[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: RFS: sanduhr (restoring removed package)

scripsit Neil Williams:
> On Mon, 27 Aug 2007 12:26:28 -0700
> Thanasis Kinias <tkinias@kinias.org> wrote:
> > Dear mentors,
> > 
> > I am looking for a sponsor for the new version 1.93-2 of my package
> > "sanduhr" (currently removed from unstable).
> This just isn't good enough - not by a long way. Packages are not
> removed from unstable without due cause so be OPEN about the cause -
> this is open source, this is Debian - "we don't hide problems".

I certainly have no desire to be opaque; I erred on the side of
brevity as I thought an excessively verbose RFS would be ill-received.
I see that I was most profoundly mistaken, and I will try to provide 
all information that may be helpful.

> At the absolute minimum you *must* include the bug numbers that caused
> the package to be removed from unstable in the first place AND specify
> those in the list of bugs that would be closed because when a package
> is reintroduced, the bugs that were closed by the removal become
> relevant again! 

The reason for removal as cited in the package developer's info page
<http://packages.qa.debian.org/s/sanduhr.html> was bug 360269
<http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=360269>.  This was the
bug by which the developer and previous maintainer orphaned the package
on 2006-03-31.  On 2007-07-16 the bug was changed to `RM: sanduhr --
RoQA; abandoned upstream, orphaned for over a year'.  On 2007-07-20,
Mohammed Adnène Trojette posted to the bug:

1] On Mon, Jul 16, 2007, Mohammed Adnène Trojette wrote:
1] > reassign 360269 ftp.debian.org
1] OK, I may have been a bit quick on this one: today, Clint has uploaded
1] packages for sanduhr. Shall I reassign the bug to wnpp?

and Holger Levsen replied:

2] Hi,
2] On Monday 16 July 2007 02:00, Mohammed Adnène Trojette wrote:
2] > OK, I may have been a bit quick on this one: today, Clint has uploaded
2] > packages for sanduhr. Shall I reassign the bug to wnpp?
2] I'd say so, yes.

However, this was never done and on 2007-07-22 ftpmaster deleted the
package from unstable and closed the bug.

Earlier today (2007-08-27) I unarchived and reopened the bug, changed it
to ITA and took ownership, and merged it with the ITP I had filed on
sanduhr.  That is, I merged 360269 (the reason for the package's
removal) with 439824 (my ITP), which my package closes.  I see now that
I ought to have included a clearer explanation of that in the RFS; I
suppose to one without access to the inside of my head it mightn't have
been nearly as clear as it seemed to me :(

> This package had two RC bugs before removal - what have you done to
> ensure that these are fixed and that the package will not introduce
> new ones?

I may be misreading the bug report
but that is not what I understand it to say.  There have only ever been
two RC bugs against the package -- 98795 (fixed in 0.3.52-1 on
2001-05-28) and 139954 (fixed in 0.99-2 on 2002-03-27).  There were four
open bugs at the time of the packages removal:  one Normal (283548,
which I cannot reproduce with the current version) and three Wishlist.

If I am not reading the bug report correctly, please tell me where I'm
going wrong and accept my apologies for being contrary.

> Too many RFS emails are still blindly following the template without
> any thought.

That is regrettable.  I apologize; I quite honestly believed that the
template was provided precisely because we were meant to follow it quite
narrowly.  I will not make that mistake in the future.

> Sponsors are not a dumping ground for bad packages that have already
> been removed once - if you want a sponsor to help you (and yes, being a
> sponsor is entirely voluntary) you have to do some work. Sending an
> RFS template with no more data than could have been included from a
> 10 line shell script is woeful.
> If you would like me to sponsor this package you are going to have to
> do a LOT more to make up for such a bad start. Frankly, I'm not sure
> that the package deserves to be back in Debian so, please, justify your
> request and try to persuade me. 

I would submit that the package should be in Debian because:

1) It was been in Debian from 2001-05-24 until 2007-07-22, during which
   time the only RC bugs were a failure to build on m68k and an
   improperly specified text encoding on an omf file.  While it is not
   widely used, it had circa 50 popcon votes and circa 180 popcon insts
   prior to removal -- so I'm not the only person who uses it -- and the
   package has no history of causing any security or other problems for

2) It is GPL licensed and thus a good fit with Debian's social contract.

3) It is, in this version, a GNOME2 application, well-integrated with
   Debian's default desktop (including the GNOME help system).

4) It perfectly embodies the UNIX philosophy of using simple,
   well-designed tools to do specific tasks, and while this philosophy
   is not fashionable in this age of 250MB do-everything-in-one-suite
   software design, that is not a reason to exclude it from Debian.

> Bear in mind that your RFS has led to you starting from a deeply
> negative position so your responses had better be very good. I'm not
> going to do your work for you on this one, work out what you should
> have done from the documentation that already exists.

It grieves me that my RFS has so profoundly offended you; such was most
definitely never my intent.  As I mentioned above, my brevity was an
ill-conceived attempt to be considerate by not wasting anyone's time
with needless verbosity.  I hope that my above explanations are more

If I am still lacking, I would be grateful if you could point me toward
the relevant documentation from which I could learn what I need.  Other
than the Debian New Maintainer's Guide
<http://www.debian.org/doc/maint-guide/>, Debian Policy Manual
<http://www.debian.org/doc/debian-policy/>, and the mentors.debian.net
Introduction for Maintainers
<http://mentors.debian.net/cgi-bin/maintainer-intro>, with all of which
I am familiar, I do not believe I have found any documents which deal
with policies and procedures for packaging and package maintenance.

Best regards,
Thanasis Kinias
Doctoral Candidate, Department of History, and
  Instructor, Professional Enhancement Programs
Arizona State University, Tempe, Arizona, U.S.A.
Je ne viens d'aucun pays, d'aucune cité, d'aucune tribu.  Je suis fils de la
route, ma patrie est caravane, et ma vie la plus inattendue des traversées.
  -- Amin Maalouf, _Léon l'Africain_

Attachment: signature.asc
Description: Digital signature

Reply to: