[lsb-discuss] Thinking about future LSB features

Wichmann, Mats D mats.d.wichmann at intel.com
Wed Feb 25 09:07:51 PST 2009

Theodore Tso wrote:
> On Wed, Feb 25, 2009 at 08:21:22AM -0700, Wichmann, Mats D wrote:
>> I think we ought to watch PackageKit carefully, as it looks like
>> it will provide *some* of what we've been asked about.  At the
>> moment the PackageKit API is not stable (the authors say so: there's
>> no promise of complete API stability until 1.0.0), and in an informal
>> discussion on irc the other day it was noted it's not well accepted
>> in the debian variants, not least because it follows the rpm
>> design philosophy of allowing zero interaction during the install
>> process.
> I actually happen to agree with that no-questions-at-install-time
> philosophy, but I wonder how strong of an issue it really is in
> practice, especially when the most common use of Packagekit is likely
> going to be an application which is interested in installing a
> plugin, or codec, or font. 
>> There's a lot of resistance within Ubuntu, it looks like -
>> I'm reading stuff where Ubuntu people consider the PackageKit UI
>> really horrible...
> UI's can be fixed, or customized; though.  And mabe part of the
> problem is that the backend support for dpkg isn't as good as it ought
> to be (but that requires the developers from the Debian derivitives to
> become more involved).  Heck, even the UI issue can be solved by
> having a dialog box pop up so that the user can answer the package's
> questions three --- and if the user is annoyed by the package stopping
> and asking for the air-speed velocity of an unladen swallow, well,
> that's the fault of the package, not of PackageKit.  :-)
> In any case, these aren't things we can settle, but we can encourage
> people to work together on whichever projects has the most momentum,
> which at the moment seems to be PackageKit.  Hopefully the Debian
> variants will also participate and help make PackageKit better for
> Debian's design points. 

Right... all of this I agree with, I was just trying to reinforce
the "looks useful, but not ready for LSB" position, plus suggest
that we don't just wait, but be a little more active in pushing
people to figure out if this is a direction everybody wants to go in.

African, or European?

More information about the lsb-discuss mailing list