On 8/27/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
Seriously, though, I'm still looking for a response on my request for details when you said "analysis has been done, results will be published" regarding the initial disabling of anon page creation. Who did the analysis? Or did you answer in a different thread that I haven't been reading? I know I'm being a hardass about this, but so far this has been the only lead I'm aware of indicating that any such analysis might actually be taking place.
I'd lost that thread, which is why I hadn't replied.
It's not really my place to talk about papers being written by others, which is why I hadn't answered in more detail.
You're not being a hardass... I'm just not the right person to ask, the only research I've done is informal subjective measurements (which I posted about months ago), I've also posted the link to the new articles over time which was a part of that informal research.
[[NBER]] presented at Wikimania about their statistical research into Wikipedia editing, and I have worked with them some to help with collecting data for their work. One of the researchers working on the subject is currently off the net, but I'll ping the other person (Eric Garrison) and find out what the status of his research is...
I found it amusing that everyone in the thread has assumed that I'm arguing against turning back on anon page creation, because that is certantly not my position. I've only argued that we shouldn't act without evidence when evidence is so easy to obtain.
If you don't believe me that people are working on this, feel free to do your own research it's not like you need anyone's permission. If you need help getting data in a convenient form and can clearly articulate what data you need, please contact me... I'll be glad to help.
Gregory Maxwell wrote:
It's not really my place to talk about papers being written by others, which is why I hadn't answered in more detail.
I just wanted to know who was doing it so that I could follow up on it and simply confirm that the work was being done. Thanks for giving a name, if you're pinging him yourself already I'll wait a little before pestering him with questions myself.
If you don't believe me that people are working on this, feel free to do your own research it's not like you need anyone's permission. If you need help getting data in a convenient form and can clearly articulate what data you need, please contact me... I'll be glad to help.
I don't have the tools or skillset to be doing this sort of research myself, or frankly the time or interest - I'm here on Wikipedia as an editor, all I want to do is edit. I am, however, concerned about the process behind these sorts of changes. I want to be sure that Wikipedia is being run in an open, inclusive manner since that's one of the major reasons I like this place and continue to stick around.
I try to assume good faith, but the long lack of information about this started to make me strongly suspicious that no experiment was intended, Jimbo wasn't interested in the actual impact on editing and just wanted to make an appearance of "doing something" so press releases could be issued to counter the bad PR of the Siegenthaler matter. While countering bad PR is certainly a good and worthy goal, I would rather not have random tempests-in-teapots spawn restrictions on Wikipedia editing with no plan or options for ever repealing them if they turn out to be counterproductive.
Simply having the means to verify that this "experiment" _is_ in fact an experiment, and not just something that had been called an experiment to brush off any complaints that may have been raised over its implementation, goes a long way toward easing my concerns.
On 8/28/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
I try to assume good faith, but the long lack of information about this started to make me strongly suspicious that no experiment was intended, Jimbo wasn't interested in the actual impact on editing and just wanted to make an appearance of "doing something" so press releases could be issued to counter the bad PR of the Siegenthaler matter. While countering bad PR is certainly a good and worthy goal, I would rather not have random tempests-in-teapots spawn restrictions on Wikipedia editing with no plan or options for ever repealing them if they turn out to be counterproductive.
Well, Jimbo could be evil. Alternatively, he thought "here's an idea for how to fix the problems we've been having, let's see how it goes". 6 months or whatever later, it seems to have been going fairly well, and the only outcry is over the lack of formal experimentation, rather than the result itself.
Simply having the means to verify that this "experiment" _is_ in fact an experiment, and not just something that had been called an experiment to brush off any complaints that may have been raised over its implementation, goes a long way toward easing my concerns.
If you mean he always intended for this to be a permanent change with no chance of being repealed, and only called it an "experiment" to make it easier to swallow, I think you're being unnecessarily harsh and cynical.
Steve
On 8/28/06, Steve Bennett stevagewp@gmail.com wrote: ....
Simply having the means to verify that this "experiment" _is_ in fact an experiment, and not just something that had been called an experiment to brush off any complaints that may have been raised over its implementation, goes a long way toward easing my concerns.
If you mean he always intended for this to be a permanent change with no chance of being repealed, and only called it an "experiment" to make it easier to swallow, I think you're being unnecessarily harsh and cynical.
Steve
Judge by what people do, and not what they say. Perhaps that was not his intention in the beginning, but thus far that has been the practical effect.
~maru
On 8/28/06, maru dubshinki marudubshinki@gmail.com wrote:
Judge by what people do, and not what they say. Perhaps that was not his intention in the beginning, but thus far that has been the practical effect.
Ironically today he has just posted that he's strongly considering undoing it, due to lack of perceived benefits.
Steve
Steve Bennett wrote:
On 8/28/06, maru dubshinki marudubshinki@gmail.com wrote:
Judge by what people do, and not what they say. Perhaps that was not his intention in the beginning, but thus far that has been the practical effect.
Ironically today he has just posted that he's strongly considering undoing it, due to lack of perceived benefits.
Yeah, but until he posted that I also shared Maru's feeling - I had no way of knowing his motives at the time but even if they were perfectly sincere the _result_ was still a haphazard change to policy with vague promise that it would eventually be evaluated but no actual indication anything more would ever be done.
Trust, but verify.
On 8/28/06, Steve Bennett stevagewp@gmail.com wrote:
On 8/28/06, Bryan Derksen bryan.derksen@shaw.ca wrote:
I try to assume good faith, but the long lack of information about this started to make me strongly suspicious that no experiment was intended, Jimbo wasn't interested in the actual impact on editing and just wanted to make an appearance of "doing something" so press releases could be issued to counter the bad PR of the Siegenthaler matter. While countering bad PR is certainly a good and worthy goal, I would rather not have random tempests-in-teapots spawn restrictions on Wikipedia editing with no plan or options for ever repealing them if they turn out to be counterproductive.
Well, Jimbo could be evil. Alternatively, he thought "here's an idea for how to fix the problems we've been having, let's see how it goes". 6 months or whatever later, it seems to have been going fairly well, and the only outcry is over the lack of formal experimentation, rather than the result itself.
Back in June Jimbo posted this:
On 6/20/06, Jimmy Wales jwales@wikia.com wrote:
Guy Chapman aka JzG wrote:
On Tue, 20 Jun 2006 17:37:21 -0400, "The Cunctator" cunctator@gmail.com wrote:
Sorry. This is mainly a reference to the whole "Today, as an experiment, we will be turning off new pages creation for anonymous users in the English Wikipedia." I still think you were being a bit disingenuous (if unintentially) about the experimentality of that decision.
It worked, though.
It is not clear to me that it did. I would love for us to have some serious analysis of that.
I don't think anyone is claiming that Jimbo is evil. But to call this an "experiment" is terribly inaccurate considering that months later the main proponent doesn't have any clue whether or not it worked and is asking for others to do the analysis.
Assume good faith. In other words, assume incompetence. In any case, I'd say it's time to correct the mistake. Gregory says "that we shouldn't act without evidence when evidence is so easy to obtain". To that I'd respond that the "experiment" was conducted so poorly that good evidence is impossible to obtain. There are far too many other changes that have been made between the start of the "experiment" and today.
Anthony
Sometimes when people say "experiment", they mean "let's measure the reliability of a hypothesis using months worth of carefully worked out calculations and data measurements". Sometimes when people say "experiment", they mean "let's try this new thing and see how well it works out". It astonishes me that people are attributing a nakedly cynical motive to a different connotation of a commonly used word.
On 8/28/06, Rob gamaliel8@gmail.com wrote:
Sometimes when people say "experiment", they mean "let's measure the reliability of a hypothesis using months worth of carefully worked out calculations and data measurements". Sometimes when people say "experiment", they mean "let's try this new thing and see how well it works out". It astonishes me that people are attributing a nakedly cynical motive to a different connotation of a commonly used word.
