On 6/16/07, Guy Chapman aka JzG guy.chapman@spamcop.net wrote:
On Fri, 15 Jun 2007 16:29:00 -0700, "Joe Szilagyi" szilagyi@gmail.com wrote:
Is it appropriate for a CheckUser to disclose on someone's RFA the methods of *how* they connect to edit Wikipedia?
Yes, if they are using TOR. TOR is verboten, for good reason.
Guy (JzG)
http://www.chapmancentral.co.uk http://en.wikipedia.org/wiki/User:JzG
jayjg jayjg99@gmail.com to English show details 8:52 am (3 hours ago)
Well, as explained before, I've already answered one of the questions,
and you're neither a prosecutor nor a judge. There's no particular reason I should answer questions from an obviously hostile questioner who has been applying outrageous double standards in this incident from the very start.
If the TOR was forbidden for good reasons, Charlotte should have simply been asked, long before the RfA, to stop using it. I have no doubt she would have complied with a request. There was no need for a revelation of information gained through use of check user tools to sink her RfA.
That is the real Occam's razor coupled with AGF in this incident: ask her to stop, when you first find out, explaining it is verboten, for good reason.
KP
On 6/17/07, K P kpbotany@gmail.com wrote:
On 6/16/07, Guy Chapman aka JzG guy.chapman@spamcop.net wrote:
On Fri, 15 Jun 2007 16:29:00 -0700, "Joe Szilagyi" szilagyi@gmail.com wrote:
Is it appropriate for a CheckUser to disclose on someone's RFA the
methods
of *how* they connect to edit Wikipedia?
Yes, if they are using TOR. TOR is verboten, for good reason.
Guy (JzG)
http://www.chapmancentral.co.uk http://en.wikipedia.org/wiki/User:JzG
jayjg jayjg99@gmail.com to English show details 8:52 am (3 hours ago)
Well, as explained before, I've already answered one of the questions,
and you're neither a prosecutor nor a judge. There's no particular reason I should answer questions from an obviously hostile questioner who has been applying outrageous double standards in this incident from the very start.
If the TOR was forbidden for good reasons, Charlotte should have simply been asked, long before the RfA, to stop using it. I have no doubt she would have complied with a request. There was no need for a revelation of information gained through use of check user tools to sink her RfA.
That is the real Occam's razor coupled with AGF in this incident: ask her to stop, when you first find out, explaining it is verboten, for good reason.
KP
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: http://lists.wikimedia.org/mailman/listinfo/wikien-l
Asking someone first is indeed a good first step, but I fail to see what the "good reason" for forbidding it would be. As long as the user isn't hiding behind the proxy to get away with edit warring and such there's no reason we should disallow it. (Is there evidence of any editing abuse?)
If anonymity services are really that much of a problem we should've blocked AOL ages ago (we know it is possible). It is offering more or less the same functionality.
On 6/17/07, MacGyverMagic/Mgm macgyvermagic@gmail.com wrote:
If anonymity services are really that much of a problem we should've blocked AOL ages ago (we know it is possible). It is offering more or less the same functionality.
Since AOL proxies now pass XFF headers, they are no longer useful for anonymising; thus the problem there has gone.
The issue with AOL, before XFF was available, was that many Internet users were set to use those proxies by default. This is not the case with TOR, the use of which is optional.
-Matt
Asking someone first is indeed a good first step, but I fail to see what the "good reason" for forbidding it would be. As long as the user isn't hiding behind the proxy to get away with edit warring and such there's no reason we should disallow it. (Is there evidence of any editing abuse?)
But that's the whole point - there won't be any evidence of abuse even if she's the worst edit warrer ever since all the evidence would be hidden by the fact that she's using TOR. We have no way to know if she's hiding behind it or using it legitimately, and that's why it is banned.
On 6/17/07, Thomas Dalton thomas.dalton@gmail.com wrote:
Asking someone first is indeed a good first step, but I fail to see what the "good reason" for forbidding it would be. As long as the user isn't hiding behind the proxy to get away with edit warring and such there's no reason we should disallow it. (Is there evidence of any editing abuse?)
But that's the whole point - there won't be any evidence of abuse even if she's the worst edit warrer ever since all the evidence would be hidden by the fact that she's using TOR. We have no way to know if she's hiding behind it or using it legitimately, and that's why it is banned.
In addition, I believe this issue was debated before the policy made.
KP
Thomas Dalton wrote:
But that's the whole point - there won't be any evidence of abuse even if she's the worst edit warrer ever
Well, besides all the prima facie evidence of being the worst edit warrior ever.
since all the evidence would be hidden by the fact that she's using TOR.
If someone is being abusive, is suspected of being a sockpuppet, and is found to be using TOR, that's another pretty good nail in their coffin.
We have no way to know if she's hiding behind it or using it legitimately, and that's why it is banned.
Not quite. It's banned (as I understand it) because if editor A is editing from 12.345.678.910, and editor B is editing via TOR, and B is suspected of being A's sockpuppet, we can't say, "Aha! You're using the same IP address as A!" But it seems to me we could note that they're using TOR and block them as a sockpuppet on that basis -- *after* they're suspected/accused of being a sockpuppet.
The claim that productive contributors can't/shouldn't/mustn't use proxies is tenuous at best.
On 6/17/07, Steve Summit scs@eskimo.com wrote:
Thomas Dalton wrote:
But that's the whole point - there won't be any evidence of abuse even if she's the worst edit warrer ever
Well, besides all the prima facie evidence of being the worst edit warrior ever.
since all the evidence would be hidden by the fact that she's using TOR.
If someone is being abusive, is suspected of being a sockpuppet, and is found to be using TOR, that's another pretty good nail in their coffin.
We have no way to know if she's hiding behind it or using it legitimately, and that's why it is banned.
Not quite. It's banned (as I understand it) because if editor A is editing from 12.345.678.910, and editor B is editing via TOR, and B is suspected of being A's sockpuppet, we can't say, "Aha! You're using the same IP address as A!" But it seems to me we could note that they're using TOR and block them as a sockpuppet on that basis -- *after* they're suspected/accused of being a sockpuppet.
The claim that productive contributors can't/shouldn't/mustn't use proxies is tenuous at best.
Well, besides all the prima facie evidence of being the worst edit warrior ever.
Oh, thanks for stating the obvious.
KP
On 6/17/07, Steve Summit scs@eskimo.com wrote:
We have no way to know if she's hiding behind it or using it legitimately, and that's why it is banned.
Not quite. It's banned (as I understand it) because if editor A is editing from 12.345.678.910, and editor B is editing via TOR, and B is suspected of being A's sockpuppet, we can't say, "Aha! You're using the same IP address as A!" But it seems to me we could note that they're using TOR and block them as a sockpuppet on that basis -- *after* they're suspected/accused of being a sockpuppet.
The claim that productive contributors can't/shouldn't/mustn't use proxies is tenuous at best.
It's for the exact same reason, just in reverse: If an apparently good-faith contributor is using an anonymizing service like TOR, it's impossible to detect that they're also trolling (or whatever) from their actual IP. It also becomes entirely impossible to identify more subtle forms of disruption such as sockpuppetry.
On 6/18/07, zetawoof zetawoof@gmail.com wrote:
On 6/17/07, Steve Summit scs@eskimo.com wrote:
We have no way to know if she's hiding behind it or using it legitimately, and that's why it is banned.
Not quite. It's banned (as I understand it) because if editor A is editing from 12.345.678.910, and editor B is editing via TOR, and B is suspected of being A's sockpuppet, we can't say, "Aha! You're using the same IP address as A!" But it seems to me we could note that they're using TOR and block them as a sockpuppet on that basis -- *after* they're suspected/accused of being a sockpuppet.
The claim that productive contributors can't/shouldn't/mustn't use proxies is tenuous at best.
It's for the exact same reason, just in reverse: If an apparently good-faith contributor is using an anonymizing service like TOR, it's impossible to detect that they're also trolling (or whatever) from their actual IP. It also becomes entirely impossible to identify more subtle forms of disruption such as sockpuppetry.
I think it could be both; one could have different sockpuppets using different TOR exit nodes, alongside a real IP account.
zetawoof wrote:
On 6/17/07, Steve Summit scs@eskimo.com wrote:
Not quite. It's banned (as I understand it) because if editor A is editing from 12.345.678.910, and editor B is editing via TOR, and B is suspected of being A's sockpuppet, we can't say, "Aha! You're using the same IP address as A!" But it seems to me we could note that they're using TOR and block them as a sockpuppet on that basis -- *after* they're suspected/accused of being a sockpuppet.
It's for the exact same reason, just in reverse: If an apparently good-faith contributor is using an anonymizing service like TOR, it's impossible to detect that they're also trolling (or whatever) from their actual IP.
But we don't use checkuser to detect that apparently good-faith contributors are trolling or sockpuppeting or whatever. We use it in an attempt to confirm that apparently bad-faith editors are.
G'day Steve Summit,
zetawoof wrote:
It's for the exact same reason, just in reverse: If an apparently good-faith contributor is using an anonymizing service like TOR, it's impossible to detect that they're also trolling (or whatever) from their actual IP.
But we don't use checkuser to detect that apparently good-faith contributors are trolling or sockpuppeting or whatever. We use it in an attempt to confirm that apparently bad-faith editors are.
Hear, hear. "I hold in my hand the names of 205 administrators who are also trolling as IP addresses!"
If I'm a troll/puppetmaster with different accounts from this IP, you have no way of knowing ... because you have no reason to check. Zetawoof seems to be proposing that we check all users to ensure they aren't running multiple accounts[0]. Zetawoof's Wikipedia is not a Wikipedia to which I would wish to contribute.
[0] Which is not, in itself, a violation of policy or the Right Thing.
As I understand it, any admin logging in from a possibly non secure terminal is supposed to use either the ip address or an alternative user name. DGG
On 7/4/07, Mark Gallagher m.g.gallagher@student.canberra.edu.au wrote:
G'day Steve Summit,
zetawoof wrote:
It's for the exact same reason, just in reverse: If an apparently good-faith contributor is using an anonymizing service like TOR, it's impossible to detect that they're also trolling (or whatever) from their actual IP.
But we don't use checkuser to detect that apparently good-faith contributors are trolling or sockpuppeting or whatever. We use it in an attempt to confirm that apparently bad-faith editors are.
Hear, hear. "I hold in my hand the names of 205 administrators who are also trolling as IP addresses!"
If I'm a troll/puppetmaster with different accounts from this IP, you have no way of knowing ... because you have no reason to check. Zetawoof seems to be proposing that we check all users to ensure they aren't running multiple accounts[0]. Zetawoof's Wikipedia is not a Wikipedia to which I would wish to contribute.
[0] Which is not, in itself, a violation of policy or the Right Thing.
-- Mark Gallagher "'Yes, sir,' said Jeeves in a low, cold voice, as if he had been bitten in the leg by a personal friend."
- P G Wodehouse, /Carry On, Jeeves/
WikiEN-l mailing list WikiEN-l@lists.wikimedia.org To unsubscribe from this mailing list, visit: http://lists.wikimedia.org/mailman/listinfo/wikien-l
On 7/4/07, Mark Gallagher m.g.gallagher@student.canberra.edu.au wrote:
zetawoof wrote:
It's for the exact same reason, just in reverse: If an apparently good-faith contributor is using an anonymizing service like TOR, it's impossible to detect that they're also trolling (or whatever) from their actual IP.
But we don't use checkuser to detect that apparently good-faith contributors are trolling or sockpuppeting or whatever. We use it in an attempt to confirm that apparently bad-faith editors are.
Hear, hear. "I hold in my hand the names of 205 administrators who are also trolling as IP addresses!"
If I'm a troll/puppetmaster with different accounts from this IP, you have no way of knowing ... because you have no reason to check. Zetawoof seems to be proposing that we check all users to ensure they aren't running multiple accounts[0]. Zetawoof's Wikipedia is not a Wikipedia to which I would wish to contribute.
I'm not proposing any such thing. There is, in fact, already an "implicit checkuser" in place for such situations, in the form of the autoblock: blocking an account will also place a temporary block on the IP address they last edited from, which serves as adequate protection from good hand / bad hand sockpuppetry. What I was pointing out was that allowing TOR makes this existing mechanism entirely ineffective.
On 05/07/07, zetawoof zetawoof@gmail.com wrote:
On 7/4/07, Mark Gallagher m.g.gallagher@student.canberra.edu.au wrote:
zetawoof wrote:
It's for the exact same reason, just in reverse: If an apparently good-faith contributor is using an anonymizing service like TOR, it's impossible to detect that they're also trolling (or whatever) from their actual IP.
But we don't use checkuser to detect that apparently good-faith contributors are trolling or sockpuppeting or whatever. We use it in an attempt to confirm that apparently bad-faith editors are.
Hear, hear. "I hold in my hand the names of 205 administrators who are also trolling as IP addresses!"
If I'm a troll/puppetmaster with different accounts from this IP, you have no way of knowing ... because you have no reason to check. Zetawoof seems to be proposing that we check all users to ensure they aren't running multiple accounts[0]. Zetawoof's Wikipedia is not a Wikipedia to which I would wish to contribute.
I'm not proposing any such thing. There is, in fact, already an "implicit checkuser" in place for such situations, in the form of the autoblock: blocking an account will also place a temporary block on the IP address they last edited from, which serves as adequate protection from good hand / bad hand sockpuppetry. What I was pointing out was that allowing TOR makes this existing mechanism entirely ineffective.
I know it isn't entirely inline with guidelines but what about editors who do edit controversial articles with a second log-in? They are good editors who just don't want controversial articles showing on their contrib log. mike
On 7/4/07, michael west michawest@gmail.com wrote:
I know it isn't entirely inline with guidelines but what about editors who do edit controversial articles with a second log-in? They are good editors who just don't want controversial articles showing on their contrib log.
They can go on doing that. However, they also have no reason to use anonymizing services (such as TOR) in doing so.
Well, besides all the prima facie evidence of being the worst edit warrior ever.
If they are edit warring using sockpuppets it may not be possible to spot that they are doing so. If someone became suspicious that something odd was going on with a particular article and ran a checkuser on all the contributors the best they could find is that lots of people are using TOR, but if we allow people to use TOR then it is quite a leap to assume that it is just one person using TOR. The same can be said about people editing from the same city, or same ISP, of course, but I would imagine the abuse rate on TOR is higher than for any ISP or city by virtue of the fact that any vandal can choose to use TOR, you don't have the same choice with where you live, so the situation is going to arise more often with TOR than it does otherwise.
Thomas Dalton wrote:
If someone became suspicious that something odd was going on with a particular article and ran a checkuser on all the contributors
Is that routinely done? I was under the impression that checkuser was applied only to specific users about whom suspicions had arisen.
the best they could find is that lots of people are using TOR, but if we allow people to use TOR then it is quite a leap to assume that it is just one person using TOR.
That's why I've been musing about the possibility of changing the current
You may not edit using TOR or open proxies
(if that's indeed what the current policy is supposed to say) to instead read
If you edit using TOR or open proxies, you run the risk of being falsely accused as a sockpuppet.
On 19/06/07, Steve Summit scs@eskimo.com wrote:
Thomas Dalton wrote:
If someone became suspicious that something odd was going on with a particular article and ran a checkuser on all the contributors
Is that routinely done? I was under the impression that checkuser was applied only to specific users about whom suspicions had arisen.
Pretty much. My use of it tends to be assisting other admins in tracking long-term trolls. (One of whom is an utterly reliable open proxy canary. Now if only he would get bored and go away ...)
That said, checking Tor IPs is a pretty reliable way of spotting trouble quickly. Because the good editors (Charlotte Webb, Armed Blowfish) going through them really are the exceptions, sadly.
"We are blocking Tor as a firehose of sewage." "WHY DO YOU HATE FREEDOM!" "Guess we must just be like that."
the best they could find is that lots of people are using TOR, but if we allow people to use TOR then it is quite a leap to assume that it is just one person using TOR.
That's why I've been musing about the possibility of changing the current If you edit using TOR or open proxies, you run the risk of being falsely accused as a sockpuppet.
"bad-faith editor", I suggest.
- d.
On 17/06/07, K P kpbotany@gmail.com wrote:
If the TOR was forbidden for good reasons, Charlotte should have simply been asked, long before the RfA, to stop using it.
I'm glad someone else understands.
On 6/18/07, James Farrar james.farrar@gmail.com wrote:
On 17/06/07, K P kpbotany@gmail.com wrote:
If the TOR was forbidden for good reasons, Charlotte should have simply been asked, long before the RfA, to stop using it.
I'm glad someone else understands.
Yes, everyone using them should be asked to stop.
KP