Re: Developer Workload and Sustainability

[Ansgar]

Hi Simon,

On Thu, 2023-06-29 at 00:32 +0900, Simon Richter wrote:
> According to Policy as currently published, systemd units are 
> encouraged, and init scripts are mandatory.

Please stop lying:

+---
| Packages that include system services should include systemd service
| units to start or stop those services. [...]
|
| If the package does not include a service unit (if, for example, no
| one has yet written one), including an init script, as described
| below, to start the service is encouraged.
|
| Packages including a service unit may optionally include an init
| script to support other init systems.
+---[ Policy 9.3.1 ]

And no, this isn't exactly new. Except apparently for you.

The real exhausting thing is people lying, FUD, spreading conspiracy
theories (ominous commercial sponsors (Rothschilds? Soros?)), endless
revisiting of decisions (should we prepare to revert usrmerge because
the attention of ominous commercial sponsors might shift elsewhere?),
claiming systemd is rot/cancer/an infection/the Windows registry and so
on.

I agree that this isn't a technical issue though.


> That's the thing: few people want init scripts. I don't want init
> scripts either.

This very thread only exists because people want init scripts.  I agree
that this is very few people though.

> What I want is an init system that can be maintained inside a community 
> project like Debian without burning out people and endangering the long 
> term viability of the project.

And claiming this isn't possible with systemd is one thing: FUD.

> That is where systemd fails us as a community project, because the 
> environment in which most of development is happening is not
> hospitable to community building efforts, and the complexity of the
> project constitutes a high barrier to entry, which acts as a further
> selection filter for contributors.

More FUD.

> I don't yet see a clear path out of this. The only thing that is
> obvious to me is that this is not a technical problem[3].

I think we should consider removing sysvinit and init scripts from
Debian.  The non-technical cost of having them is too high.

Ansgar

Re: Developer Workload and Sustainability

[Russ Allbery]

Simon Richter  writes:

> According to Policy as currently published, systemd units are
> encouraged, and init scripts are mandatory.

I thought I found and fixed all of the places init scripts were said to be
mandatory, but apparently I missed one.  Could you file a bug and point it
out?  That obviously need to be fixed.

Policy certainy needs work.  For example, it still makes no mention of
triggers whatsoever, which is a huge gap in its documentation of
maintainer scripts.  Unfortunately, while there is some energy to document
new and changing things, there isn't as much energy for going back and
properly documenting things we're already doing.  If we somehow miss the
window while the change or new thing is being worked on, it can slip
through the cracks.  I used to have a lot more time to work on language
for that than I do now.

-- 
Russ Allbery (r...@debian.org)  

Re: Developer Workload and Sustainability

[Simon Richter]

Hi Russ,

On 6/29/23 01:58, Russ Allbery wrote:


According to Policy as currently published, systemd units are
encouraged, and init scripts are mandatory.



I thought I found and fixed all of the places init scripts were said to be
mandatory, but apparently I missed one.  Could you file a bug and point it
out?  That obviously need to be fixed.


It appears I have misread that paragraph. Neither are mandatory, just 
encouraged, which makes sense to me.



Policy certainy needs work.


What is definitely missing is the feature set systemd units can expect.

Since we're effectively delegating part of Policy to systemd upstream, 
this reference should be made explicit, and versioned. We certainly do 
not want to maintain our own copy of that documentation.


What is unclear to me is whether we want to include smaller policy items 
imported from systemd verbatim or by reference (i.e. is usrmerge a 
Debian policy, or a systemd policy?).


   Simon

Re: Developer Workload and Sustainability

[Luca Boccassi]

On Wed, 28 Jun 2023 at 17:20, Ansgar  wrote:
>
> Hi Simon,
>
> On Thu, 2023-06-29 at 00:32 +0900, Simon Richter wrote:
> > According to Policy as currently published, systemd units are
> > encouraged, and init scripts are mandatory.
>
> Please stop lying:
>
> +---
> | Packages that include system services should include systemd service
> | units to start or stop those services. [...]
> |
> | If the package does not include a service unit (if, for example, no
> | one has yet written one), including an init script, as described
> | below, to start the service is encouraged.
> |
> | Packages including a service unit may optionally include an init
> | script to support other init systems.
> +---[ Policy 9.3.1 ]
>
> And no, this isn't exactly new. Except apparently for you.
>
> The real exhausting thing is people lying, FUD, spreading conspiracy
> theories (ominous commercial sponsors (Rothschilds? Soros?)), endless
> revisiting of decisions (should we prepare to revert usrmerge because
> the attention of ominous commercial sponsors might shift elsewhere?),
> claiming systemd is rot/cancer/an infection/the Windows registry and so
> on.
>
> I agree that this isn't a technical issue though.
>
>
> > That's the thing: few people want init scripts. I don't want init
> > scripts either.
>
> This very thread only exists because people want init scripts.  I agree
> that this is very few people though.
>
> > What I want is an init system that can be maintained inside a community
> > project like Debian without burning out people and endangering the long
> > term viability of the project.
>
> And claiming this isn't possible with systemd is one thing: FUD.
>
> > That is where systemd fails us as a community project, because the
> > environment in which most of development is happening is not
> > hospitable to community building efforts, and the complexity of the
> > project constitutes a high barrier to entry, which acts as a further
> > selection filter for contributors.
>
> More FUD.

Let's look at some community numbers for two init system projects in
question, we like to look at hard data after all, don't we. To spicy
it up, without mentioning which is which:

$ git shortlog -s | cut -c8- | wc -l
2318

$ git shortlog -s | cut -c8- | wc -l
7

Just from this alone, it certainly seems true that one of those is a
project that "fails as a community project" and that is "not
hospitable to community building efforts". Anyone want to guess which
is which?

