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

Re: Review of proposals

Lucas Nussbaum writes ("Review of proposals"):
> In Proposal D:

Hi.  Thanks for the review.

> Concern about length of proposal D
> ==================================

I'm going to deal with this first.

> I am a bit concerned about the length of proposal D, and the fact that
> it is both very detailed and very vague.

It's detailed about process.  This is because my experiences over the
past few years have been that process has been a problem.

> For example, it raises a (probably valid) concern about
> "non-init-related [declarative] systemd facilities", but:
> 1/ it mixes it with an argument that declarative facilities are better.
> Well, maybe I can agree with that. I'm not sure it's something
> the project needs to issue a statement on through a GR.

I have heard more than one person say that they are unhappy that the
current situation has been blocking specifically this kind of

> 2/ It says:
> > If policy consensus cannot be reached on such a facility, the
> > Technical Committee should decide based on the project's wishes
> > as expressed in this GR.
> Well, given that some of the criterias are not super-clear (see above),
> it's likely that policy consensus will be hard to reach. Then the TC is
> left with deciding based on the project's wishes as expressed in this
> GR. But assuming that proposal D wins, where is that project's wish on
> non-init-related declarative systemd facilities expressed?

That is referring to the list of criteria in my clause 9
(contextualised bn with the rest of of the whole proposal including
(for example) the principles in clauses 1 and 2).

What this is saying is that if we can't get policy consensus on
whether the clause 9 criteria are met, the TC should decide the same

I'm sorry that this isn't clear to you.  I'd be happy to improve that.

> I'm also not sure about the part about "being excellent to each other".
> How does this part of this GR interact with the CoC? Is it expected that
> the CoC includes special cases about init systems if this proposal wins?

I think your criticism here is addressed primarily at my clause 11.

Directly in answer to your question, no.  If I wanted to modify the
CoC I would say so explicitly.  The CoC is quite generic, whereas this
GR is quite specific.  I don't know if you have been hanging around on
-devel and reading the discussions there ?  Clause 11 is trying to
address some of the specific pathologies we see in those discussions.

Do you see the same pathologies ?  If so, what do you think should be
done to try to address them ?  The existing CoC etc. do not seem to be

> For those reasons, I am not sure if I will rank proposal D above FD. I
> would very much prefer if it were compressed to a proposal of about the
> same length as proposals B or C.

I am sorry it is so long, indeed.  It's just that I don't see what to
cut.  The init systems wars have been so wide-ranging and demonstrated
a large variety of different pathologies, and poor arguments, as well
as of course high-level differences of approach.

I'm open to further conversation on this.

> I don't understand this part:

> > 9. systemd provides a variety of facilities besides daemon startup.
> > For example, creating system users or temporary directories. Current
> > Debian approaches are often based on debhelper scripts.
> > 
> > In general more declarative approaches are better. Where
> > 
> > - systemd provides such facility
> > - a specification of the facility (or suitable subset) exists
> > - the facility is better than the other approaches available in Debian, for example by being more declarative
> > - it is reasonable to expect developers of non-systemd systems including non-Linux systems to implement it
> > - including consideration of the amount of work involved
> >
> > the facility should be documented in Debian Policy (by textual incorporation, not by reference to an external document).
> I don't understand the role of the last item ("including consideration
> of the amount of work involved") and it doesn't sound grammatically
> correct to me. The first four items are formulated as criterias, but not
> that one. Should it be rewritten as, for example: "the amount of work
> involved in implementing alternate implementations is reasonable"?

The grammar is a bit awkward because as you say the last entry on the
list is not a criterion.  I want the question of the amount of work to
"cut both ways" and affect the reading of the whole clause.  For
example, if the amount of work is very small, then a very nugatory
specification, or a very minor improvement, might be sufficient.

I don't want to pull that item out of the list; the list is a set of
things to consider.  But maybe it could be reworded.


Ian Jackson <ijackson@chiark.greenend.org.uk>   These opinions are my own.

If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.

Reply to: