation in earlier Ubuntus, so relatively important
too for many people: - serpentine: lost .desktop translations.
https://bugs.launchpad.net/ubuntu/+source/serpentine/+bug/200539)

- help  support main page is not translated. fix committed, thanks!

Anyway, thanks for contributions so far to make Ubuntu 8.04 the best
localized Ubuntu so far! Each release has been getting better, though
each release always seems to have its own problems, too.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Reminders: 1. UI freeze == string freeze, notifications needed for translators. 2. please remember i18n.

[Timo Jyrinki]

2008/3/8, Scott Ritchie [EMAIL PROTECTED]:
 Does string freeze include changes to package descriptions?

Not sure if I'm the correct person to answer this, but I'd say no.
Package description translations is not what Ubuntu translators do
every day, or at least Ubuntu doesn't offer a framework like Rosetta
to do those. I personally think package description translations
should mostly be done in Debian anyway.

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Deprecating slocate for desktop users?

[Timo Jyrinki]

Chris Jones kirjoitti:
 Speaking for myself, I regularly use slocate to find things that are
 outside my home directory (not that I use tracker for things that are in
 my home directory - I put them where they are, so I know where they are ;)

I also use slocate sometimes, and I don't use tracker myself at the
moment since I know where my files generally are.

But: if The GUI search tool nowadays is tracker, and any ordinary person
basically only searches for files inside the home directory (and if not,
they should go to Indexing Preferences to set additional directories for
tracker), and no-one uses slocate except for those who actually know it
beforehand, why include it by default and cause daily hard disk churn
for every Ubuntu user? Anyone who knows the slocate tool, and command
line in general, can apt-get install slocate at will.

And the slowdown is not just daily, ie. sometime at night, since
people don't generally have their computers on 24h/day. Basically it
happens every time the computer is started on a new day. cron.daily is
run, which includes running slocate on the whole hard disk even though
99%+ of the users probably never utilize the database generated by it.

So still, I argue that slocate should be _at least_ moved to
cron.weekly, with the additional steps I'd hope for too:

1. move to cron.monthly instead of cron.weekly
2. switch from slocate to mlocate
3. remove mlocate/slocate dependency from Ubuntu default desktop
installation (leave it on server installations)

-Timo

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Raising i18n awareness among developers

[Timo Jyrinki]

Hello,

I'd like to raise some internationalization issues Ubuntu has faced and
is facing currently. As a translator and user of Ubuntu in my native
language I've sometimes felt that it would be useful to have better
awareness of i18n issues among the developers in general. I hope this
will help people to remember that when doing any new stuff, keep i18n in
mind and tested.

I also hope this will help finding people to fix not just the bugs
(which are usually not _that_ bad) but the wider issues I describe
towards the end of this post. Maybe there'd even be some hired person(s)
specifically looking at these kind of issues in the future.

Contents of this post: History, Gutsy, Wider Issues

== History ==

As some history, we've, for example, had a bug in Rosetta preventing
complete translations at a release time until edgy [1], and several bugs
in each released version about non-complete i18n of applications and
various other aspects like installer. For example at least [2], [3],
[4], [5], [6] in Feisty (Feisty is better than any release before,
though!). Also Ubuntu documentation translation updates haven't yet been
done for Feisty, they were supposed to be done because there was so
little time to translate documentation for Feisty. Hopefully the
problems [7] can be resolved quickly (there's reportedly a problem with
generating the docs).

[1] https://bugs.launchpad.net/rosetta/+bug/102382
[2] https://bugs.launchpad.net/rosetta/+bug/106756
[3]
https://bugs.launchpad.net/ubuntu/+source/software-properties/+bug/103292
[4] https://bugs.launchpad.net/ubuntu/+source/python-apt/+bug/103917
[5] https://bugs.launchpad.net/ubuntu/+source/bluez-gnome/+bug/95796
[6] https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/45741
[7] https://bugs.launchpad.net/ubuntu/+source/ubuntu-docs/+bug/123963

== Gutsy ==

Currently we have some problems with gutsy, besides the fact the
installer is not translatable yet at all which is being fixed. Hopefully
[8] too, which is also in Feisty. The installer currently doesn't
install support for any language besides English [9], plus both on the
live CD and in an installed Ubuntu Examples folder isn't translated
[10]. The new xdg-user-dirs currently also works suboptimally [11], of
which there's going to be further investigation indeed before the
release (thanks!), I hope that it will work in an optimal way in the
release.

I'm currently not seeing any actual applications that couldn't be
translated at all, besides problems with Rosetta lagging behind badly
and missing a feature to keep restricted manager translated at the
moment [12]. I do believe they'll be solved near/after the string freeze
and before release this time. For an example about a new Ubuntu program,
displayconfig-gtk seems very properly translatable though currently
translations have to be sent to the bzr repository until Rosetta catches up.

As a whole, gutsy is looking better than feisty again, which is great,
but there are big issues still left, read more below.

[8] https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/103925
[9] https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/131294
[10] https://bugs.launchpad.net/ubuntu/+source/example-content/+bug/45489
[11] https://bugs.launchpad.net/ubuntu/+source/xdg-user-dirs/+bug/123435
[12] https://bugs.launchpad.net/rosetta/+bug/130138

== Wider Issues ==

I don't doubt the automatic language support downloading wouldn't be
fixed to some extent soon, but most probably not optimally. That brings
me to the bigger problem of the difficulty of having Ubuntu in one's
native language:

(0. the Win32 part of the live cd doesn't have translations)

1. When starting Ubuntu from the live CD, it starts in English unless
the user (let's suppose a newbie) strikes F2 in the boot menu

2. When the desktop shows, even if the person chose non-English
language, generally support for any other language is not on the CD
(more languages are needed to be dropped in each release), and besides
the installer which includes all translations most stuff is in English

3. In the installer, it's not indicated that the person should _really_
have network connection enabled for the language support to be
downloaded, plus even with the connection it's currently broken like
stated ([[9])

4. Before or after installation, if the user happens to find Language
Selector, and has the default language as one's native language (ie.
selected in the boot menu or in the installer if already installed), the
selector does not suggest installing support for the Default Language
selected if there was no network connection during installation [13]

[13] https://bugs.launchpad.net/ubuntu/+source/language-selector/+bug/135752

To fix the numbered problems, besides including instructions for 1. and
3. in our ubuntu-fi.org's installation guide [14] that hopefully some
Finnish people read before installing, I've done one blueprint [15] to
collect ideas and other blueprints.

The blueprint was briefly discussed