Don't report new bugs for Rmail??

Don't report new bugs for Rmail??

[Eli Zaretskii]

If you go to http://emacsbugs.donarmstrong.com/, make sure "package"
is checked in the "Find bugs by:" selector, type "rmail" into the
"What to search for:" field, and click the "Find" button, you will see
a page with the following on its 2nd line:

  There is no maintainer for rmail. Please do not report new bugs against this package.

??? Why does it ask not to report new bugs?

Re: Don't report new bugs for Rmail??

[Don Armstrong]

On Sat, 31 Jan 2009, Eli Zaretskii wrote:
> If you go to http://emacsbugs.donarmstrong.com/, make sure "package"
> is checked in the "Find bugs by:" selector, type "rmail" into the
> "What to search for:" field, and click the "Find" button, you will see
> a page with the following on its 2nd line:
> 
>   There is no maintainer for rmail. Please do not report new bugs against this package.

Because there is no Rmail package.

If someone wants one, they kind of need to ask.


Don Armstrong

-- 
Where I sleep at night, is this important compared to what I read
during the day? What do you think defines me? Where I slept or what I
did all day?
 -- Thomas Van Orden of Van Orden v. Perry

http://www.donarmstrong.com              http://rzlab.ucr.edu

Re: Don't report new bugs for Rmail??

[Eli Zaretskii]

> Date: Sat, 31 Jan 2009 02:44:03 -0800
> From: Don Armstrong <don@donarmstrong.com>
> 
> On Sat, 31 Jan 2009, Eli Zaretskii wrote:
> > If you go to http://emacsbugs.donarmstrong.com/, make sure "package"
> > is checked in the "Find bugs by:" selector, type "rmail" into the
> > "What to search for:" field, and click the "Find" button, you will see
> > a page with the following on its 2nd line:
> > 
> >   There is no maintainer for rmail. Please do not report new bugs against this package.
> 
> Because there is no Rmail package.

Sorry, I'm not following: if there's no Rmail package, how come the
bug tracker finds bugs by that package?  What am I missing?

Re: Don't report new bugs for Rmail??

[Juanma Barranquero]

On Sat, Jan 31, 2009 at 12:19, Eli Zaretskii <eliz@gnu.org> wrote:

> Sorry, I'm not following: if there's no Rmail package, how come the
> bug tracker finds bugs by that package?  What am I missing?

At the moment, "packages" in the Emacs bugtracker are just arbitrary
labels. The tracker allows searching for these labels.

A package, apparently, is a label plus some other data, like a
maintainer name & address, etc. There's no package defined for Emacs.

    Juanma

Re: Don't report new bugs for Rmail??

[Eli Zaretskii]

> Date: Sat, 31 Jan 2009 12:28:41 +0100
> From: Juanma Barranquero <lekktu@gmail.com>
> Cc: Don Armstrong <don@donarmstrong.com>, emacs-devel@gnu.org
> 
> On Sat, Jan 31, 2009 at 12:19, Eli Zaretskii <eliz@gnu.org> wrote:
> 
> > Sorry, I'm not following: if there's no Rmail package, how come the
> > bug tracker finds bugs by that package?  What am I missing?
> 
> At the moment, "packages" in the Emacs bugtracker are just arbitrary
> labels. The tracker allows searching for these labels.

It's a bad thing to call "package" to two different things.  How about
calling the labels "labels" instead?

> A package, apparently, is a label plus some other data, like a
> maintainer name & address, etc. There's no package defined for Emacs.

Fine, but the "don't submit bug reports" request is misleading and
should be removed or rephrased, IMO.

Re: Don't report new bugs for Rmail??

[Don Armstrong]

On Sat, 31 Jan 2009, Eli Zaretskii wrote:
> From: Juanma Barranquero <lekktu@gmail.com>
> > On Sat, Jan 31, 2009 at 12:19, Eli Zaretskii <eliz@gnu.org> wrote:
> > > Sorry, I'm not following: if there's no Rmail package, how come
> > > the bug tracker finds bugs by that package? What am I missing?

Because there are bugs assigned to that package.

