[Testing] Testing Summary, Auckland - 9 June 2012

[Chris Leonard]

On Mon, Jun 18, 2012 at 6:54 PM, Daniel Drake <dsd at laptop.org> wrote:
> We're expecting a quite major WebKit modernisation for the next major
> release. Having seen WebKits progress over the last few months I think
> these kinds of issue will decrease and decrease as we move forward.

Having lurked on their IRC channel,  I concur that WebKitGTK+ seems to
be moving to address issues quickly.

There is one area where I have a lingering concern about WebKitGTK+
and that is in the realm of localization (i18n/L10n).  There has been
a longstanding issue with WebKitGTK+ generating their POT file and
getting it updated in it's L10n home on the Gnome Damned Lies server

You can see the errors reflected on this page:
http://l10n.gnome.org/module/webkit/

I bugged the WebKitGTK+ folks about this and got them to land some PO
files that had been awaiting commit for a while.  mrobinson and kov
were actually quite helpful and friendly, but the real issue looks to
be upstream of WebKitGTK+

The root cause of this apparently traces upstream to an intltool issue:

https://bugs.launchpad.net/intltool/+bug/823217

But I have had no success in getting the intltool developer to priortize it.

In general, localizers are staying away from working on WebKitGTK+
until the POT generation errors get resolved.   If we want good
upstream L10n bits (and I am pretty sure we do), it would be helpful
if a developer could investigate how hard a fix this would be to land
in intltool.

cjl