[License-discuss] discussion of L-R process [was Re: [License-review] Approval: Server Side Public License, Version 2 (SSPL v2)]

Rick Moen rick at linuxmafia.com
Fri Mar 15 21:17:39 UTC 2019


Quoting Ben Hilburn (bhilburn at gmail.com):

> Luis offered some great suggestions for possible improvements. I've
> never used Discourse, so can't comment on it directly, but the general
> goals of improving transparency, clarity, and inclusion are acutely
> important and exactly the sort of things we should be trying to
> improve moving forward, in my opinion.

I can comment briefly on Discourse.

It's a heavily AJAX-based discussion medium with continuous scrolling,
which makes it nearly impossible to (locally, independently) save
archived copies of anything.  Basically, all of the disadvantages of Web
forums are present (because it's an example of that), somewhat worsened
by the continuous-scrolled presentation.  It also has a non-threaded
'topics' model (lack of threading being another egregious fault IMO)>
It does realtime updates (notifications, etc.) and is mobile-friendly,
with smartphone apps for Android and iOS based on Facebook's React.js
Javascript toolkit.

In fact, one might say that it reflects a mobile-centric mentality in
the sense that you are in a constant scrolling experience, e.g. must
read all the way to the bottom of a topic to get it marked read, then go
on to the next topic via either the links at the bottom or top, or back
up to category view, just as with a smartphone or tablet where you are
forced into a serialised UI rather than being able to see the structure
of all discussions and navigate freely (as with a good mail or netnews
reader).

Organisation administrators tend to smile on Discourse because of its
support for crowd-sourced moderation, in its case 'flagging' of posts 
anyone deems inapproriate, that are then hidden pending staff review.
(I have nothing against crowdsourced moderation, except that for the
most part I'd greatly prefer the decision about what to killfile be
delegated to each subscriber.)

Being written in Ember.js and Ruby on Rails, Discourse is slow compared
_even_ to other Web-forum software, heavily loads its servers, and doesn't
scale well (what one might called the GitLab problem).  


I avoid using Discourse, to the extent possible.  In my experience,
technical computer users tend to feel likewise, and a typical discussion
would go:  Them: 'What's Discourse like?'  Me: 'It's a non-threaded Web
forum with continuous scrolling, on Ruby on Rails with gobs of
Javascript, that's archiving-resistant.'  Them: 'Oh, never mind.'

Another reason for technical-users' leeriness of Web forums generally is
their established history:  Their lifecycle tends to involve several years of
operation followed by either shutdown or a webmaster deciding to replace
it with something newer and shinier, in the latter case discarding all
existing forum contents.  Web forums are as close to migration-proof as
makes no difference, whereas mailing lists with 20-year continuous
history are routine, partly because of use of a standard data store,
usually Unix mbox format.  For example, Silicon Valley Linux User
Group's mailing list has traffic back to 1997 because it was a trivial
task to incorporate the initial Majordomo mailing list archive into that
of its replacement, GNU Mailman.

So, a technically knowledgeable contributor will know that any
substantive contribution to a Web forum will probably evaporate in a
short number of years.  Add to that the fact that continuously scrolled
Web-forums have slim-to-no representation on Web search engines:  E.g.,
while important mailing lists from 2002 are still archived at
web.archive.org, no Discourse forums are, and probably never will be.


I cannot help wondering if Luis is seeking to solve the wrong problem.
(As co-author of an essay on seeking help on technical problems, 'How to
Ask Questions the Smart Way', I've seen a good bit of that.)  If the
main problem is 'Sometimes, net.random participants on OSI mailing lists
are misunderstood to be somehow reflecting the position of OSI', then 
perhaps the appropriate remedy is a stronger social convention for
official OSI representatives to always identify themselves as such and
state when they are speaking officially.

You could also alter the Mailman listinfo pages and new-subscriber text
for license-review and license-discuss to stress these being public
mailing lists open to any member of the public willing to abide by the
code of conduct, and should NOT be assumed to speak for OSI unless so
indicated.  However, to be rather blunt, I believe the recently
disgruntled commenters are entirely aware of that fact, and, if they are
not merely attempting passive-aggressive kickback against License
Committee decisions they didn't like, are not easily distinguished from
that interpretation.


(Oh:  Wiews expressed are mine and in no way speak for OSI, with whom I
have no institutional connection.)

-- 
Cheers,              "I am a member of a civilization (IAAMOAC).  Step back
Rick Moen            from anger.  Study how awful our ancestors had it, yet
rick at linuxmafia.com  they struggled to get you here.  Repay them by appreciating
McQ! (4x80)          the civilization you inherited."           -- David Brin



More information about the License-discuss mailing list