> > At the moment, "packages" in the Emacs bugtracker are just
> > arbitrary labels. The tracker allows searching for these labels.

They're not "just arbitrary labels".
 
> It's a bad thing to call "package" to two different things. How
> about calling the labels "labels" instead?

Because they're not labels, they're packages. Debbugs allows people to
assign bugs to packages that don't have a maintainer, but suggests
that people not do so, because in the typical case it means that no
one looks at them.

If rmail (or anything else) is going to be used as a valid package to
assign bugs to, it should have a maintainer, and whoever wants it
created should tell me that they want it created and what the
maintainer is.

> > A package, apparently, is a label plus some other data, like a
> > maintainer name & address, etc. There's no package defined for
> > Emacs.

There is a package defined for emacs. There is also one for bzr,
aquamacs, octave, and a few other things. There isn't one for rmail.

> Fine, but the "don't submit bug reports" request is misleading and
> should be removed or rephrased, IMO.

Bug reports against packages that don't have a maintainer shouldn't be
submitted, because there's no indication that someone is going to look
at them. In this case, the error is that no one has asked for the
rmail package to be created.

There's no difference between rmail and the foobarbaz package in this
case.


Don Armstrong

-- 
One day I put instant coffee in my microwave oven and almost went back
in time.
 -- Steven Wright

http://www.donarmstrong.com              http://rzlab.ucr.edu

Re: Don't report new bugs for Rmail??

[Eli Zaretskii]

> Date: Sat, 31 Jan 2009 13:19:29 -0800
> From: Don Armstrong <don@donarmstrong.com>
> 
> On Sat, 31 Jan 2009, Eli Zaretskii wrote:
> > From: Juanma Barranquero <lekktu@gmail.com>
> > > On Sat, Jan 31, 2009 at 12:19, Eli Zaretskii <eliz@gnu.org> wrote:
> > > > Sorry, I'm not following: if there's no Rmail package, how come
> > > > the bug tracker finds bugs by that package? What am I missing?
> 
> Because there are bugs assigned to that package.

But you just said that a package named "rmail" does not exist, didn't
you?

> If rmail (or anything else) is going to be used as a valid package to
> assign bugs to, it should have a maintainer, and whoever wants it
> created should tell me that they want it created and what the
> maintainer is.

Rmail's maintainer is "FSF".  Which means that Emacs maintainers are
collectively responsible for maintaining Rmail.

> > Fine, but the "don't submit bug reports" request is misleading and
> > should be removed or rephrased, IMO.
> 
> Bug reports against packages that don't have a maintainer shouldn't be
> submitted

Well, that request can easily be interpreted as meaning that bug
reports against Rmail are not welcome.  I think this is not what we
want to say.

> In this case, the error is that no one has asked for the rmail
> package to be created.

It's IMO inconsistency to allow bugs be filed against a package that
doesn't exist.

Re: Don't report new bugs for Rmail??

[Don Armstrong]

On Sat, 31 Jan 2009, Eli Zaretskii wrote:
> But you just said that a package named "rmail" does not exist,
> didn't you?

As far as the BTS is concerned, it doesn't exist. Bugs are allowed to
be assigned to packages which do not exist.
 
> Rmail's maintainer is "FSF". Which means that Emacs maintainers are
> collectively responsible for maintaining Rmail.

Doesn't really matter to me who is responsible for it; I just need a
mailing address.

> Well, that request can easily be interpreted as meaning that bug
> reports against Rmail are not welcome. I think this is not what we
> want to say.

It is actually what is meant. If a package doesn't exist, people
shouldn't file bugs against it. The problem here is that the package
*should* apparently exist, and no one has bothered to communicate that
fact.

> It's IMO inconsistency to allow bugs be filed against a package that
> doesn't exist.

The alternative is rejecting them entirely, which I think is the wrong
decision. The reports are allowed to be filed on the presumption that
somone later will look at the bug reports and assign them to the
appropriate package.


Don Armstrong

-- 
A citizen of America will cross the ocean to fight for democracy, but
won't cross the street to vote in a national election.
 -- Bill Vaughan

http://www.donarmstrong.com              http://rzlab.ucr.edu

Re: Don't report new bugs for Rmail??

[Eli Zaretskii]

> Date: Sat, 31 Jan 2009 14:24:01 -0800
> From: Don Armstrong <don@donarmstrong.com>
> 
> > It's IMO inconsistency to allow bugs be filed against a package that
> > doesn't exist.
> 
> The alternative is rejecting them entirely, which I think is the wrong
> decision.

All I care about is that sentence asking not to submit more bug
reports.  Can we please remove or rephrase that?

Re: Don't report new bugs for Rmail??

[Don Armstrong]

On Sun, 01 Feb 2009, Eli Zaretskii wrote:
> All I care about is that sentence asking not to submit more bug
> reports. Can we please remove or rephrase that?

It just needs someone to tell me what address they want as the
maintainer for the rmail package; I don't care what address it is.
[Likewise for any other packages that people want to create.]


Don Armstrong

-- 
Do not handicap your children by making their lives easy.
 -- Robert Heinlein _Time Enough For Love_ p251

http://www.donarmstrong.com              http://rzlab.ucr.edu

Re: Don't report new bugs for Rmail??

[Don Armstrong]

On Sun, 01 Feb 2009, Eli Zaretskii wrote:
> > Date: Sat, 31 Jan 2009 20:33:12 -0800
> > From: Don Armstrong <don@donarmstrong.com>
> > 
> > On Sun, 01 Feb 2009, Eli Zaretskii wrote:
> > > All I care about is that sentence asking not to submit more bug
> > > reports. Can we please remove or rephrase that?
> > 
> > It just needs someone to tell me what address they want as the
> > maintainer for the rmail package; I don't care what address it is.
> 
> Please use emacs-devel@gnu.org.

Done.
 

Don Armstrong

-- 
Science is a way of trying not to fool yourself. The first principle
is that you must not fool yourself, and you are the easiest person to
fool.
 -- Richard Feynman "What is and What Should be the Role of Scientific
    Culture in Modern Society"; 1964

http://www.donarmstrong.com              http://rzlab.ucr.edu

Re: Don't report new bugs for Rmail??

[Glenn Morris]

Eli Zaretskii wrote:

>> It just needs someone to tell me what address they want as the
>> maintainer for the rmail package; I don't care what address it is.
>
> Please use emacs-devel@gnu.org.

I think that should instead be bug-gnu-emacs@gnu.org.
Otherwise bug reports and notifications about rmail are going to start
appearing on emacs-devel rather than bug-gnu-emacs.

Re: Don't report new bugs for Rmail??

[Glenn Morris]

Glenn Morris wrote:

>> Please use emacs-devel@gnu.org.
>
> I think that should instead be bug-gnu-emacs@gnu.org.

Since we assign bugs to both the emacs _and_ the rmail package, does
this mean we will start getting two copies of things? I regret to say
I am confused.

Re: Don't report new bugs for Rmail??

[Glenn Morris]

Glenn Morris wrote (on Sun, 1 Feb 2009 at 21:20 -0500):

> Since we assign bugs to both the emacs _and_ the rmail package, does
> this mean we will start getting two copies of things? I regret to say
> I am confused.

I think Jason had it right:

    For other sub-packages that come under the umbrella of the emacs 
    developers, we have assigned bugs to "emacs,PACKAGE", so that the 
    sub-package can be used as a filter in the web interface, but the emacs 
    maintainers continue to be kept in the loop.

    [...]

    People don't file bugs against those packages, since the only address 
    for filing bugs is for emacs. But they might get reassigned later.

Re: Don't report new bugs for Rmail??

[Glenn Morris]

Glenn Morris wrote:

> I think Jason had it right:
>
>     For other sub-packages that come under the umbrella of the emacs 
>     developers, we have assigned bugs to "emacs,PACKAGE", so that the 
>     sub-package can be used as a filter in the web interface, but the emacs 
>     maintainers continue to be kept in the loop.
>
>     [...]
>
>     People don't file bugs against those packages, since the only address 
>     for filing bugs is for emacs. But they might get reassigned later.

