On the moral bankruptcy of Wikipedia’s anonymous admins

[DanMurphy]

Start with John Smith of the USA. Create a culture where it's seen as acceptable to ask people questions about their background and location, and a little dodgy to refuse to address things. An imperfect system is not an argument against making no improvements at all. Y'all are addicted to your secrecy. For most wikipedians, it makes them feel important, like they're on a secret mission. It usually just enables bad and incompetent behavior.

Think you pretty much summed up why I would leave Wikipedia immediately upon such a change.

Yes, if having to put your name behind your editing would drive people like you from Wikipedia, I would consider that a feature, not a bug. Who are you?

[HRIP7]

Bottom line, anonymity is not that big a part of the problem, and removing probably makes things worse. One thing I could get behind is an attempt to ensure no person can ever get adminship via a sock account - require ID verification to the Foundation, maybe even a token $0.01 credit card deposit (the current system of emailed ID paper scans isn't exactly foolproof). Knowing that you can't just walk away from a bad admin rep to start again would be a much better defence against wantonly bad behaviour than public identification, I think.

The idea of the $0.01 credit card payment has always appealed to me as well.

[Kelly Martin]

Bottom line, anonymity is not that big a part of the problem, and removing probably makes things worse. One thing I could get behind is an attempt to ensure no person can ever get adminship via a sock account - require ID verification to the Foundation, maybe even a token $0.01 credit card deposit (the current system of emailed ID paper scans isn't exactly foolproof). Knowing that you can't just walk away from a bad admin rep to start again would be a much better defence against wantonly bad behaviour than public identification, I think.

The idea of the $0.01 credit card payment has always appealed to me as well.

Honestly, I think full editorship in Wikipedia should be afforded only to members in a dues-paying organization. $10 to $20 a year seems reasonable. Editors who appear able to contribute but cannot afford to pay dues can petition for a waiver of dues. Full membership would be required, at a minimum, to be an administrator or any other role of responsibility beyond casual contributor.

[The Devil's Advocate]

Yes, if having to put your name behind your editing would drive people like you from Wikipedia, I would consider that a feature, not a bug. Who are you?

I get that you really have no reason to be concerned as you have a high-profile job, so ending anonymity would have no discernible impact on your life, but not all of us are so well-known and would prefer that people not have access to our personal information simply because we don't have as much ability to defend ourselves should that information be misused. Having my name associated with my editing is not inherently the issue I am concerned about. Now, would you please provide a substantive rebuttal to my previous argument, rather than trying to divert attention away from the issue?

[lilburne]

Real life names or pseudonyms it doesn't matter so long as the RL name can be accessed when required. The issue is whether people think they are anonymous or not. So if CantTieMeShoeLacesUp has put defamation on WP the defamed need to go to WP to discover the miscreant (with suitable legal protections). IOW if CantTieMeShoeLacesUp wrote that Jane Doe stole money from school and was caught giving blow jobs behind the bus shelter, then one should only need to go to WP to find out who CantTieMeShoeLacesUp is. They shouldn't have to go to the courts to get information from WP, and then go to court to extract information from an ISP based on the information from WP, only to then find that the entry was made from some McDonald's WiFi point.

Access to the RL information can be restricted to back office staff. It doesn't need to be publicly accessible.

Now that's not going to give you all information about whether CantTieMeShoeLacesUp is some 13 yo kid, a 50 yo professor, or some up and coming journalist. But the knowledge that your actions aren't that anonymous does make one think about what one is doing. I don't suspect for one moment that Hari would have created accounts to defame his rivals if such a scheme was in place.

[HRIP7]

Real life names or pseudonyms it doesn't matter so long as the RL name can be accessed when required. The issue is whether people think they are anonymous or not. So if CantTieMeShoeLacesUp has put defamation on WP the defamed need to go to WP to discover the miscreant (with suitable legal protections). IOW if CantTieMeShoeLacesUp wrote that Jane Doe stole money from school and was caught giving blow jobs behind the bus shelter, then one should only need to go to WP to find out who CantTieMeShoeLacesUp is. They shouldn't have to go to the courts to get information from WP, and then go to court to extract information from an ISP based on the information from WP, only to then find that the entry was made from some McDonald's WiFi point.