Kind regards,
Luca Boccassi

Re: Developer Workload and Sustainability

[Sean Whitton]

Hello,

On Wed 28 Jun 2023 at 06:20PM +02, Ansgar wrote:

> On Thu, 2023-06-29 at 00:32 +0900, Simon Richter wrote:
>> According to Policy as currently published, systemd units are
>> encouraged, and init scripts are mandatory.
>
> Please stop lying:

Please assume good faith.  There is no reason to think Simon would be
lying about what he thinks is in Policy.  It's a mistake, not a lie.

> The real exhausting thing is people lying, FUD, spreading conspiracy
> theories (ominous commercial sponsors (Rothschilds? Soros?)), endless
> revisiting of decisions (should we prepare to revert usrmerge because
> the attention of ominous commercial sponsors might shift elsewhere?),
> claiming systemd is rot/cancer/an infection/the Windows registry and so
> on.

Perhaps people are spreading conspiracy theories like this about systemd
outside of Debian.  But people are not doing that on Debian lists, so
the quoted text from your message seems entirely off-topic.

-- 
Sean Whitton


signature.asc
Description: PGP signature

Re: Developer Workload and Sustainability

[Sean Whitton]

Hello,

On Wed 28 Jun 2023 at 06:17PM +01, Luca Boccassi wrote:

> Let's look at some community numbers for two init system projects in
> question, we like to look at hard data after all, don't we. To spicy
> it up, without mentioning which is which:
>
> $ git shortlog -s | cut -c8- | wc -l
> 2318
>
> $ git shortlog -s | cut -c8- | wc -l
> 7
>
> Just from this alone, it certainly seems true that one of those is a
> project that "fails as a community project" and that is "not
> hospitable to community building efforts". Anyone want to guess which
> is which?

It's understandable that you'd feel frustrated by what seems like a
misrepresentation of your project's organisation and ethos, but please
try to avoid this sort of rhetoric.

You can just as well present the git statistics without hiding the
author's names for rhetorical effect, or asking rhetorical questions,
and it would keep the temperature of the discussion lower.

-- 
Sean Whitton


signature.asc
Description: PGP signature

Re: Developer Workload and Sustainability

[Ansgar]

On Fri, 2023-06-30 at 11:10 +0100, Sean Whitton wrote:
> On Wed 28 Jun 2023 at 06:20PM +02, Ansgar wrote:
> 
> > On Thu, 2023-06-29 at 00:32 +0900, Simon Richter wrote:
> > > According to Policy as currently published, systemd units are
> > > encouraged, and init scripts are mandatory.
> > 
> > Please stop lying:
> 
> Please assume good faith.  There is no reason to think Simon would be
> lying about what he thinks is in Policy.  It's a mistake, not a lie.

It's not only about this sentence, but the entire mail.

And it's also not the first instance.

> > The real exhausting thing is people lying, FUD, spreading conspiracy
> > theories (ominous commercial sponsors (Rothschilds? Soros?)), endless
> > revisiting of decisions (should we prepare to revert usrmerge because
> > the attention of ominous commercial sponsors might shift elsewhere?),
> > claiming systemd is rot/cancer/an infection/the Windows registry and so
> > on.
> 
> Perhaps people are spreading conspiracy theories like this about systemd
> outside of Debian.  But people are not doing that on Debian lists, so
> the quoted text from your message seems entirely off-topic.

People, including DDs, do several of the things I listed, including in
this thread.  So it seems entirely on-topic.

Ansgar

Re: Developer Workload and Sustainability

[Ansgar]

Hi Sean,

On Fri, 2023-06-30 at 11:14 +0100, Sean Whitton wrote:
> It's understandable that you'd feel frustrated by what seems like a
> misrepresentation of your project's organisation and ethos, but
> please try to avoid this sort of rhetoric.

Fancy idea: how about we ask people to *stop* grossly misrepresenting
other projects instead?

We've had a decade of that about systemd, probably more if one looks at
Pulseaudio, GNOME and other things. Eventually we might reach a point
where we might want to stop that. Sadly I don't see you asking for that
to happen, rather the opposite.

> You can just as well present the git statistics without hiding the
> author's names for rhetorical effect, or asking rhetorical questions,
> and it would keep the temperature of the discussion lower.

Okay, then let's just note that sysvinit has an extreme barrier of
entry, driving most contributors away.  A property it seems to share
with dpkg.

Thus both aren't a sustainable base to build a community distribution
on and we should look at solving that problem, possibly by using
community-friendly alternatives instead.

Does that help? I tried to write in the helpful style the mail I
replied to uses. I skipped stating {sysvinit,dpkg} proponents haven't
done their homework, using {sysvinit,dpkg} incurs technical debt, they
failed us as community projects, it's impossible to onboard people to
them[1], and possibly some other minor points.

Ansgar

  [1]: Of course stating this is fine even when I'm not involved with
   dpkg or sysvinit, as per the mail I replied to.

Re: Developer Workload and Sustainability

[Sean Whitton]

Hello,

On Fri 30 Jun 2023 at 12:31PM +02, Ansgar wrote:

> We've had a decade of that about systemd, probably more if one looks at
> Pulseaudio, GNOME and other things. Eventually we might reach a point
> where we might want to stop that. Sadly I don't see you asking for that
> to happen, rather the opposite.

Let me just note that you're lumping together a number of distinct
projects here, which is not fair to the sysvinit people in this thread.

> Does that help? I tried to write in the helpful style the mail I
> replied to uses.

I don't think sarcasm is constructive in this sort of contentious
discussion.

-- 
Sean Whitton


signature.asc
Description: PGP signature