I call out one area where I think additional discussion is valuable; see
after the underscores.


I'd like to thank Andrew for posting his summary of the discussion.
As I have said over the years, I think such summaries are a critical
part of driving discussions forward.
I didn't ever get a chance to do this as DPL, but one of the great
things about summaries is that anyone in the community can post an
alternate summary if they think the summary misses the mark.

So here's my summary.
I want to stress that this is a summary here. I haven't gone around
digging through vcs history or even looking at things like whether
someone has tried to revert the NMU.

We had a couple of things going on here.

1) Andrew asked the community whether we needed this package.

2) Jonathan Dowland made an NMU of fortune-mod removing
fortunes-off.

As I said I'm only focusing on the discussion. No one in the discussion
made a claim about ordering or about whether these two events were
related.


Let's take the NMU first.

In [1] Jonathan indicated that his action was reversible.

Anyone who wants to put their name to explicitly racist, sexist and
pro-nazi material in Debian is free to re-upload it.

[1]: https://lists.debian.org/[email protected] No one has disputed the assertion that people are free to revert the NMU
if they are willing to sign up to maintaining fortunes-off.

Even Steve Langasek
said that if someone wanted to do the work, they could [2]:

If someone wants to sift through the contents of fortunes-off to
separate
the wheat from the chaff, fine, let them do it. But the presence of
some
good fortunes in the package doesn't compel anyone to keep it, nor does
rightly pointing out the garbage that's in it incur an obligation to do
the
work to filter out only the stuff that conflicts with the project's
Diversity Statement.

[2]: https://lists.debian.org/[email protected]

(I have contested Steve's claim that the diversity statement is relevant
to package content.)

There were other messages that cast the idea of doing such an NMU in
more positive light.
My point is that there is a strong consensus that if someone wants to do
the work of maintaining the package, they can do so.


*However* There is a lack of consensus on what work a maintainer would
need to do in order to maintain the package.
The messages quoted above are typical of those claiming that significant
work would need to be done.
Also see messages like [3]

[3]: https://lists.debian.org/[email protected]

Other messages supported the claim that things are fine as they are.
I see no consensus here.

I claimed [4] that what we do when there's not a consensus about such
things is trust our maintainers and defer to their judgment. [4]:https://lists.debian.org/[email protected]

________________________________________

And now we come to Andrew's question.

According to Andrew, he was asking a question, not advocating for a
specific decision [5]:

Would it be a good idea to at least remove fortunes-off and the corresponding >data file for Bookworm at this point and going forwards?
It saves a tiny amount of space and a few translations and represents a small >simplification of Debian at minimal cost.

[5]: https://lists.debian.org/[email protected]

Other messages from Andrew emphasize that this is a question, not even
an advocacy for this change. I'll note that some of Andrew's questions
are fairly leading, and many people including myself took Andrew's
message as if he was advocating for this change.
But even later [6], he was stating possibilities, not claiming them as
his opinion:

It may be a fact that we don't promote Nazism but the presence of Nazi
texts
may itself be problematic to some people. There are other elements in
fortunes-off which are perhaps inappropriate within Debian.

It may be that fortunes-off doesn't constitute a corpus of (optional)
text that is worth retention? The fashion for fortunes has rather
disappeared in the last 25 years, after all.

[6]: https://lists.debian.org/[email protected]

(Which is to say were I judging consensus, I would not read Andrew's
messages as in favor of removing fortunes-off. Be careful how you
speak:-)

However, several members of our community were more explicit in their
desire to see fortunes-off go.

In his summary [7] Andrew claims that this is *hard*:

There's been a grudging consensus that this is *hard*. It's very hard indeed >to draw good conclusions as to what to do when everyone agrees that something >could be done and disagrees with what that should be.

[7]: https://lists.debian.org/Y35hKdU/7/[email protected]

I agree with Andrew's summary: there is no consensus on what to do.
Above I've posited that given a lack of consensus, we don't have a
decision ath this level, and absent something like a GR, package-level
issues fall to individual maintainers.
But, I think there's a certain area where we're talking past each other.
And we might have more of a consensus if we explore that.

It appears that people who are advocating against fortunes-off have one
idea of how that database came to be [6]:

The fashion for fortunes has rather
disappeared in the last 25 years, after all.

Or from [2]:

This package is a fossilized
collection of fortunes that some random people on Usenet found funny or
otherwise worthy of inclusion over 25 years ago. There are
subcollections
of fortunes in this package that are explicitly *categorized* as
racist,
homophobic, and misogynistic.

In general, the assumption appears to be that these were collected
because people found them funny, and that little more thought was put
into the compilation than that.

In contrast, Branden Robinson [8] appears to be arguing that the
compilation criteria was more nuanced.
[8]: https://lists.debian.org/20221120182859.k7ehjxuzt4zxpdal@illithid

My take is that Branden believes that the compilers/editors of the
collection had motivations related to historical irony, pointing out the evolution of ideas, and generally put a lot more thought into things
than "these were funny."
Certainly as a reader/user of fortunes-off Branden appears to be much
more engaged than "hey that's just a thing on my screen."
In particular, he has inferred properties of the collection as a whole, reminisced on how he wished the collection were different, etc.

In many cases, texts that provoke that level of engagement and thought
are given a higher level of protection than simple lists.

In particular, in Branden's world, it's not at all clear that the
editors agreed with the quotes or promoted their philosophy at the time
they were included.

Similarly, The Wanderer [9] explained what fortunes-off meant to them:

[9]: https://lists.debian.org/[email protected]

I have never once interpreted them as promoting Nazi ideology. Rather, I
have always interpreted them as being a useful reminder of the sorts of
things that Nazis said and stood for - and it's important to remember
those things, both in terms of remembering the things that Nazis *did*,
and so that we have the additional warning if we ever find ourselves at
risk of stepping towards the same paths.

In conclusion, it appears to me that people on different sides of this
debate are viewing the literary context of fortunes-off differently.
And that's a very complicated issue, because the reader and the context
they bring is a very important part of the experience.
Even that is showing some of my bias.
I have tried to be relatively neutral here, but it's hard.


I'd suggest several questions to explore if we want to try and build
more of a consensus.

Of anyone expressing an opinion:

* What do you think fortunes-off is?

* Why do you think the fortunes were added to that package?

* What do you think users of fortune and fortunes-off are hoping for
when they interact with the corpus?

And questions generally to us as a whole:

* How should we approach creative works when the context is ambiguous?

* Where do we draw the boundary between encouraging thought
about/exploration of uncomfortable ideas and promoting those ideas?

* How do we want to approach things where a reader/user has to infer a
lot of context and some approaches they might take are negative?

Thanks for your consideration,

--Sam

--=-=-Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iHUEARYIAB0WIQSj2jRwbAdKzGY/4uAsbEw8qDeGdAUCY36N4wAKCRAsbEw8qDeG dAirAQDKYH3boEokXm9BqTJ1Wx9syCZ+6QeamsvRZ8mIxixPuQD9HHXGkgWQxTGs BSzuw5d3+Rlg6yM+jVhUTpufP/GhxQA=5/ci
-----END PGP SIGNATURE-----

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)