And no one has proposed how we could have measured the effects without going ahead and doing it... there are just too many variables. We could have carefully quantified how much crap is created by anons vs non-anons, but we couldn't have quantified how many of the crapmakers would have simply made accounts.
Wikipedia won't be finished in a day...and some times we need to just try things out if we are to find good solutions.
On 8/28/06, Gregory Maxwell gmaxwell@gmail.com wrote:
And no one has proposed how we could have measured the effects without going ahead and doing it... there are just too many variables. We
I don't know what you mean by "without going ahead and doing it" - obviously you have to perform the experiment in order to collect data from it. I also proposed a list of data that could have been collected. I don't have any problem with the fact that no data were collected, however - there are scientific experiments, and there are lots of other experiments.
could have carefully quantified how much crap is created by anons vs non-anons, but we couldn't have quantified how many of the crapmakers would have simply made accounts.
Well, we *could* have.
Wikipedia won't be finished in a day...and some times we need to just try things out if we are to find good solutions.
Yep. My heart is not bleeding that anons couldn't make their usual truckload of crappy articles.
Steve
On 8/28/06, Steve Bennett stevagewp@gmail.com wrote:
On 8/28/06, Gregory Maxwell gmaxwell@gmail.com wrote:
Wikipedia won't be finished in a day...and some times we need to just try things out if we are to find good solutions.
Yep. My heart is not bleeding that anons couldn't make their usual truckload of crappy articles.
So are a truckload of crappy articles no longer being made, or do you think you can identify the people who are making them?
Anthony
At 20:14 -0400 28/8/06, Anthony wrote:
On 8/28/06, Steve Bennett stevagewp@gmail.com wrote:
On 8/28/06, Gregory Maxwell gmaxwell@gmail.com wrote:
Wikipedia won't be finished in a day...and some times we need to just try things out if we are to find good solutions.
Yep. My heart is not bleeding that anons couldn't make their usual truckload of crappy articles.
So are a truckload of crappy articles no longer being made, or do you think you can identify the people who are making them?
Anthony
I have found two articles that needed attention within the past few days. I probably came to Wikipedia via a Google.com search. In other words, I did not start in Wikipedia.
// pauses to reflect
In one case, an obituary from "The Times" had been posted verbatim when the person in question had died in March 2006. By August 2006, the full copyright text was still there, so I removed it and created a biographical entry. I then bought the book that this person had written, and started to add references (the book and its ISBN etc) and other information, and linked to the Times article as an external link.
In the second case, the person is still living I believe. He was born in 1941, and is household name in the East End of London (UK). I added a couple of redirects and will probably "wikify" the article next time around. It has these: {{tone}}, {{references}} and {{wikify-date|August 2006}}...
I have often stopped from creating an article for a living person, and I note that there is now clear policy on this, which I welcome.
http://en.wikipedia.org/wiki/Wikipedia:Biographies_of_living_persons
http://en.wikipedia.org/wiki/Wikipedia:WikiProject_Biography
When I started being editing, and I am a late comer (January 2004), there was no clear policy (or I was unaware of it).
Gordo
On 8/28/06, Steve Bennett stevagewp@gmail.com wrote:
On 8/28/06, Gregory Maxwell gmaxwell@gmail.com wrote:
And no one has proposed how we could have measured the effects without going ahead and doing it... there are just too many variables. We
I don't know what you mean by "without going ahead and doing it" - obviously you have to perform the experiment in order to collect data from it. I also proposed a list of data that could have been collected. I don't have any problem with the fact that no data were collected, however - there are scientific experiments, and there are lots of other experiments.
"No data was collected" ... What single piece of data did we *fail* to collect? Okay.. there are a few things, but I've yet to see any suggested here.
Collection isn't the problem. It's the effort it take to sit down and make sense of the fire hose...