Just this goy...

Wednesday, November 10, 2010

Memoirs of impressions of the Offline Strategy Task Force, as it stood in January 2010.

Any discussion of an offline strategy for the Wikimedia Foundation is, essentially, a discussion of technology and infrastructure. There are two primary target markets: those who already have access to WMF products, and those who do not. The largest offline product-consuming market will always be those who already have access to WMF products, but it is a primary element of the WMF mission to provide access to those markets who do not have access.

With the above assumptions, and the reality the Wikimedia Foundation is a charitable organization with finite resources, the primary strategy should be enabling third-party development of technologies and infrastructure for offline WMF-derived products. How the Foundation does so should not be restricted, but there are certain goals we found to be of particularly high value:
  • XML article format, which can support semantics, with project-specific DTD.
  • Create and publish a Mediawiki parser specification.
  • Create and support a reference parser, and companion writer.
Related to these goals we found supporting intra-project standardization, and implementing international information categorization schema, to be important efforts to encourage communities to pursue (and support them in pursuit of) in order to make their output more valuable for offline applications.

When considering how to make WMF published content available to people who do not have regular internet access, the largest potential market could be realized through cellphone platforms. In the most-underserved internet markets cellular phone service has good to excellent availability. Retaining the theme of supporting third-parties rather than investing extensively in technology, it's noted there are already several cellular-based products and projects using WMF content. This second strategy for the Foundation is to focus on the cellphone specifically as the world's most ubiquitous hardware platform, with several goals related to this publication medium:
  • Target upstream from end users: cellular network providers for dynamic content systems, and OEM for static content systems as well as built-in 'apps'.
  • Prioritize support for third-party development: open offline storage standards, readers which use those standards, and finally proprietary products.
  • Where possible, support non-internet distribution systems, for example an SMS article retrieval system.
School systems are a natural partner for WMF in regions under-served by Internet. They are centers for learners, a primary target audience for the Foundation's projects. Schooling systems themselves are the target audience for the Wikibooks project, as well as other more-specific wikiprojects. This tertiary strategy did not develop concrete targets, but rather a laundry-list of desired outcomes including:
  • Work with schools, and support organizations which do so.
  • Customizable content collations.
  • Use multi-project approaches: draw content from any relevant WMF project.
  • Use multiple distribution channels: Internet, static digital, print, &c.
Most WMF published content is general, and targets an adult audience. The offline markets with the lowest internet access rates are disproportionately younger, and content targeting a youth audience should be specifically supported. Third-party publishers tend to have target markets for whom they wish to tailor their content, as do schools and governments. Technologies already exist to provide best versions, as well as semantic tagging to improve article relevance. A fourth strategy focus involves improving the ease of creating custom collations, both through content initiatives and through technology development.

Tuesday, November 09, 2010

Sue Gardner's recent [blog entry] discusses what she sees as "changes that stuck", highlighting the [[en.Wikipedia]] issues of Biographies of Living People and license migration, and the supposedly Foundation-wide Strategy Project.

It's my opinion the 9 points she raised are fundamentally wrong as regards that latter 'change that stuck'.

As someone who was involved in the Offline Task Force, devoting rather a large number of hours to research, consultation, and discussion as well as editing our recommendations, [my impression of the process] is it was futile, a waste of our time, and ultimately co-opted by WMF staffers who altered, changed, deleted, or substituted their agenda for the issues discovered, researched, and presented by the community. Perhaps one cannot test the quality of the strategy until after it is begun to be implemented, but one can certainly judge the honesty with which it is collected and reported.

Take, for example, [[Offline]], the primary link for offline Strategy. All but three edits are by staff members, nearly 90% of the content was written within the past 6 weeks. Only one of the contributors ever attended an [[Offline Task Force]] [[meeting]]; Philippe, who hasn't touched the page in more than 6 months. And nothing on the page is related to the Offline Task Force's [[recommendations]].

  1. The person/people leading the change did wait for it to happen naturally, then like a wiki-troll, they co-opted the community process and replaced the findings of the task force with their desired outcomes. Maybe only this one task force, but I wouldn't trust it to be such a unique situation.
  2. A single person didn't make it happen, but a group of employees did.
  3. There is no evidence of any effort made to understand the global/meta issues in the current pages. The trail of evidence disappears back in April, to be replaced in October with atheoretical and unsupported conclusions.
  4. If the process was carefully designed to ask the right people the right questions at the right time, this was a carefully designed ruse to convince those volunteers and contextual experts that their efforts would ultimately shape the documents guiding Wikimedia Foundation's strategy 2010-2015, for none of the research is represented in the current pages.
  5. Lots of people did lots of real work. No dispute about that. But "making it happen"? As far as the Offline Task Force is concerned, clearly what we found should happen is never going to happen because it has been excised from the offline strategy pages.
  6. No, it appears almost none of the work which has gone into the current offline strategy has taken place in public. The primary contributor, a new employee, has never edited on a single talk page, has never asked a question nor answered one. So where did all the verbiage come from?
  7. Yes, almost every bit of work on the current offline strategy took place in private, probably within the Wikimedia Foundation offices. No, [this is not a good thing]. To be blunt, leadership is about making decisions when they need to be made, publicly, and being responsible for them. It is better to make a bad decision than to not make a decision. As a sailor, often I must make commitments without enough data to be sure of safety, and yes I have run aground. But I cannot get anywhere if I wait in port until I know everything about the route, the weather, and the destination anchorage because I can never know enough to eliminate all risks of error or mischance. Operating in private until you won't be embarrassed means you will always be [operating secretly].
  8. Did people put their credibility on the line? That's a tough call. Sue Gardner put her credibility on the line when she said Strategy would be a community-centered and driven process, with results drawn from the community. As regards the offline strategy that is clearly not the present case - none of the findings of the community task force are represented in the strategy documents.
  9. Most people wanted - more than anything else - to advance the Wikimedia mission... Hmm. I think, among the most active members of the offline task force, I can agree with this statement. But I don't believe that is true of a majority of those involved with the offline strategy, certainly was not true of the majority of the individuals I interviewed while doing research related to the strategy. And there is a huge caveat: to advance their personal interpretation of the Wikimedia mission, however illogical or distorted that interpretation might be.
Maybe Ms Gardner thinks this pattern is a positive model. Ultimately, she claims responsibility for both the conception and implementation of the Strategy project. To me this means she is personally responsible for the current situation regarding Offline Strategy. She is responsible for encouraging the community involvement, the research, but also the disregarding of both. As a "change that sticks" I would see it as an anti-pattern.

Saturday, August 28, 2010

Otourly's blog entry translated to English:

A circular irrigation water distributor in Taketa, Oita Prefecture, Japan.Water distribution at Taketa in Japan - public domain - Picture by Tsutsui Mizuki - And it make me curiously think about wikimedia commons' logo...
When we speak about Wikimedia Commons to Wikipedians, not natively anglophone in particular, they say Wikimedia Commons is good but too English-centric, so they use it only for essentials. But, the search tool of the site is too English as well so that it is really difficult for a non-English speaker to find what he is searching for. Briefly, an English search request on Commons will be generally more precise and find more matches than an equivalent request in French.

The true problem is not the English, because there must be a common language to link all the projects, but the problem is the categories.

To explain: the categories on commons are very complicated for someone who is not at ease with English. Often this person will only add one category (one of the main aims of the media collection) or even none to files they upload, leaving the confirmed contributors to clear up according to their English level. And I can assure you that it takes time, and sometimes these contributors would rather take photos, uploading photos, write descriptions, or revert some vandalism... I sometimes wonder how many confirmed contributors have less than 5% of their edit count fixing categorization...

Often my uploads define the categories that I will fill, correct, and, in a word, improve. There is meanwhile a tree view of the Commons Categories that I follow thanks to a little javascript code, created by [[User:Chphe]] called [[SuiviCat.js]]; a French name and as a consequence not a lot used on this anglophone project. I still remember when I requested the adaptation of SuiviCat for Commons...

In fact if we have a look outside the Wikimedia Foundation, I mean on other media websites (such as FlickR), the category system is often replaced by a tag system. Tags are useful, quick, simple... but tend toward disorder, in particular with homonyms. Wrong path, so back to our beloved categories.

In fact categories, to definitively solve the problem, should ideally be 100% multilingual. But this require sysops who are multilingual too. Otherwise vandalisms in Hindi could certainly be more difficult to detect... At the moment its seems to be an unattainable Utopia, but fully multilingual categories would have a big advantage; this could reconcile the different contributors of Wikipedias, Wiktionaries, Wikibooks, Wikinews...

Wikimedia Commons would became a truly international project, the research would be more usable by everybody. For all Wikipedias, each contributor would understand categories and subjects of the media... Besides, often it is the categories alone which best describe a file. For the good reason that not everyone is named Otourly, and as a consequence not everyone everyone links to Wikipedia in the Commons descriptions. Worst, sometimes there is only a minimal description, which does not describe it very optimally.

But it is not the fault of the English language, I must admit it... If we have a look this file for example : [[File:Church_of_the_Nativity_of_the_Theotokos_(Gora_Pnevits)_05.jpg]] only the title and the category are understandable for someone who has only few notions of English. There is no link to Wikipedia in the description which could help us to know what it is exactly.

But reconciling Wikipedia and Commons is possible, anyway, the [[Projet Monuments historiques]] is a good example; who better than French people to take photographs, sort them, geolocalize them (or just geolocalize categories) and offer historical monuments in France to everybody? It is true that these files on Wikimedia Commons are often only described in French, but they are mainly well categorized.

Monday, August 23, 2010

This blog entry is for Danny Pickle.

You signally failed to ask "Why the hell do you think anything related to humans you can sense is free culture?" I'm guessing you just decided to dismiss me and the whole conversation as a troll, unless you went to bed mid-discussion.

Like telling everyone to refuse to use proprietary software in developing free culture - as though this would suddenly make the proprietary hardware irrelevant. Oh, sorry, I shouldn't expect consistency even though you do.

The Suzuki recorder's shape is trademarked. Did you know that? Its exact ratios, decorative elements, are a corporate asset. And if I composed a jingle on it, slapped some free license acceptable to FSF on the tune and published it far and wide on the internet, the tune would be free culture. The tool has no influence on the product - I could use the instrument to write a melody for the piano, or even a Bede-esque poem composed of tone phrases.

It would be free culture, even though it undoubtedly would be influenced by Pachelbel's Canon in D, Queen's More of that Jazz (Youtube), and Joplin's Pleasant Moments. Its derivation from these intellectual properties - for that is what they are, whether or not currently protected by copyright law - has nothing to do with whether it is free culture. The idea that something is not free simply because of some artificial legal proscription is small-minded and just as artificial as the law cited.

As a digression, in the mid-90's in the USA it was illegal to "desecrate" any of the thousands of minute-man statues the US military dotted about the country's university campuses wherever they happened to have have a reserve officer training program. At the time there was a wave of protest on campuses against allowing the programs standing in accredited schools which had anti-discrimination policies - the US military clearly discriminates on the basis of gender and sexual orientation. Time after time activists requested permits to hold demonstrations in front of ROTC offices, and were turned down.

Rather than follow the rules, I and friends posted a notice in the paper of our celebration to be held in front of the ROTC offices. We then went down and enjoyed an afternoon of diversity, including dressing the statue up in a nice dress, while the ROTC students mustered at the side of the field. After the festivities died down, we cleaned everything up and went home, and sent out thank you notes to the CO and the University.

The cogent element of the event should be clear: we did not assume we needed permission to gather, to dress up (without causing permanent harm) the statue, or to carefully record our actions and then distribute the pictures/tapes - all of which the law clearly proscribed.

If you haven't gotten my point, I should probably be more blunt: there is no way any cultural item can avoid influencing its audience. Just as all video games are, in a sense, derivatives of pong, so every major software or artistic or hardware development has illegitimate offspring in the eyes of the law if literally and exactly enforced. But it would be impossible to prevent every infringement, especially if everyone just ignored an untenable legal position and just did it.

Trying to define what is free relative to an illegitimate legal standard is far more harmful than refusing to accept that illegitimate legal standard in the first place.

Thursday, March 04, 2010

I wish I'd written this:
Unfortunately, what the CNCI document describes sounds like the opposite of innovative, which isn't a surprise, given that the impulse to conceive and develop it entirely in secret was the opposite of democratic. In the end, there's no better way to stifle innovation than to let a single entity control it. But that's the kind of wrong-headed groupthink that takes root and grows in small groups that are restricted by excessive secrecy.[1]
This is exactly why I think the secrecy of ComCom - and the WMF public relations efforts in general - is stupidly wrong. It discourages innovation, and attacks constructive criticism. And it does so for no measurable benefit.

If you are not doing wrong, transparency can only help you. Living in a fishbowl can be disconcerting, but if you develop a thick skin and work directly toward [[SMART]] goals - avoiding digressions and distractions - you accomplish much and gain a reputation for honesty and resolution.

Are there topics which should not be discussed publicly? of course. But nothing else should be confidential, full stop.

Wednesday, March 03, 2010

English encyclopedia articles should be understandable by non-specialist native speakers. If the topic or concept cannot be given a generic overview readily understood by a layperson, it is not a topic to be included in a general encyclopedia.

[[L'Hôpital's rule]] clearly fails this metric.

(The counter argument - that a specialist encyclopedia may cover its subjects at any depth - is a [[strawman]]. A specialist encyclopedia has a targeted audience; it is in fact the definition of such a work.)

Sunday, February 14, 2010


okay, so I've been wrestling with this script... it's blown up a couple of times after many many hours of work, and I've done a number of not-very-pristine-code bits to get it back up and running. I'm not good at babysitting a script.

Anyway, after getting home and seeing it dead and fixing it and rebooting it with toes crossed... I just discovered a logic failure in one of my four primary categories. "Former" indicates a user with plenty of activity on one or more Wiktionary languages, but on which the user has not edited within the past year. However, the script does not value recentness of editing activity over number of edits - so someone who has been editing (albeit rarely) on a language but was formerly active on a different language will be listed as formerly.

That's more of a value judgement than a logic problem, I guess. It's just that I personally would value recentness over number of edits.

Saturday, February 13, 2010

Nothing quite like working the hell out of a new piece of code.

I've been casually working on a php mediawiki API class for, uhm, months. But got motivated this past week and rebuilt it a lot. Decided to hammer out a quick proof-of-concepts today and ended up with some junk code which will need to be rewritten.

On the other hand, right now it's chewing through a 1013 long list of names, checking the contributions of each on up to 171 different Mediawiki website, and for each site it's creating an instance of my class, connecting and logging in, requesting the name's contributions, parsing them if any, logging out, and moving on to the next site (assuming the user has not made enough contributions.) With so many website connections and communications the script is not quick, and each user can take several minutes.

I think I can safely say the log-in and user contributions routines are good.

On the other hand, the editing routines haven't been touched yet. They don't get any use until after working through the list of users the third time (first time builds the list of unique voters, the second time builds the list of users ordered by how active they are on Wiktionary projects, third time builds the wiki syntax for writing to the wiki.) By my estimations, three and a half days from now it will (finally) get done. That's a pessimistic guess though.

So I'm just watching the terminal as it ever-so-slowly scrolls along... we're all the way up to the An* usernames.

Monday, February 01, 2010

I love Strunk's Elements of Style. Yes, the 1918 version, before E.B. White joined his name to the foundational style text. Most especially, I love Rule #13, and this brilliant one-paragraph essay on writing:

Vigorous writing is concise. A sentence should contain no unnecessary words, a paragraph no unnecessary sentences, for the same reason that a drawing should have no unnecessary lines and a machine no unnecessary parts. This requires not that the writer make all his sentences short, or that he avoid all detail and treat his subjects only in outline, but that every word tell.

Today I received the following, excerpted, bit:

I think reducing the bylaws to the boilerplate in order to speed up the process would be overkill. Can we look through Amgine's minimum version and make sure that every minimum clause is represented in the longer draft bylaws?

Can anyone explain to me why one would wish to include more than what is necessary?

Monday, January 25, 2010

I've tried to approach this subject three or four different ways. Face it, my writing skills suck, and I'm far too blunt at this time of night. Stripped of all rhetorical embellishments:

The Wikimedia Foundation Strategy project is [still] doomed.

This is not the simple pronouncement of someone who hates the idea, has done nothing to try to make it work. I put my bits of sweat and research into the process. I have every hope some vestiges of my efforts will make it into any reports, proposals, and recommendations.

My conclusions really bother me. While reading on the project, and working on a task force, I came into contact with a number of very wonderful people. While not entirely selfless, they were working in good faith on something they view as larger than themselves, larger than the Foundation itself really. Their visions of the project are inspiring, admirable, even at times beautiful.

I just don't think, in the larger picture and over the longer term, this project is very relevant to the future of the Wikimedia Foundation. Here are some of my reasons, which I think have evolved since my last entry on the subject:

  • There is nothing tying any of this work to implementation. Not only is there no mechanism or created process going from envisioning to application, there is no transparent and public statements that anything will be implemented. There is no moral difference between the task force's efforts and [[Walter Mitty]]'s fantasies - that is, escapism from the reality of the Foundation as it is.

  • The process fundamentally alters the relationship between the Wikimedia Foundation and the communities which it formerly served. With this process the future of the projects is clearly top-down, from the ED's office to the people who volunteer, rather than bubble-up from the people to the office. This may reflect a reality of long standing. That does not make it any more palatable. In fact, it fucking sucks.

  • For all intents and purposes, the entire point of the Strategy project is to talk about other people doing things. This is 100% in opposition to the Wiki Way. In the Wiki Way, people work on what interests them. They do not work on what other people think they should work on.

  • For all intents and purposes, the Strategy project is about talking about changes. I hate that. In my opinion, if you see something you'd like to change, you act on getting it changed. {{sofixit}} is the mantra of wikis. If anything is my primary point here, this is it.

There is one further point to be made: Everything is about Wikipedia.

No, to be honest, very little is about Wikipedia. At least, not if you're not sucked into the narrow, focused, blindered [[weltanschauung]] of the Wikimedia Foundation. Schools in third world countries don't want wikipedia: they want to teach students. Google news doesn't want wikipedia; they want free backgrounder information which is hyper-current. Random readers don't want wikipedia: they want reasonably reliable information instantly on the topic in which they are interested [this microsecond].

But wikipedia contributors - like contributors to each of the projects - are all about wikipedia.

Sunday, January 24, 2010

en.WP should not contain literary criticism. Not that it is good or bad, but it is not encyclopaedic.

The title [[If on a winter's night a traveler]] is a good indicator of this novel which is reminiscent of Laurence Sterne's Tristram Shandy. The book commences on a hypothesis of novelistic elements ("If...") on a when, a someone...would do what? According to this book, the entire novel, even its plot, is an open trajectory where even the author himself questions his motives of the writing process. This theme — a writer's objectivity — is also explored in Calvino's novel Mr. Palomar, which explores if absolute objectivity is possible, or even agreeable. Other themes include the subjectivity of meaning (associated with post-structuralism), the relationship between fiction and life, what makes an ideal reader and author, and authorial originality.

Wednesday, January 20, 2010

Communication to and from the Wikimedia Foundation is strictly controlled. This includes all communication to the various communities, and from them. It is impossible to tell just how completely this practice of filtering and manipulation of communications streams is happening within the Foundation, controlling what either the left or right hands know or believe to be true[1]. A fair assumption is: probably most of it.

No one I have spoken to was able to name a single member of the Communication Committee who is primarily a Sister's Project community member. Let me restate that: there is no official communications link from the Foundation to any non-Wikipedia project. Full stop.

Listening in on various communications tools one is struck by the silence. Comproj seems to have had exactly one real application in about a year: an attempt to organize PR materials in one, central location. Of course the Communications Committee list is restricted to members.

An interesting - and telling, imo - symptom of the balkanization of the Foundation is the Wikimedia-Boston mailing list: Social list for Boston-area Wikimedia gatherings. "This is a list of Boston-area Wikipedians and other Wikimedia supporters."[2] - after all, some projects are more equal than others.[3]

However, speaking of Wikimedia-lists, there are plenty of them. Ostensibly the naming convention suggests they are to be used by/for Wikimedia Chapter formation/internal communications; few seem to actually be actively used for such. A relatively high percentage of the lists restrict archive access to members, so it's impossible to generalize accurately, but of those countries which do have a Wikimedia chapter few have any appreciable use; several have never had a single message posted. A very tiny sampling of lists in English show almost no Foundation<->Chapter community communications, with the noteworthy exception of WMUK.

So how is the Foundation communicating with its contributors, especially its niche communities?
  • Secret communications
    By far this is the most common channel for dispersion of information. Via restricted-access Wikis, restricted mailing lists, restricted IRC channels, private e-mails and discussions. In the recent Fundraiser dozens of individuals - site bureaucrats and community leaders - were quietly threatened via back channels to prevent any site modifying the centralNotice Banner Advertisements even though these ads were often offensive to the local contributing editors. This highly-successful intimidation campaign is only one example of the coordination of WMF communications out of sight.

  • Official communications:

    Board resolutions, press releases on [[Foundation wiki]] plus pronouncements on Meta or en.Wikipedia are probably the second largest channel for "communicating with the masses." These extremely impersonal messages dehumanize the Foundation; it becomes the corporate machine dictating to the projects, with no chance of response.

    The Mediawiki techs are a hugely important element of the Foundation, and they use [[bugzilla]] and [[wiki]] as their formal communications routes, [[supplemented]] with [IRC] and [[mailing lists]]. However, the reality is the bugzilla is ignored in favour of internal/personal communications networks in determining tech priorities. Even though the developers tell you the best way to get an issue addressed is to define the problem, develop a solution for the problem, and submit it this is not true: the best way is to throw a hissy fit involving dozens of well-placed people and make a public-relations headache - at which point your issue will magically develop priority and may actually get attention. (no guarantee it will actually be resolved, however: examples include Wiktionary's [[Extension:Transliterator]], community approved in August 2009, committed in October after languishing for months, never reviewed or approved by WMF devs as of this writing.)

  • Semi-official communications:

    This is the human face of the Foundation: staff members like [[Bastique]] and [[Philippe Baudette]] who every day work in contact with various contributor communities, and serve as channels to bring issues into the Foundation staff meetings, and give depth and interpretation to the Foundation's less personal messages.

    These individuals have both an inappropriate stress and authority due to the public perception they are speaking in a semi-official capacity for the Foundation, which is increasingly remote from its average contributor.

  • Informal communications:

    This is the way work actually gets done, no matter how carefully communications channels are monitored and prescribed. A phone call from the right person, an SMS heads up to someone who should be aware - these are the real networks which need to be recognized and - where appropriate - encouraged. They are prone to abuse, and to causing as many problems as they solve, where impediments to constant sharing of information are put in place.

    There are occasions where communication must not take place in public; they are the rare exception. There are occasions where communication is one-way - an instruction or announcement - but more commonly it should be a discussion with mutual agreement as to what was said, what compromises reached.

    Unfortunately, this channel is the least accepted or exploited by the Wikimedia Foundation, and in fact it appears the Foundation is actively campaigning to restrict informal communications between itself and the contributing community for whom it exists.

Well, there've been a few months of frenetic activity involved with the WMF which should be blogged about, so I'm working on three articles atm.

  • Wikimedia Foundation Strategy: Excellent goodwill and good faith.
  • On Mediawiki Bots: Mote, motivation, métier.
  • Communication Corps

About Me

Owned by Njørđson, a Cape Dory 25D.