Mailinglist Archive: opensuse (783 mails)

< Previous Next >
Re: [opensuse] openSUSE merits a better home page
On 26 October 2017 at 07:56, Per Jessen <per@xxxxxxxxxxxx> wrote:
Roger Price wrote:

My impression is that you will need to present very strong arguments
to convince the board to change tack.

I think anyone who comes up with an improved home page will be
considered.

Indeed

But this is not the correct mailinglist to discuss this topic

This is opensuse@xxxxxxxxxxxx - openSUSE's support mailinglist. This
is not a support topic

Please relocate this discussion to opensuse-project@xxxxxxxxxxxx (if
you wish to discuss the Project's direction, Board's decisions and
such) or opensuse-web@xxxxxxxxxxxx (if you want to actually discuss
the openSUSE website)

As both are in question, I will however breach with my new protocol of
not providing insight into the workings of the project on this list to
provide a little more context as to how the current openSUSE website
and the Projects direction came to be.

Before I start, a general reminder - the Board is openSUSE's _elected_
leadership body, responsible for leading and steering the Project.
5 of the 6 seats are directly elected by the community.
Only a maximum of 2 of those seats can possibly be held by employees
of the same company (eg. SUSE).
The openSUSE Membership can force a re-election of the Board, if 20%
of the membership request it.
The 6th seat is appointed by SUSE as the Board's Chairman.
The current Board Chairman (myself) was a community elected Board
member before being appointed, and a user / community contributor for
8 years before being employed by SUSE 4 years ago.
The 5 elected Board members can, at any time, request that SUSE
replace their Chairman.
The Board (as the Project as a whole) is independent from SUSE and
only the Chairman has a responsibility to represent SUSE's needs on
the Board.
Therefore any suggestion that the Board makes any decision on behalf
of SUSE is misrepresenting the function of the Board, the facts that
the Board answers to the community, the Chairman answers to the Board
as much as SUSE, and the openSUSE Project is independent from SUSE.

Over the years we (the Project) have repeatedly looked at the
statistics of our userbase and release downloads to give us an
indication of the general health of the openSUSE Project.
The Board has used this statistics in part of their decision making
processes, but as should be obvious from the story below were not the
only factors in play.

By the end of November 2014 there were some clearly apparent problems
with the Project.

Since 2009

- our number of monthly downloads of our distributions had steadily
declined, suggesting a constant apathy in the adoption of the
Project's main output
- our number of downloads around each release had steadily declined,
suggesting that each release was failing more than the last at
gathering significant traction with the broad "Linux of everyone"
target audience the openSUSE Project had at the time
- our number of unique users of our official download repos had
steadily increased, suggesting however that we had a steady, loyal,
and growing userbase

When taken on balance this mean that openSUSE could be argued to be
_at best_ stagnant, or at worst in a shallow decline.

This led the Board to do significant analysis of the results, asking
lots of our users lots of questions. The conclusion of the Board at
around 2014, and since reinforced by subsequent Boards, was that the
Project was failing to connect with the Board "Linux for everyone"
target audience, as shown by the declining download numbers.

However, the strong and stable repo userbase numbers gave us some
hope. The Board's analysis of the users found a pretty clear and
obvious trend - one of our strongest usergroups that were consistently
adopting, keeping, and growing their openSUSE usage were developers,
power users, and sysadmins. And of course these userbases were also
highly represented in Tumbleweed/Factory, which was growing well at
the same time.

Put simply, this was the primary motivation for the Board pushing the
Project in the direction we have, focusing on these userbases.

Since making that decision, we have seen openSUSE's downloads and
userbase numbers for both Leap and Tumbleweed grow.
We have also seen the number of projects beyond the distributions
under the openSUSE umbrella grow.
I personally have no doubt we made a good decision based on the
information we had to hand, and I think we did a good job of
rationalising the facts and figures with the Board's personal and
intimate knowledge we had of the Project, as we are all part of it.

In parallel to the above analysis and the raw numbers, there were
other factors at play, especially in the openSUSE Leap part of the
story.

No one can deny that the 12.2, 12.3, 13.1, and 13.2 releases were very
problematic to develop. The delays and close calls speak for
themselves.

The simple fact was that the Project was struggling to find volunteers
to actually help produce the distribution. This often led to
significant delays.

By the release of 13.2, despite the fact we did manage to get the
release out of the door, there were so few contributors left working
on the release that we basically had no one left to do any future
releases.

In the end of 2014 and early 2015, as Chairman, I was faced with the
very real and depressing problem of having to find a way of informing
the community that there would be no more releases of the openSUSE
Distribution.
I kept this problem privately between me and the Board at the time, in
the hope there would be a solution to save me from having to do that.
I spent a lot of time trying to find new volunteers to step and drive
future releases, with no real success.

This was despite the fact that openSUSE Tumbleweed was going very
well, with a very strong number of contributors, merging with openSUSE
Factory, and growing it's userbase at an incredible rate.

Luckily, within SUSE there was their own efforts to find a way of
making the SLE sources more readily available and more useful to a
broader ecosystem than just the existing SUSE developers and partners.

Along with a number of SUSE employees who are also openSUSE
contributors (including within SUSE Management), we were able to steer
things within SUSE that the idea morphed into the direct provision of
the SLE sources into OBS, with the suggestion to the community that
the community base a stable distribution on those sources.

In a prototypical form I presented the idea at oSC 15, and from that
point on it was driven by the community. It has since evolved into
what everyone now knows as openSUSE Leap.
Instead of openSUSE no longer having a stable distribution due to lack
of contributor interest, Leap has found new contributors in addition
to the ones we effectively 'stole' by milking SUSE's SLE efforts for
everything they were worth.
It is also a development model which is easier for our contributors
more interested in Tumbleweed to still work with and help shape and
maintain Leap in a sustainable way.
And as a result Leap has been a bigger success than I had ever hoped for.
As long as we had the contributors to keep building it, I would have
been satisfied with Leap continuing the 'trend of stagnation' or 'slow
decline' that the old stable releases were suffering from, and we
could have relied on Tumbleweed to provide the growth that openSUSE
needs to keep on getting new blood into the Project.

But instead of that, Leap has bucked the trend and exceeded my expectations.
42.1 and 42.2 both showed signs of halting the stagnation/decline, and
both showed signs of increase in the metrics we measure.
42.3 was the first release since 2009 to totally reverse that trend
and grow significantly in it's own right across every metric we have
been able to measure so far (Release downloads & repo users).

This is despite Leap 42.3's marketing, on purpose, taking a
significantly LESS detail orientated approach.
Instead of bombarding uses with endless detailed outputs of every
package version, complex deep diving sneak peaks and other such
breakdowns we used to do for every release (and like people are
advocating here for the website), we purposfully took an approach to
42.3's release marketing to follow the same approach as the current
website.

Like the website, we focused Leap 42.3's marketing on the general
'themes' of the release. It's closeness to SLE. The general topics of
how and where we expect users to find Leap 42.3 useful.

And the results have been greater than we expected. The numbers of
users, and the sustainability of the development process, both speak
for themselves (and we need both for openSUSE to continue long term)

So, to summarise

- the general direction of Leap was seen by the Board as a natural
decision in the light of the realities the Project faced at the time
- these realities included what could be described at best as
'stagnation' or at worst as a 'steady decline', depending on how you
wanted to cut the numbers.
- the alternative to Leap would have been the cessation of openSUSE
producing a regular release distribution, due to lack of contributor
interest.
- since launching the new openSUSE website we have seen both increases
in downloads & users to both openSUSE Tumbleweed & Leap
- Since 2015, the Project is growing in a way that we haven't seen since 2009
- Leap 42.3's marketing campaign seems to reinforce the hypothesis
that saturating potential users with too much information can actually
hinder growth, even when your target audience is Developers, Sysadmins
& Power Users.
- The website is fully open source and contributions are welcome at
https://github.com/openSUSE/landing-page

And so, that is the story behind what is going on here. While
obviously I think the history invalidates some of the concerns raised
in this thread, I'm open to discussing those points further, but not
on this mailinglist, and this is not the correct place, as already
stated.

Please relocate this discussion to opensuse-project@xxxxxxxxxxxx (if
you wish to discuss the Project's direction, Board's decisions and
such) or opensuse-web@xxxxxxxxxxxx (if you want to actually discuss
the openSUSE website but not contribute to it directly on
https://github.com/openSUSE/landing-page)

Regards,

Richard Brown
openSUSE Board Chairman

--
To unsubscribe, e-mail: opensuse+unsubscribe@xxxxxxxxxxxx
To contact the owner, e-mail: opensuse+owner@xxxxxxxxxxxx

< Previous Next >
This Thread