Fedora 18 issues with translations and keymaps

Robyn Bergeron rbergero at redhat.com
Thu Jan 3 17:24:20 UTC 2013


On 01/03/2013 08:05 AM, "J├│hann B. Gu├░mundsson" wrote:
> On 01/03/2013 02:09 PM, Josef Skladanka wrote:
>> I'm absolutely +1 with jeischmann and jreznik on this matter.
>> Keyboard layouts are broken almost every other (if not every one) 
>> Fedora release. The same exact encrypt-prompt issue was (at least) in 
>> F16, and we released it anyway.
>> I'm not saying that we should just blindly say "it's late, let's ship 
>> it", but quite a few of these bugs (which adamw mentioned) IMHO are 
>> either present (not saying the exact ones, but the 'type' of the 
>> bugs) in Fedora for quite some time already, or just are not clear 
>> show-stoppers.
>
> Which to me makes it more urgent to become a showstopper.
>
> Ignore that broken behavior and just releasing has not gotten those 
> thing properly fixed now has it...
>
> JBG
I think the main thing here that adam was getting at is not the 
"individual bugs that we keep saying we'll fix and have accepted at 
release in the past" (but never seem to, as an aside point), but more of 
the sum of all things combined.  Yes, the release criteria provides a 
bit of wiggle room here, but there is also guidance about using 
judgement about things like # of users potentially affected, the ease of 
fixing things for both knowledgable users and uninformed users, etc.

My concern here is mostly that we all keep to be pointing to workarounds 
but I question how many of these workarounds are going to basically land 
someone on the doorstep of another language-related bug. ie: "oh, you 
can work around the encryption issue by doing X, except X doesn't work 
because your translation isn't done," etc.  And this is what adam more 
or less pointed out in 889562.

The fact that bengali is 33% done being translated does jump out at me a 
bit, as does chinese defaulting to the US keyboard.  These are not 
insignificant populations.

Most of the indic languages defaulting to the guajarati keyboard is 
something we might look at - I almost wonder if that is somehow 
intentional, all of the languages that are affected seem to support 
inscript. (It may be related to the 
http://fedoraproject.org/wiki/Features/Inscript2_Keymaps feature, or 
not, but it might be worth asking the feature owner for any useful 
insight he might have about keyboard usage.)  But if it's not 
intentional and it's screwing things up, well, that's a pretty 
significantly large population as well.

-r




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.fedoraproject.org/pipermail/test/attachments/20130103/b2c9219b/attachment.html>


More information about the test mailing list