[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-release-team
Subject:    Re: [Nepomuk] kde-4.4, virtuoso-6.1.0, and virtuosoconverter
From:       Sebastian Trueg <trueg () kde ! org>
Date:       2010-02-04 22:18:18
Message-ID: 4B6B47AA.9050106 () kde ! org
[Download RAW message or body]

Sebastian Kügler wrote:
> On Thursday 04 February 2010 16:16:34 Tom Albers wrote:
>> Op Thursday 4 February 2010 16:02 schreef u:
>>> On 02/04/2010 07:43 AM, Sebastian Trueg wrote:
>>>> Rex Dieter wrote:
>>>>> FYI,
>>>>> http://kde-apps.org/content/show.php/Nepomuk+Virtuoso+Converter
>>>>>
>>>>> 1.  The requirements for kde-4.4/nepomuk have changed pretty late in
>>>>> the game (after all 4.4rc's).
>>>> No, they did not. I always said that Virtuoso 6 would be the dependency
>>>> for KDE 4.4.
>> You can not add a new dependency for KDE 4.4 the day before the release
>> without testing. I consider it a new feature which is not allowed since
>> ages. for KDE 4.5 I've planned to add it to the release schedule
>> explicitly to make it even more clear. It is too late in the cycle for 4.4
>> to change this dependency. Really. Some packagers can not add a new
>> dependency to their repository in one day and have prepared the KDE 4.4
>> packaging based on the 4.4 release candidates we have given them. Please
>> don't do this.
>>
>>> Fact is that we're in a bad situation of now requiring something that's
>>> been untested and it's data is incompatible with what was tested.
>> +1
>> Unacceptable imho.
> 
> What will happen is that distros will ship Virtuoso 6 anyway, because they don't want 

That has been the plan for a long time now and the major distros already
knew that. Like I said: Debian, SuSE and Mandriva have been in contact
with OpenLink for a long time to clean up the code tree (licence-wise)
and figure out the best packaging scheme.

> to introduce V5, and then have to convert users' data again (first time from the 
> sesame2) backend. I don't know in how far V6 is different from V5, other than the 
> dataformat, but we might consider recommending V6 anyway, because it has a clearer 
> road forward.
> 
> It's a bit unfortunate that there was a bug in V6.0 that prevented us from testing 
> with it, that seems fixed in 6.1?
> 
> I'd also be interested in more (dis)advantages of V6 over V5, and how different V5 
> and 6 actually are from a "what can break KDE?"-point of view. Sebastian, can you 
> give more details?

From my point of view there are no disadvantages, only advantages.
Better performance, lower mem footprint, more features, better support, ...

>>>> Anyway, the converter is hacky tool allowing early adopters of KDE 4.4
>>>> who already tries Nepomuk with Virtuoso 5 (and thus, compile
>>>> themselves!) to easily convert their data.
>> Distro's who have provided RC's to their users are now more or less obliged
>> to run that script. Virtuoso was made a hard dependency (at least for
>> anyone compiling kdepim) so all svn users also need to update.
> 
> Bottom line is that distros will probably shop Virtuoso 6 to their end-users anyway, 
> so we need to make sure it works well with it. I don't think it's as easy as saying 
> "it's too late to require V6", because it's not. Note that I don't like the 
> situation, but to me it's more complex ...
>
> There's also the question what we require. As far as I know, KDE 4.4.0 works with 
> virtuoso 5 and 6, so what would be natural is requiring at least V5, recommending V6 
> (for the upgrade path, BUT we haven't tested it yet).

And that is exactly what is happening. It works with both v5 and v6 but
nobody will package v5. Thus the converter.

Cheers,
Sebastian
_______________________________________________
release-team mailing list
release-team@kde.org
https://mail.kde.org/mailman/listinfo/release-team

[prev in list] [next in list] [prev in thread] [next in thread]