Rui Correia wrote:
... tell me whether we are bleeding new or old members.
http://strategy.wikimedia.org/wiki/Editor_Trends_Study/Results
http://meta.wikimedia.org/wiki/Research:Editor_lifecycle
and
http://meta.wikimedia.org/wiki/Research:Editor_classes
agree: we lose experienced editors at about the same rate we always have, but what plummeted after 2007 is the rate at which we attract new editors. That's why there was so much enthusiasm for the Visual Editor, but it was misplaced because being able to figure out wikitext is an excellent attribute in new editors (analogously, being able to figure out that wikitext has ambiguities equivalent to the halting problem would have been an excellent attribute in VE architects....) None of the other technical solutions (Huggle, Wikilove, two click thanking, etc.) have made a dent in the numbers, so it is time to consider this the social problem that it is, and not just some technical problem that can be coded around with a fancy new feature, fewer bots, or addressed with nicer template warnings. Since the typical editing tasks continue to transition from creating new articles to maintaining the accuracy of old articles, that is even more reason to want to attract highly educated editors who will be able to overcome technical learning curves and social hurdles with their own minds, not a Mediawiki extension.
Consider the supply and demand of both editors and their leisure time by educational attainment:
http://www.informationisbeautiful.net/2010/cognitive-surplus-visualized/
There is no shortage of new editors to attract. But how much free time do those potential new editors have? For the typical highly educated potential male editor, or the potential female editor of any educational attainment level in the vast majority of the English-speaking world, things have been getting a lot worse:
http://cdn.theatlantic.com/static/mt/assets/business/Screen%20Shot%202013-05...
http://cdn.theatlantic.com/static/mt/assets/business/Screen%20Shot%202013-05...
(from http://www.theatlantic.com/business/archive/2013/06/how-did-work-life-balanc... in case those URLs expire)
These are all pertinent to whether strategic priorities should include direct action to improve the extent of leisure time among highly educated people in the developed world. Do that, and there will be plenty of new Mediawiki and Wikidata extensions to choose from as potential symbiotic solutions to both editor recruitment and the transition from creation to maintenance. If I had more free time, I would do this one:
http://strategy.wikimedia.org/wiki/Proposal:Develop_systems_for_accuracy_rev...
That is on topic, because if we had that feature, maintaining accuracy would be a lot easier in that it would take less volunteer time. But I don't think for a minute that any of the external strategic priorities I've listed would do less if they came to fruition.
Best regards, James Salsman
On 05/29/2014 08:57 PM, James Salsman wrote:
but it was misplaced because being able to figure out wikitext is an excellent attribute in new editors
I think that statement fails on two aspects: for one, saying that the enthusiasm 'was misplaced' is rather premature as VE itself is rather incomplete - we do not yet know its potential.
Secondly, and more importantly in my mind, "being able to figure out wikitext" might be a good attribute, but making it a requirement pretty much sacrifices any hope we have of getting rid of our systemic bias. The vast majority of the planet cannot - or will not - have the time and resources to learn an arcane and overcomplicated mishmash of markup languages; yet many of those have knowledge and skill to share.
In 2004, when articles were mostly unformatted, that argument made sense. Most anyone with minimal computer skills (and that's already a very restricted slice of the population) could edit a page to fix a typo or add a statement or two without much difficulty.
Nowadays? Not so much. For the untrained eye, even finding the glaring typo you saw in a reference is nearly impossible after you hit the edit button.
-- Marc
"even finding the glaring typo you saw in a reference is nearly impossible after you hit the edit button." -- Marc
Yes, it was, as references were getting longer and longer (almost to the point of including the author's likesa and deslikes and what he or she had for breakfast. That was 'solved' by the new <ref=.....>, that is really not the easiest to figure out. And oddly enough, I don't ever see anywhere any form of a tutorial on changes - such as the new ref method, hoveing footnotes, etc.
Other then clicking edit on another page to see how it is done, there is no gjuidance whatsoever.
Rui
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
"being able to figure out wikitext" might be a good attribute, but making it a requirement pretty much sacrifices any hope we have of getting rid of our systemic bias....
Individual editors' skill with wikitext should be independent of almost all of the systemic biases from which we suffer except perhaps math articles with sections ordered by utility to the typical reader. The effect, if it exists, should be stronger on the Simple English Wikipedia.
I'm not opposed to further improvements and measurements of the Visual Editor, but I am opposed to the absence of honest cost-benefit evaluations of supplemental opportunities for increasing participation.
On 05/31/2014 08:27 PM, James Salsman wrote:
Individual editors' skill with wikitext should be independent of almost all of the systemic biases from which we suffer [...]
Seriously?
I have (non-CS) engineer friends that, upon hitting that edit button, basically went "Gak! No way!"
Wikitext resitricts editing to pretty much only "computer science professionals, highly computer-literate professionals (which excludes most of Academia -- have you ever done IT support for a university?), and westerners with enough leisure time to learn it the hard way".
This is, optimistically, 1-2% of the world, only a small fraction of which are women.
There's no way to *not* have a catastrophic systemic bias with those demographics that pretty much excludes the vast majority of academia, most cultures, and selects strongly against women.
-- Marc
On 1 June 2014 03:40, Marc A. Pelletier marc@uberbox.org wrote:
On 05/31/2014 08:27 PM, James Salsman wrote:
Individual editors' skill with wikitext should be independent of almost all of the systemic biases from which we suffer [...]
Seriously? I have (non-CS) engineer friends that, upon hitting that edit button, basically went "Gak! No way!"
I'm a Unix sysadmin and I frequently hit it and go "Gak! No way!" Wikitext is not humanly usable.
The question is not whether we need a better interface, it's the implementation.
- d.
wikimedia-l@lists.wikimedia.org