What should the FSFE provide

Bernhard E. Reiter bernhard at fsfe.org
Fri Aug 31 11:16:57 UTC 2018


Hi Paul,

Am Donnerstag 30 August 2018 21:56:58 schrieb Paul Boddie:
> Thank you for indulging me in this discussion which has probably covered
> more ground already than many previous discussions on some of these topics.
> My apologies if this is a long message.

no problem, you are welcome! Thanks for caring for Free Software.
Given the heated tangled discussion on this list, please take my apologies for
a brief answer. There is probably more time for longer exchanges at a later 
point of time.


== FSFE offerings of personal blogs

> Still, one might hope that the blogs act as that "conversation starter",
> with discussion possibly moving to places like this list, but we don't
> really see much of that, either. So it seems that regardless of the
> technology employed, the FSFE does not necessarily provide a particularly
> effective forum for driving Free Software discussions, development and
> strategy. At least from the perspective of a distant collaborator or
> outsider.

What worked best in many years were to bring people together at event, 
supporter meeting, local group, conferences and so on. People that already 
had an connection to each other via FSFE. If you want to become more involed 
in this, try to go to one or organise one yourself.

It is true that we haven't found a cool response to the challenges that new 
electronic media has brought to high quality exchanges. Should be wo more 
micro-blogs, social medias, chats or what else? What keeps the coherency.
We are not the only group that has this problems. Combined with more topics in 
society generally, less political actions by young people and growing success 
of Free Software worldwide leading to growth problems we do have a 
combination of challenges. The best inspritation I've found is the book of 
Tina Rosenberg, called "Join the club".


== Preserving knowledge

> I don't think that the scope of what the FSFE would need is anywhere near
> to what archive.org does, either in their Web archiving role or just for
> cultural artefacts. It just needs to keep its published content in place,
> which is admittedly something that many organisations fail to do as they
> "refresh" their Web sites. (Even supposed custodians of knowledge such as
> universities are surprisingly appalling at this, I find.)

Keeping our contents and history in place is something that we do (as far as I 
know). Not every revision of course, but everything significant. Do you have 
examples of what contents you are missing.

> the organisation does need to avoid
> the trap of outsourcing everything because people do then rightly wonder
> about the value provided by the organisation.

And we don't however, FSFE's aim is not to directly produce software, we want 
others to be able to do it in several aspects. For many is much better if 
there are several choises of service providers that are not FSFE. It helps us 
to be in a position to evalutate those providers.

> I could make observations about other organisations in this domain. For
> instance, the Python Software Foundation chose SourceForge for repository
> hosting back when people were still using CVS/Subversion. It then switched
> to self-hosting using Mercurial, but people started to clamour for Git
> usage and then GitHub usage with all the usual dubious assertions that this
> entails.

This always is a hard question for an organisation: What does this contribute 
to our core mission? A question like this has to be weighted for each case. 
If the core missiong of the PSA is to further Python, they may profit from 
the lower entry barrier of github and live with the drawbacks. Just like the 
FSF took advantages at some time of computers with proprietary compilers, 
proprietary operating system, proprietary microcode on chips and other stuff 
in order to raise the chances to have more Free Software in the future.

Personally I had hoped that the PSA would have decided differently. (I am one 
of the co-maintainers of an old issue tracker technology they are still 
using, called roundup. We have a hard time competing with a development based 
on proprietary software money like github, bitbucket or gitlab.) But if 
Python lost important young people because of the higher barrier, the github 
move may contribute for them being there and relevant in 2 years and not 
leaving the field to languares more geared towards the interest of large 
corporations like Swift or Go (meanwhile both Free Software languages and 
technically good, the influence of lage computing companies and their needs 
is still very strong.)

> I understand that people don't want to do their own hosting, but at some
> point someone has to do the hosting, and then one needs to consider the
> burden on individuals. 

In my eyes a long term solution needs professionals running them, otherwise
there is a too great risk of not having the necessary stability (for support, 
security, updates, maintenance). FSFE could run this, like any organisation 
could, but why should it. What is bad about buying bread from a professional 
bakery, produce from a farmer, news from a journalist and get thaught by a 
teacher? I could probably do each one of this myself, but not everything 
together.

