Mailinglist Archive: opensuse-project (230 mails)

< Previous Next >
Re: [opensuse-project] openSUSE 2016: taking a picture of openSUSE today

On Thursday 28 November 2013 08:26:54 Andrew Wafaa wrote:
On 26 November 2013 19:38, agustin benito bethencourt <abebe@xxxxxxxx>

Once openSUSE 13.1 has been released, it is time for the openSUSE Team to
focus on the future. We want to share some ideas we have about the project
in general and factory in particular. The topic is not easy. so this mail
is a little long and dense, but hopefully worth it. It won't be the last
one so let me know how to improve it.


This is the first of a series of mails we will publish the following days
with different ideas. The process we are proposing has no intention of
pointing at anybody, revisiting the past or enforce any situation within
the community. Our goals are:

* Share a picture as a starting point of discussion.
* Use the discussed picture as a reference to agree on actions we all
can/want to execute.


One of the first things we did was digging into numbers that provided us
information about the status of the project. Data cannot be the only
source to create a complete picture, but it is helpful as first step.

In order to better understand the rest of the mail, you probably want to
look the following references:

* Alberto Planas talk at oSC13: openSUSE in Numbers[1]
* Alberto Planas' slides from the above talk[2]
* First openSUSE Team blog post: Numbers in openSUSE[3]
* Second openSUSE Team blog post: More on statistics[4]
* Jos post about numbers[5]

One important note about the numbers: since most of the behaviors of the
variables reflected on the graphs were consolidated, at some point we
decided to stop adding effort in collecting numbers until 13.1 was
released. Once the Release is well established, we will update them and
evaluate the influence of this Release in the global picture.

I won't try to go very deep in the analysis. It would be too long. There
are many interpretations that can be done based on the graphs. I will
just point out the most relevant for our purpose. Feel free to add
Statistics are a marketing tool, and generally everyone acknowledges
they can mean whatever you want them to mean and as such are
realisticly useless.
We (openSUSE) have been making a lot of noise about how we have more
users than Fedora. I don't really believe that stat, but I'll go along
with it for now; one thing that we m,ost certainly do NOT have is more
developers using openSUSE compared to Fedora.

This is confirmed by our data. As mentioned by Alberto in his keynote, I think,
we used Fedora as reference because the rest of the distros do not publish
their numbers together with the methodology and the technical info required
for us to compare it.

Ultimately this is what
we should be looking at doing growing openSUSE as a good developer
platform, not just as a good end user platform (Windows desktop

I think both are deeply linked. We cannot have one without the other. We need
to work on both rather than the current compromise we have. That is a relevant
point of the proposal.

Following Alberto Planas' order from his slides[2]...

1.- Downloads

The number of downloads do not measure our user base, but provide hints
about the impact of the work done every 8 months, the potential new users
we might bring to the project and, looking at pre-release downloads, the
number of testers.

Taking a look at the graphs, we can see that the overall number of
downloads is growing at a slow path (slope). This behavior is not
consistent in every release. For instance, 12.1 was more downloaded that
12.2 or 12.3. More and more people uses zypper for updating the
distribution though.

Not sure I understand what you're trying to get at here. Sure, keep an
eye on trends, but that's all what they are. Don't fixate on it.

The overall number of downloads grow if you take the full time window shown on
the graph. If you count the number of downloads per release, you would see
many variations from release to release.

@Alberto, am I correct? I am sure you can explain better.

2.- UUIDs (installations that update regularly)

* Looking at the number of machines that regularly update against openSUSE
repositories (daily, weekly and monthly), we can easily conclude that the
situation is very stable. The speed of growth (daily and weekly stats) or
decline (monthly) is low.

* What the graph do not show is the acceleration. It has been negative
(small in value) for quiet some time now.

Your stats don't take into account people that update from
local/on-site mirrors, this is a very common thing in areas with
slow/un-reliable internet.


On the other hand we are counting people that have several virtual machines in
the same iron. It is the best we can do for now. This is why we approach the
numbers looking for trends and consolidate behaviors, not looking for details.

* When looking at the architectures, we see that x86_64 is more popular
than i586. This behavior is accelerating, as confirmed in the download
numbers collected for 12.3

* When looking at the mediums where those installations come from, we
clearly see three dominant ones: .iso (dvd version), ftp (net installs)
and Live CD.
This is nothing new, and was known before. We already knew the trend
was in x86_64's favour, and we adjusted the media available as a
result to reflect the three you mentioned.

The shift in the downloads of openSUSE happened in 12.3, not so long ago.

* There is a relevant detail that Alberto mentioned in his talk. More than
half, almost 2/3, of openSUSE installations are not using the last version
many weeks after Release date. There is also a significant amount of
installations using unmaintained or Evergreen versions.

3.- Factory and Tumbleweed installations/"users"

Factory is our ongoing development effort. As you can see in the graph,
number of Factory installations is constant. Tumbleweed was very
when it came out. Many developers and bleeding edge users liked it. Its
popularity is decreasing though.

Factory is almost always a constant percentage of installs vs the
number of contributors. We haven't been making much noise about
Tumbleweed so it's only natural that its usage will drop if it isn't
known about. Many new users know about the wealth of repos on OBS so
they just add the repos they want, also Tumbleweed isn't suited to
those with proprietary driver needs, which unfortunately is a fairly
high number.

There are several possible explanations for Tumbleweed number of users
behavior. I am not qualified to provide a solution here. I think this is one of
those points for discussion at the Tumbleweed mailing list, for instance. I am
interested on the conclusions of such a debate.