Sorry for taking so many mails to figure this out, but that means we
_don't_ want rmail defining as a package. We _don't_ want bugs
reporting against "rmail", because these will not show up in the list
of "emacs" bugs.

As has been said, we are using packages just as labels for
administrative convenience. People don't need to worry about packages
at all when they report a bug. (People who know what they are doing
can use "emacs,rmail" to save us a bit of admin, but that need not be
expected.)

Yes, the web-pages are a bit misleading in this regard. But this is
just an aspect of bug#750 IMO.

Re: Don't report new bugs for Rmail??

[Glenn Morris]

Glenn Morris wrote:

> I think that should instead be bug-gnu-emacs@gnu.org.
> Otherwise bug reports and notifications about rmail are going to start
> appearing on emacs-devel rather than bug-gnu-emacs.
[...]
> Since we assign bugs to both the emacs _and_ the rmail package, does
> this mean we will start getting two copies of things?

I see that both of the above are true. Messages from the control
server about rmail are now cc'd to both emacs-devel and bug-gnu-emacs.
This is not what we want to happen.

I think the rmail maintainer should just be "unknown" again.
It was working perfectly fine that way.

Re: Don't report new bugs for Rmail??

[Don Armstrong]

On Mon, 02 Feb 2009, Glenn Morris wrote:
> Glenn Morris wrote:
> 
> > I think that should instead be bug-gnu-emacs@gnu.org.
> > Otherwise bug reports and notifications about rmail are going to start
> > appearing on emacs-devel rather than bug-gnu-emacs.
> [...]
> > Since we assign bugs to both the emacs _and_ the rmail package, does
> > this mean we will start getting two copies of things?
> 
> I see that both of the above are true. Messages from the control
> server about rmail are now cc'd to both emacs-devel and bug-gnu-emacs.
> This is not what we want to happen.

I've set it to bug-gnu-emacs@gnu.org again. You folks need to figure
out what you actually want to happen first; if you don't understand
the consequences, ask questions.


Don Armstrong

-- 
Three little words. (In descending order of importance.)
I
love
you
 -- hugh macleod http://www.gapingvoid.com/graphics/batch35.php

http://www.donarmstrong.com              http://rzlab.ucr.edu

Re: Don't report new bugs for Rmail??

[Eli Zaretskii]

> Date: Mon, 2 Feb 2009 11:18:37 -0800
> From: Don Armstrong <don@donarmstrong.com>
> 
> You folks need to figure out what you actually want to happen first;
> if you don't understand the consequences, ask questions.

Hey, that's unfair!  I said loud and clear what I wanted: that the
sentence discouraging people from submitting bug reports about Rmail
be removed or rephrased.  You ignored that request, and instead
insisted on an email address, "any email address" -- your own words!
Perhaps you should ask yourself questions about the consequences,
before you post such deceptively simple advice.

Re: Don't report new bugs for Rmail??

[Stefan Monnier]

> I've set it to bug-gnu-emacs@gnu.org again. You folks need to figure
> out what you actually want to happen first; if you don't understand
> the consequences, ask questions.

We don't need an email for the `rmail' package.  We just need for the
message not to say "don't report bugs for this package" and instead say
"report bugs for this package to the `emacs' package".


        Stefan

Re: Don't report new bugs for Rmail??

[Eli Zaretskii]

> From: Glenn Morris <rgm@gnu.org>
> Date: Mon, 02 Feb 2009 13:48:28 -0500
> Cc: emacs-devel@gnu.org
> 
> I think the rmail maintainer should just be "unknown" again.
> It was working perfectly fine that way.

I'm fine with that, if the sentence about not sending bug reports is
removed or rephrased.

Re: Don't report new bugs for Rmail??

[Stefan Monnier]

> I think the rmail maintainer should just be "unknown" again.
> It was working perfectly fine that way.

Indeed,


        Stefan

Re: Don't report new bugs for Rmail??

[Jason Rumney]

Don Armstrong wrote:
>> Rmail's maintainer is "FSF". Which means that Emacs maintainers are
>> collectively responsible for maintaining Rmail.
>>     
>
> Doesn't really matter to me who is responsible for it; I just need a
> mailing address.
>   

For other sub-packages that come under the umbrella of the emacs 
developers, we have assigned bugs to "emacs,PACKAGE", so that the 
sub-package can be used as a filter in the web interface, but the emacs 
maintainers continue to be kept in the loop.

> It is actually what is meant. If a package doesn't exist, people
> shouldn't file bugs against it.

People don't file bugs against those packages, since the only address 
for filing bugs is for emacs. But they might get reassigned later.

Re: Don't report new bugs for Rmail??

[Juanma Barranquero]

On Sun, Feb 1, 2009 at 07:22, Jason Rumney <jasonr@gnu.org> wrote:

> we have assigned bugs to "emacs,PACKAGE", so that the sub-package can be
> used as a filter in the web interface, but the emacs maintainers continue to
> be kept in the loop.

100% agreement.

    Juanma

Re: Don't report new bugs for Rmail??

[Don Armstrong]

On Sun, 01 Feb 2009, Jason Rumney wrote:
> For other sub-packages that come under the umbrella of the emacs
> developers, we have assigned bugs to "emacs,PACKAGE", so that the
> sub-package can be used as a filter in the web interface, but the
> emacs maintainers continue to be kept in the loop.

I think usertags and completely reassigning the bug to the other
package will serve you all better in the end, but that's your choice.

> People don't file bugs against those packages, since the only
> address for filing bugs is for emacs. But they might get reassigned
> later.

People can, actually. And file in this case also means reassign; you
get warned if you assign a bug to a package which does not exist.


Don Armstrong

-- 
More than any other time in history, mankind faces a crossroads.
One path leads to despair and utter hopelessness.
The other, to total extinction.
Let us pray we have the wisdom to choose correctly.
 -- Woody Allen

http://www.donarmstrong.com              http://rzlab.ucr.edu

Re: Don't report new bugs for Rmail??

[Stefan Monnier]

> As far as the BTS is concerned, it doesn't exist. Bugs are allowed to
> be assigned to packages which do not exist.

Indeed and we use that as labels by assigning bugs to emacs,<label>.
You mentioned "usertags" as another (better?) way to do the same, but
I don't know what that is.  Any URL?
 
>> Well, that request can easily be interpreted as meaning that bug
>> reports against Rmail are not welcome.  I think this is not what we
>> want to say.
> It is actually what is meant.  If a package doesn't exist, people
> shouldn't file bugs against it.  The problem here is that the package
> *should* apparently exist, and no one has bothered to communicate
> that fact.

I think the message should be changed to say that bugs should be
assigned to the `emacs' package instead.  I.e. we don't want to imply
that bugs shouldn't be reported, but just that they should be reported
elsewhere (to the "parent" package).


        Stefan

Re: Don't report new bugs for Rmail??

[Juanma Barranquero]

On Sat, Jan 31, 2009 at 22:19, Don Armstrong <don@donarmstrong.com> wrote:

> They're not "just arbitrary labels".

Packages are not arbitrary labels. "Packages" are. I said that
packages are "packages" (i.e., labels) plus other things.

What I was saying, in other words, is that you can assign a bug to any
arbitrary label, but that the label is not a real package unless the
package exists.

> Bug reports against packages that don't have a maintainer shouldn't be
> submitted, because there's no indication that someone is going to look
> at them. In this case, the error is that no one has asked for the
> rmail package to be created.

But still, you can search for the label "rmail" even if no package exists.

    Juanma

Re: Don't report new bugs for Rmail??

[Don Armstrong]

On Sun, 01 Feb 2009, Juanma Barranquero wrote:
> Packages are not arbitrary labels. "Packages" are. I said that
> packages are "packages" (i.e., labels) plus other things.
> 
> What I was saying, in other words, is that you can assign a bug to
> any arbitrary label, but that the label is not a real package unless
> the package exists.

You don't assign a bug to an arbitrary label. There's no such thing as
a label in debbugs. You assign a bug to a package.

The package(s) a bug is assigned is orthogonal to whether that package
actually exists and the other goodies that come with packages that
have been properly defined in debbugs.

I'd rather not waste time trying to work around problems and
misconceptions brought about by the use of invented non-standard
nomenclature to describe how debbugs operates. [In actual fact, I
probably won't spend the time dealing with such issues at all.]

