Disappointed: My feature was removed without notifying me

Toshio Kuratomi a.badger at gmail.com
Wed Oct 8 19:20:30 UTC 2008


Christoph Wickert wrote:
> Am Mittwoch, den 08.10.2008, 12:38 -0400 schrieb Bill Nottingham:
>> Kevin Kofler (kevin.kofler at chello.at) said: 
>>> Bill Nottingham <notting <at> redhat.com> writes:
>>>> There are descriptions which are translated as well....
>>> Well, leave it blank, leave it in English (untranslated), put just a URL for 
>>> the upstream web page as the description
>> ... or, actually follow the string freeze, and the string freeze break
>> policy.
> 
> The first constructive statement I hear from a FeSCo member. In fact you
> are the _only_ FeSCo member who responded to my mail at all.
>>> too, ... - anything is better than having a set of packages all packaged and 
>>> reviewed, but hidden from users (not listed in comps, and rejected as a feature 
>>> on the ground that it isn't in comps).
>> I'm sorry, but I'm missing the outrage -
> 
> I will try to explain it to you, please read below.
> 
>>  the feature page listed the
>> following things, under 'Scope':
>>   
>>   * Package Reviews  (were not fully done)
> 
> But that's not my fault I guess.

There is no fault here... there is only work done or not done.  If the
Feature says that the packages are needed for the feature to be complete
and they are not reviewed then the feature is incomplete.

Steps you could take:
  * Get other lxde interested packagers to perform reviews
  * Trade reviews with people
  * If the packages are "very nice to have but not truly required" you
could drop them from the list of packages that must be in for the
feature to be complete.

>>   
>>   * Comps: new group with id "lxde-desktop", name "LXDE"...
>> 	o mandatory: lxde-common, lxpanel, openbox, pcmanfm
>> 	o default: gpicview, leafpad, lxappearance, lxtask, lxterminal, obconf, xarchiver
>> 	... (wasn't done)
> 
> Wasn't done because
>       * of the missing reviews
>       * I expected someone from FeSCo to give me a go.
> 
This would be miscommunication, then.  AFAIK, comps is open to
maintainers to make changes to.  Maintainers are expected to do that.

>> and, under 'Test Plan':
>>   
>>   As most of the packages are already in fedora they are proven to be stable.
> 
> Sorry to interrupt here, but this is important: It was well documented
> that there already LXDE packages in Fedora for quite a while. There also
> is a detailed list of which package is available for what release(s).
> 
>>  So we need to
>>   
>>   1. get the 2 missing packages into rawhide ASAP so they are included in the F10 Beta
>>   2. add packages to comps
>>   3. test that installation with yum-groupinstall works correctly
>>   4. test fresh installs with the beta 
>>
>> It wasn't even to having step 1 completed. So, exactly what else is there to
>> do when even the basic scope and test plan of the feature isn't ready?
> 
> Bill, I _do_ understand that my feature has been dropped because it
> is/was not testable and it breaks the string freeze, but I can not
> accept the way _how_ this decision was made. Let me quote the complete
> discussion about LXDE:
> 
>> bpepple: Feature: https://fedoraproject.org/wiki/Features/LXDE
>> nirik: for lxde, sadly 2 packages are still in review... so I think it needs to go to 11 (ha ha).
>> dgilmore: so not testable
>> bpepple: nirik: do we know what's holding up those 2 reviews?
>> nirik: unless those packages aren't critical for testing, but I think they are.
>> nirik: bpepple: I glanced at them and there was a lot of discussion about what should be a subpackage, etc.
>> wwoods: none of the LXDE packages are even built, AFAICT
>> nirik: wwoods: some are, but thats because they were already in... like openbox is the windowmanager.
>> notting: well, if they try and add the comps group now, they'll be breaking the string freeze
>> nirik: yeah, so I think this should move back to next release... if they get it sooner great, we can trumpet it in 11
>> bpepple: nirik: +1
>> notting: +1
>> bpepple: anyone disagree with pushing LXDE, otherwise we can move on.
> 
> Obviously the only one who was at least a little informed is nirik, he's
> also the only one who apologized for "the lack of communication" in a
> conversation we had yesterday. But nirik is wrong too: Those LXDE
> packages that are already in Fedora are not there because they have been
> forever but because they were part of my feature (lxtask, lxterminal,
> lxlauncher, ...).
>  
> So the decision was made under the assumption that _none_ of the LXDE
> packages has ever passed a review. How can somebody who has read the
> feature page claim that? Why did nobody answer him back?
> 
> Summing it all up: My outrage is because
>       * nobody contacted my to ask me for status or to tell me the
>         feature was dropped. Nobody contacted me at all.

There's definitely a need for better communication.  There have also
been features dropped as not-a-feature this year that I thought shouold
have been better communicated so that there was a chance to rewrite the
proposal so it was a feature.

We should have a rule like:
 Communication happens on the Feature Discussion Page
or:
  Communication happens via private email between the feature owner and
the feature wrangler
or:
  Communication happens via the FESCo Meeting Summary

so that everyone knows that there's a one-stop shop to put out and
receive information.

>       * the wiki says [1]: "The feature owner is responsible for
>         watching any owned pages for state changes, using the wiki watch
>         feature." That's what I did, nevertheless I did not get
>         notified.

I've heard that mediawiki doesn't work like people expect for watching
pages.  I'll admit that I haven't watched pages since the change to
mediawiki so I'm not sure how this works.

>       * I think I did everything I could to rescue my feature. I someone
>         contacted me I would have responded.
>       * the decision was made by people who obviously did not read the
>         feature page and have no interest in the feature. If somebody
>         has not read the page he should not speak up on that topic or
>         decide about it, and if he has further question he should ask.

These are a bit unfair.  The Feature Page shows that the feature is not
done.  Checking bugzilla shows that the page is up-to-date in regards to
the package review status.  Beta is a deadline for features and that has
come and gone.  So the Feature is plainly not completed whether you were
contacted or not; whether the people who commented knew all the
particulars or only some.

Can you be angry that communication wasn't good enough?  Sure.  But it
works both ways -- if the packages can't be reviewed in time but they
aren't necessary why not be proactive and take them off the list?  If
you feel you need them and they can be reviewed after Beta but before
final, why not ask for an extension?  If you're waiting for FESCo to
approve the new comps group, why not ask FESCo if you can move ahead on
that portion?

FESCo members should be reading the feature page but they will never be
able to be intimately familiar with all the Features being proposed.  In
other words, we can't expect them to know any more about a feature than
what is written on the Feature Page.  If you look at the Feature Page
and it looks like the Feature is not completed or doesn't express what
the Feature really is or leaves out something that might lead FESCo to
think it's not a feature it needs to be clarifies on the feature page so
they can understand it better.

>       * no FeSCo member except from Bill reacted to my previous mails.

heh.  I interpret that as a bunch of polite people not jumping in to say
 "Me too" but it could be frustrating.

-Toshio

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: OpenPGP digital signature
URL: <http://listman.redhat.com/archives/fedora-devel-list/attachments/20081008/71df882f/attachment.sig>


More information about the fedora-devel-list mailing list