4.- Contributors to factory and devel projects

The numbers of users that are submitting request to factory/devel projects
is increasing. Now we have more non SUSE contributors. SUSE ones remain
constant. The overall growth is about 27 new contributors per year, a
little bit more than 2 new contributors per month.

Why have the contributions from SUSE remained constant? Surely we
should see a growth across the contribution base?

Ralf Flaxa pointed reasons in his keynote and what is his intention in this

It is a challenge all the Free Software companies (also close ones) face. They
need to maintain code for a long time for their customers while the
development area moves fast and you have limited chances to control that

The success or failure of a software company in a long term depends also in
how they face this challenge.

5.- Social media and comparison with Fedora

openSUSE is, in the social media channels evaluated, in the range of
Fedora. Comparing our numbers, I guess we all agree with this general
trend that states that openSUSE is a more user oriented distribution than
Fedora is. We have less downloads but more users (installations updating
This is a pointless metric. Yes we need to evaluate our standing with
our peers, fixating on it is going to do nothing good and will detract
from stuff that needs to be done.

The compassion is what we think is valuable. Maybe it should be extended to
other distros. We are afraid we might not like some of the results ;-)


All the above pieces shows a stable picture. Every sign of growth or
decline is, in absolute and/or relative numbers, small except social
media, due to their explosion as communication channels (which I do not
think is way different from what other Free Software communities are
I disagree, I genuinely think the picture is positive and not stable.
We have more growth than you make out, this was pointed out to you at
oSC13 but you chose to dismiss it.

Can we agree at least on the idea that we have much more potential that what
we are delivering right now? Can we agree that this is a good time, from the
internal and external perspective, to make a move that allow us to increase
our relevance as a distribution?

If we do, disagreeing on the current picture is not optimal, but we can share
the ride.


openSUSE coexist with other "coopetitors" (Free Software competitors +
cooperators) and competitors (closed sources distributions).
Touchscreens, cloud, big data, games...the Linux ecosystem is evolving and
there are new users with new needs.

New players are consolidating their positions: Arch, Chakra, Mint...
Ubuntu is moving to the mobile space, Debian is getting some attention
back from previous Ubuntu users....

On the other hand, some distros that were relevant in the past have
disappeared, our 13.1 has got more attention than previous ones, SUSE is
healthy and willing to invest more in openSUSE in the future ...

In the above context, how is our "stable" situation perceived? How
do we think it should be perceived?

The perception is positive, they see good improvements with only minor
niggles. Many peers regularly look at what we are doing and feed off
of that. I don't think we should worry about what X perceives about
us. We know who/what we are and what we stand for.

Worry...not. I agree. By putting an eye in what others are doing and how they
are solving similar problems, what results they get..... that is something
that I am interested by default.


If we agree that the overall number of users of Linux based server +
"traditional" desktop OS (let's remove the mobile/embedded space and
cloud for now), is growing, not following the "market" growing trend might
be perceived as a wake up call, a clear sign that improvements needs to
be done.

But if we agree that we are playing in a risky and challenging field,
stability can be perceived as a healthy sign.

After these months of analysis and discussions with both, contributors and
users, I would like to ask you if you agree with the the idea that the
first picture is more prominent than the second one. But, does the second
one provide us a good platform to improve our current position?

We need to be in both "pictures". SUSE and openSUSE used to be
innovators, we need to get that innovation streak back. We are growing
and we are stable, we now need to spice things up and become a

I agree with you. This is not about fixing our problems, it is about deprecate


Let me propose you some questions:

1.- What other variables we should put in place to create an accurate
picture of the current state of the project?

An easy way would be to actually talk to people at events,
FOSDEM/oSC/SCALE/etc and listen to them. Don't try and put words in
people's mouths. Take a notepad and write down everything you hear.

2.- What is the perception you think others have from the project?

What does it matter what I think others think of openSUSE? I'm not
insecure about openSUSE, sure I sometimes wonder WTF is going on at
SUSE but openSUSE gives me no concern as I have a lot of trust in
those that actually call the shots and are willing to stick their neck
on the line.

3.- What is your perception, your picture?

As I said, we are strong, stable and growing. We now need to
capitalise on that and start shaking things up.

To get some context you might want to take a look at the following

* Current strategy[6]
* Ralf Flaxa keynote at oSC'13[7]
* Jos article: Strategy and Stable[8]
* Jos article: Strategy and Factory[9]


Please point us to other relevant references:

[1] Alberto Planas talk at oSC13: openSUSE in Numbers:
[2] Alberto Planas' slides from the above talk:
[3] First openSUSE at SUSE team blog post: Numbers in openSUSE
[4] Second openSUSE at SUSE team blog post: More on statistics
[5] Jos article about numbers:
ng.html [6] Current strategy:
[7] Ralf Flaxa keynote at oSC'13:
[8] Jos article: Strategy and Factory:
[9] Jos article: Strategy and Stable:

Agustin Benito Bethencourt
openSUSE Team Lead at SUSE
To unsubscribe, e-mail: opensuse-project+unsubscribe@xxxxxxxxxxxx
To contact the owner, email: opensuse-project+owner@xxxxxxxxxxxx

Agustin Benito Bethencourt
openSUSE Team Lead at SUSE
To unsubscribe, e-mail: opensuse-project+unsubscribe@xxxxxxxxxxxx
To contact the owner, email: opensuse-project+owner@xxxxxxxxxxxx

< Previous Next >