> It is a bit like discussions about "social media" 
> platforms, but I doubt that anyone participating in this list would
> advocate moving everything to Facebook and have fsfe.org redirect to some
> Facebook URL.

Facebook's service is far away from FSFE to be considered a vendor.
Still we may need to go there to talk to people who we want to get educated 
about Free Software and we could only reach there. (If only for telling them 
why they should not be on Facebook.)

> Meanwhile, there are arguably dormant organisations like Qi Hardware which
> still manages to maintain all its Web resources, and it may even be the
> case that most of those resources are still fully usable and not read-only
> archives. Although I acknowledge the challenges of hosting services,
> sometimes I think that people make arguments based on those challenges to
> pursue another agenda.

There are many very interesting approaches not just Qi Hardware. I like them.
We need more of them. And they need more buyers.

FSFE is about making it possible they can get more fraction and stay allowed. 
Think about the radio frequency initiative, that would just outrule to use 
some devices, no matter how clever they be constructed. Think about how we 
try public and private procurement to make it possible to prefer Free 
Software (or other open resources) or at least buy it at all.

> This extends to arguments about using more "agile" tools for communication.
> People seem to think that e-mail is too much effort and that forums would
> be better, and so on. But this is often motivated by people wanting to
> spend as little effort as possible on their interactions (hence
> observations about messages sent from people's phones). The result is often
> low-quality "tweeting" of opinions and unsustainable collaboration.
>
> (Given your own considerable experience, I imagine that you have also seen
> situations where advocacy for "low-threshold editing" on things like wiki
> services, to pander to people who want to do "quick edits", ends up with
> incoherent content and spam that is someone else's job to tidy up.)

A low barrier does not guarentee good contents, however a high barrier also 
does not. A low barrier to me means that I could contribute more in many 
situations, so I do prefer a low barrier to contributors and well supported 
tools. Github has developed some advances in this field, just as Sourceforge 
did. And of course with many people supporting them with their attention and 
feedback. What can we do to keep Free Software tools to keep up with those 
improvements, how can we make other advantages of Free Software more popular 
(think decentralised communication where XMPP with OpenPGP was way ahead of 
many messengers, but still failed to get the large traktion)?

Professionally I stay away from Github as much as I can, but I cannot fully, 
because otherwise I cause trouble for people I am developing Free Software 
for, which would mean that I couldn't to it up to the extend that it wouldn't 
be Free Software in the end.

== Places to discuss

> Indeed. And we lack mechanisms to take such discussions further. Here, the
> Python community provides another lesson. Up to about ten years ago or so,
> the concept of "special interest groups" was actively cultivated. 

I'm a big fan of Python and its decision structure. Still they are also having 
problems to maintain the consistency and the rate of innovation. And were 
using a single person as technical lead (Guido van Rossum).
The main point is: Their mission is quite different from FSFE's.

> However, another phenomenon emerged which involves people promoting
> specific solutions and denigrating others. Instead of reaching a consensus,
> the goal of some people was to try and dominate others and persuade the
> audience to pick their solution as the winner. 

I believe that some competition is good. Having several solutions to chose 
from is good. Hopefully people will select on merits and not just style of 
presentation of course.

== Groupware 
> But it is interesting to see how some of the software somehow lost momentum
> and relevance, meaning that people were no longer considering Kolab (version
> 2 and earlier) as a competitive solution.

For historic reasons I am still using it each day and it is old, but quite 
stable. Missing a number of features and so on. I still have customers who 
use it with surprising stability. :) But its old and not competitive anymore 
for a new installation. My story about this is, that together with others I 
had to hand over the torch to somebody else. It did not turn out how I had 
hoped it would, but as the former coordinator, I am not in a position to 
criticise my followers in detail, they did the best they could and followed 
what they perceived was the right way.

[..] 
> it became apparent that there was a gulf between what 
> we wanted to suggest to potential adopters and what was available and 
> acceptable to them.

> And again, this is where organisations like the FSFE have to be able to
> substantiate and support their advocacy with activities that make their
> claims a reality. 

I've learned many lessons about this the hard way. I don't think this would 
work. If we ever meet, we should have a beer over this conversation. 

== Funding Free Software

> A huge structural problem that has increasingly been noticed is that no-one
> wants to fund Free Software in general. For the most part, companies will
> fund things to get their products out the door, Free Software or otherwise,
> and if they can avoid paying for "infrastructure components" then they
> will. Here's a recent apt commentary with some strong language about this
> phenomenon:
>
> https://carlchenet.com/foss-passive-consumerism-kills-our-community/

(Sorry, stopped reading the link very soon, as it does not promise a thought 
out contribution to the debate.)

My assessment of the situation is different, I believe more and more companies 
are funding Free Software development, in some areas it is the only way now 
to stay competitive. Examples Google with Android, Chromuim, Angular, Go. 
Facebook with React, Twitter with Bootstrap, Microsoft with Visual Code,
and many many others like IBM, Redhat with Linux.
 

> Now, I don't necessarily expect the FSFE to directly fund software
> development. Meanwhile, since the FSF wishes to see software development
> done, I would expect it to pay people a decent wage for the work it wishes
> to see done. 

The funds of FSFE and FSF are soooo little compare to what it take to do 
anything of medium significance in software engineering. We may be able to do 
a few very important missing links, small stuff at the maximum.

> Unfortunately, there is a culture of people doing work "for 
> exposure" which means that people are tempted into doing things for free to
> get their name known.

It is one of the motivations for people to work on Free Software, but we need 
more professionals and thus we need more money for Free Software which means 
more companies and customers of companies that want this.

> The consequences are often (1) nothing actually getting done, (2) people
> burning out as they try and fit their volunteer work in around everything
> else in their lives, 

To any volunteer, my advise as a volunterr is:
Always prepare for the long run. Don't overstretch.

> (3) the perception that Free Software is a product of 
> hobbyists who can be paid with pennies. 

This perception is fuelled by people who do not want to see the number of 
professional offerings or who do not want to pay for software or service at 
all.

> So while I accept that policy activities are vital to allow Free Software to
> be deployed, it also must be  recognised that without a sustainable
> development culture there will be no Free Software to deploy.

FSFE does foster development culture and we may even do more. Still it is more 
important that if a person starts a Free Software business he or she is 
allowed to do so and get a fair chance to win over business.


> > == license compliance

> > For some of the activities see:
> >   https://fsfe.org/activities/ftf/legal-conference.en.html
>
> My problem with this is that it doesn't seem very transparent. And again,
> there are individuals who have sought to mischaracterise other
> organisations who have tried to discuss general matters of compliance
> openly. Meanwhile, a vacuum has formed around licence compliance that has
> been exploited by various opportunists.

This maybe a area we need to get into the details. Within FSFE we had a number 
of discussions about this lately and they have also been tied with specific 
persons more often than I would have liked. Being effective on such a topic 
means that we find, hold and motivate people and sometimes for personal or 
other reasons there are gaps. (I consider this a daily operational question 
where a public debate about it would just be unhelpful.)

> What this communicates to outsiders like myself is that other organisations
> are (1) more likely to offer practical support and advice below the level
> of formal legal consultation (which is, to be honest, where most developers
> want to be operating), (2) more open about the legal and regulatory
> challenges facing Free Software, and (3) more active with regard to
> upholding Free Software licences and defending well-intentioned Free
> Software usage in Europe.
>
> Again, this may be an unfair impression, but I can only work with the
> visible facts I have to hand. And when an organisation solicits support, it
> surely falls on that organisation to make a clear and convincing case for
> such support. On this particular matter, and in contrast to the outreach
> done by other organisations, I have difficulties seeing such a case.

You maybe right in that FSFE has been less public about what it does in recent 
years. We believe that we need to help that more people understand, can 
follow and come to their own conclusions about licensing "enforcement" and 
how the different players interact. Otherwise FSFE too much becomes the 
organisation who forces decisions, which has long term disadvantages. If 
people like us to put in more resources on this topic, it is good to get this 
as feedback. 

> Once again, my apologies for a long message which still didn't discuss
> everything it needed to. However, I hope it provides another perspective
> that, to some extent, maybe some other people also share.

Thanks for sharing your perspective and giving me a chance to respond.

Best Regards,
Bernhard

-- 
FSFE -- Founding Member     Support our work for Free Software: 
blogs.fsfe.org/bernhard     https://fsfe.org/donate | contribute
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.fsfe.org/pipermail/discussion/attachments/20180831/680a7e90/attachment.sig>


More information about the Discussion mailing list