> But still, you can search for the label "rmail" even if no package
> exists.

No. You can search for bugs which have package "rmail". There is no
such thing as a label in debbugs. The package a bug is assigned to has
nothing to do with whether the package is known to exist. Debbugs only
cares about what is in the package field of the bug when searching for
bugs.


Don Armstrong

-- 
A Democracy lead by politicians and political parties, fails.

http://www.donarmstrong.com              http://rzlab.ucr.edu

Re: Don't report new bugs for Rmail??

[Juanma Barranquero]

On Sun, Feb 1, 2009 at 05:02, Don Armstrong <don@donarmstrong.com> wrote:

> I'd rather not waste time trying to work around problems and
> misconceptions brought about by the use of invented non-standard
> nomenclature to describe how debbugs operates. [In actual fact, I
> probably won't spend the time dealing with such issues at all.]

It is clear that what I'm describing as a "label" is just what you
call a "package that has not been created". The fact they are
initially arbitrary labels is highlighted by the fact that

  reassign N this-nonexistent-package

works (with a warning), and then you can search for
this-nonexistent-package. If packages were so first-order objects, it
would be impossible to assign one to a nonexistent one.

> No. You can search for bugs which have package "rmail".

Yes. That's exactly what I was saying. But, until you create the
package rmail, "rmail" is only a label (even if you and debbugs don't
call it so).

> There is no
> such thing as a label in debbugs. The package a bug is assigned to has
> nothing to do with whether the package is known to exist. Debbugs only
> cares about what is in the package field of the bug when searching for
> bugs.

Which agrees 100% with what I said, except by nomenclature.

    Juanma

Re: Don't report new bugs for Rmail??

[Juanma Barranquero]

On Sun, Feb 1, 2009 at 12:15, Juanma Barranquero <lekktu@gmail.com> wrote:

> would be impossible to assign one to a nonexistent one.

s/one to/a bug to/

    Juanma

Re: Don't report new bugs for Rmail??

[Don Armstrong]

On Sun, 01 Feb 2009, Richard M Stallman wrote:
> There's something I don't understand:
> in what way does an Emacs bug report specify a "package"?

In the Package: pseudoheader; if there isn't a package specified in
the pseudoheader, the instance of debbugs on
emacsbugs.donarmstrong.com uses emacs as the default package.

Don Armstrong

-- 
The attackers hadn't simply robbed the bank. They had carried off
everything portable, including the security cameras, the carpets, the
chairs, and the light and plumbing fixtures. The conspirators had
deliberately punished the bank, for reasons best known to themselves,
or to their unknown controllers. They had superglued doors and
shattered windows, severed power and communications cables, poured
stinking toxins into the wallspaces, and concreted all of the sinks
and drains. In eight minutes, sixty people had ruined the building so
thoroughly that it had to be condemned and later demolished.
 -- Bruce Sterling, _Distraction_ p4

http://www.donarmstrong.com              http://rzlab.ucr.edu

Re: Don't report new bugs for Rmail??

[Richard M Stallman]

    > There's something I don't understand:
    > in what way does an Emacs bug report specify a "package"?

    In the Package: pseudoheader; if there isn't a package specified in
    the pseudoheader, the instance of debbugs on
    emacsbugs.donarmstrong.com uses emacs as the default package.

I don't know what a pseudoheader is, so I can't understand that
answer.  However, someone else provided the missing information: users
never specify these packages.  They are purely internal to the
development team.

Re: Don't report new bugs for Rmail??

[Stefan Monnier]

> If you go to http://emacsbugs.donarmstrong.com/, make sure "package"
> is checked in the "Find bugs by:" selector, type "rmail" into the
> "What to search for:" field, and click the "Find" button, you will see
> a page with the following on its 2nd line:

>   There is no maintainer for rmail. Please do not report new bugs against this package.

> ??? Why does it ask not to report new bugs?

The message is misleading, indeed.  It would be the same if you tried
"firefox" instead of "rmail".
A better message would be "There is no such package named `rmail'" or
something like that.


        Stefan