It's lovely to see such operatic vision! And I for one would love to see
some of those things happen.
But, just to bring it down a bit; the technological issues rear their ugly
heads. Engineering-wise, hosting Wikipedia is a tough problem. Distributing
Wikimedia hosting across the globe is very definitely a "hard" problem. If
it could even be considered in a 5 year project scope that would be IMO an
aggressive timescale :)
Also, I am not sure the WMF has attitude for decentralisation to chapters;
nota bene the work relating to Labs and Toolserver. So commercially that
might be a tough sell.
However, despite this, I hope enough people see something in your vision to
push forward change.
On 7 April 2014 14:39, Ting Chen <wing.philopp(a)gmx.de> wrote:
Hello dear all,
From 2008 on until recently the Wikimedia Foundation (WMF) had seen a
staggering growth to fulfill its mission, and it had pulled a great deal of
the resources, in money, but as well as in talent, manpower and volunteer's
effort of the movement.
From the beginning hosting of the Wikimedia projects was the core
competency of the WMF. A big part of the WMF budget and staff is dedicated
to the operation of the servers. Meanwhile the main server farm is moved
from Tampa, Florida to Ashburn, Virginia.
In the last years the WMF had evolved to the main development party of the
MediaWiki software. The software and product development had drawn many
resources and talents from around the world to San Francisco. Many
developers were relocated to join the WMF team.
With the increased prominence of especially Wikipedia the WMF and its
projects were facing more and more legal challenges in the past years. Law
suits from around the world were reported since 2005. Because of this the
WMF had expanded its legal team.
To improve its role as the leader of the movement and to settle the
disputes between the WMF and chapters about the processing and distribution
of the funding the WMF had evolved since 2010 into a grant making
All in all the WMF is without doubt the center peace of the movement and
claims four fifth of the expanses of the entire movement.
The recent dispute about the URAA motivated massive content deletions on
Wikimedia Commons highlights the problem of this strong centralized
In basic, the storage solution of the Wikimedia projects is still a very
classical approach with two central database centers, both of them located
in the US. This approach had repeatedly induced conflicts about what
content can be stored and what cannot. It does not reflect the
international character of the projects and had repeatedly induced critics
on the Wikimedia projects to be US biased and it is, measured on today's
storage technology, outdated. Even though currently the US law is one of
the most liberal in relation to freedom of speech it does has its bias. The
US copy right law for example is meanwhile one of the most restrictive and
backward looking copy right laws in the entire world. Another example of
the potential hazardous result of this approach are the image files that
are currently stored in the individual projects. For example on Chinese
Wikipedia images that are free according to the Chinese and Taiwanese copy
right laws are stored directly there, and not on Commons. These images are
nevertheless not free according to the US law and are stored in servers
that are located in the US and distributed from there. This poses potential
problems for all parties that are involved here: for the Foundation, for
the project, for the community that is curating these images and for the
users that are using these images.
In a larger sense the problem is not constrained to the file repositories,
but also to the content. Even though the Foundation had increased its legal
department and had tentatively tried to work out an approach to support its
community in legal conflict basically it is still working with the old
strategy: In case there is a legal case in a foreign country the Foundation
will avoid the call of the court while the Chapter will deny any
responsibility for the content. This leaves in the end all potential
hazards to the volunteer who contributed the content. In case of a court
suit he is probably the one that have the worse legal support and had to
take the charge privately, even if he handled legally and in good will.
In my opinion, since the technology is ripe, it is time for the movement
as a whole and WMF especially to seriously consider the approach of a
distributed hosting. Files and contents that let's say are legal in the EU
but not in the US should be able to be stored on a server located in the EU
and distributed and operated from there. Files and contents that are legal
in PRC and Taiwan and may violate copy right law in the US should be able
to be stored in a server say in Taiwan or Hongkong and be distributed from
there into the world. This approach is meanwhile technical viable and is
used by almost all major international internet providers today.
This also means that the chapters, as far as there is one, should be able
to take the responsibility for the content and the hosting of those servers
in their country. They should be obliged to provide legal consultation and
defense to the community, which means a distribution of the legal defense
from a central point into the world, to the chapters and directly to the
communities. Indeed the legal consultation and protection of the community
is in my opinion one of the most missed duty of the chapters and the
Foundation to the movement.
Every country, that meets a certain standard of freedom of speech, freedom
and media and freedom of justice is a potential place to set up such a
server and in which the chapter can be entitled to claim the responsibility
of the content that is stored there. There are meanwhile pretty many
renowned independent organizations that provide such standards and measure
the status of a country against these standards, like Reporters sans
frontières, Human Rights Watch, etc.
Also software and product development can be done distributed. Many
commercial companies do this successfully, many open source projects do
this successfully. The WMF is not unfamiliar with distributed software
development. One of the most prominent developer of the WMF, Tim Starling
is for example never relocated to San Francisco. Also in the past decades
many important impulses came from outside of San Francisco, the last one is
WikiData, initiated and developed by Wikimedia Deuschland (WMDE). Wikimedia
Serbia had offered in the past to hire developers in Belgrade because the
people there are well educated, talented and the wage there is low. I
believe there is no necessity to concentrated all developers at one place.
Fore sure distributed developer teams need certain trainings, standards,
communication skills and procedures to be able to doing well. But it is
possible, it is even meanwhile industrial standard. It is meanwhile a
backward looking approach to draw and concentrate developers at one place.
From organizational view it makes more sense to have these distributed
developers organized by the chapters (as far as there is one) instead of
let them work as contractors for the Foundation, which also means an
organizational decentralization of the software and product development.
For me personally there are some life experience that makes me an absolute
supporter for the decentralization.
I was born 1968, the year which marks the climax of the darkest period of
the Chinese history, the Cultural Revolution. In the year when I was born
Chine was experiencing the worst political purge since Stalin's death in
the whole world. At that time, no one could imaging, that from the boys and
girls that were born that year in China, millions will go to North America
or Europe to study there and work there and live there. No one could
imaging that some of them will go back to China because they know that
China will provide them better chances for work, research and life than in
North America and Europe. 1988 I traveled with the train throw Soviet Union
and crossed the no man's land of Berlin Wall, and at that time no one in
the whole world could imagine that less than four years later there will be
no Soviet Union any more and the Berlin Wall will fall.
Those experiences tell me not to trust any fortune teller and future
researcher. I won't bet that USA will not turn into a dictatorship within
my life time, and I won't bet that Central and West Africa won't turn into
the most prosperous and most liberal region of the world in my life time.
However unprobable this looks like. Because of that I don't trust one
central prominent hub, because however strong and well developed and well
organized, it is the single point to fail.
Decentralization, on any aspect, only works if the parties are aligned.
One of the darkest hour of my board chair's personship was by an interview
with an Austrian television. Together with me a chair's person of a chapter
board, a volunteer and a researcher of Wikipedia were interviewed. When the
reporter came to the topic of gender bias and Foundation's effort to
balance it he at first addressed the question to the chapter chair's
person. And the person answered: Well, for our chapter this is not a topic,
we concentrate our work on article quality. And for the next few seconds
before the question is addressed to me I was feverish thinking about an
answer which would not sound like I support and agree with him but also
don't like as if we will publicly take out a dispute about what is the
I think this should not happen. And if the movement really want to be
organized decentralized, we cannot afford such things to happen. It made me
sad to see that WMDE and WMUK published their strategic planning for the
coming years, each by themselves. I think it should be a strategic planning
with all organizations, agreed by all organizations and all organizations
will work together on those goals, together.
I think there should be a charter for all organizations in our movement,
signed by all organizations that want to join us, that set up standards,
set up things like working together on strategies and working together on
goals. Unfortunately, and I do blame myself partly for this, that despite
the movement roles work group, despite some other tries afterward, we were
not able to set up such a charter. And I think that one of the goals for a
movement strategic planning should be set up such a charter in the next few
So, if we decentralize the hosting, the software and product development,
the legal and the movement organizations, where is the place of WMF?
I imagine the WMF as the United Nations of Wikimedia. I can see a lot of
people now wrinkle their nose and say: What? that ugly and useless
bureaucracy? And I will tell you: No, I am not thinking about that ugly and
useless bureaucracy, I am thinking about that organization that
concentrated and coordinated the world's effort to eradicate smallpox, I am
thinking about that organization that set up standards to preserve the
world's heritages, the organization that coordinates and develops standards
for civil air and sea traffic that makes an smooth and safe international
travel possible. So, I am imagining an organization that coordinates the
movement resources, that set up and safe guard standards, but not
dominating the movement. And all in all, despite the 40 and plus partner
organizations, there are still more volunteers that don't have an
organization to support them, and there are still much to do for the
Foundation. Especially, I still see the WMF as the leader of the partner
organizations and the movement.
Looking back into the history I believe it is necessary for the Foundation
to have the last six year's growth. The Foundation had learned a lot from
this and it had repeatedly set up standards for the movement, despite all
the grudging and all the disputes, looking back, it is good to have those
standards set up. All organizations inside of the movement are profitable
from those standards.
But the growth of the WMF had more and more extincting the growth of the
partner organizations inside of the movement. Its dominance and its feeling
responsible for everything inside of the movement began to take the air
away from the other organizations, its concentration at one place had
always been felt as an alienation and is becoming more and more a problem.
A good captain of see knows when the wind turns and he need to change the
sail setting and course to cope with that change, for the Wikimedia
movement now is the time.
I want to repeat one sentence I said earlier: I see the WMF as the leader
of the partner organizations and the movement. I want to emphasize that I
want to see the WMF as the strong leader of the partner organizations and
the movement. The strong leader because he is wise and experienced, not
because he is a dictator; the strong leader that knows that every member in
his team has something that they can do better than himself and knows to
use those abilities in benefit of the group, and not the one who dominates
Wikimedia-l mailing list