Access to the RL information can be restricted to back office staff. It doesn't need to be publicly accessible.

Now that's not going to give you all information about whether CantTieMeShoeLacesUp is some 13 yo kid, a 50 yo professor, or some up and coming journalist. But the knowledge that your actions aren't that anonymous does make one think about what one is doing. I don't suspect for one moment that Hari would have created accounts to defame his rivals if such a scheme was in place.

That's the key point. Doing what Hari did only makes psychological sense if you think these things will never be traced back to you. Same with people who add unflattering material to biographies of people they know and dislike.

People wishing to edit BLPs (and articles about private companies, too) should identify, to the Foundation at least if not to the general public. Otherwise the whole encyclopedia thing is just a game of silly-buggers or, per the quip on this site's home page, a "modern defamation engine".

Admins should really identify as well – and disclose their age – to the Foundation. But Rd232's comparison to police, who are only required to identify to the public by badge number, is not without merit. Some admins on Wikipedia and other websites who are open about their real life identity have indeed been subject to appalling harassment, to the point of requiring police help. Most fair-minded people would concede that in most cases they did nothing to deserve this.

Wikipedia's admin numbers are dropping already, and there is a very real risk that people might not want to do the job if they have to identify, and would fear leaks even if the information were only held in confidence by the Foundation. Let's face it, the Foundation's track record with regard to leaks is not stellar. The Foundation and the community would not institute a change that they believe would be the equivalent of harakiri.

The thing to do is to hammer out one or two proposals that would have a realistic chance of flying, and then to harp on about the matter ad nauseam.

[HRIP7]

Bottom line, anonymity is not that big a part of the problem, and removing probably makes things worse. One thing I could get behind is an attempt to ensure no person can ever get adminship via a sock account - require ID verification to the Foundation, maybe even a token $0.01 credit card deposit (the current system of emailed ID paper scans isn't exactly foolproof). Knowing that you can't just walk away from a bad admin rep to start again would be a much better defence against wantonly bad behaviour than public identification, I think.

The idea of the $0.01 credit card payment has always appealed to me as well.

Of course, not everyone in the second and third worlds has credit cards. Scanning and faxing or e-mailing ID papers is ridiculously easy to fake, and no realistic alternative.

Could mobile phone contracts be used for automated identification / user registration?

[The Devil's Advocate]

I remember someone saying elsewhere that Wikipedia is built on misguided presumptions of people being generally good. Seems to me like you all are making the same misguided presumption on ending anonymity. Do you really think letting some 50 year-old problematic editor know that the admin who blocked him or her is a 20 year-old college student is going to do anything but create friction? What do you do when the same lack of anonymity you suggest would help identify defamers is used to identify people who are trying to remove defamation from articles on someone's opponents? I can tell you right now, the people who are likely to use the lack of anonymity in an abusive manner are going to be far more active using it than those who would use it for good reasons.

[thekohser]

...a weapon in the hands of people who want to intimidate and harass their opponents...

Why does a free and neutral encyclopedia that publicly logs all edits have "opponents" working on it? Perhaps that's the point you pro-secrecy advocates need to address?

As for Rd232's protestations against his support of pseudonymity for paid editors, it appears that at least two of us interpreted your stance that way, so why don't you enlighten us as to why pseudonymity is good for all editors, except paid editors?

[The Devil's Advocate]

Why does a free and neutral encyclopedia that publicly logs all edits have "opponents" working on it? Perhaps that's the point you pro-secrecy advocates need to address?

Because human beings are human beings and everywhere human beings interact there is confrontation. Unless you plan to fix all the world's ills, you can't honestly demand people address the human tendency towards confrontation on Wikipedia alone. It is no different than suggesting Wikipedia should have some sort of permanent solution to incivility. People get angry and there is no real way to control that anywhere except with exceedingly invasive measures.

Also, I am not a "pro-secrecy advocate", but someone who supports respecting the privacy of others. Someone's name or where someone lives is not relevant to stopping bad editing and encouraging good editing. It happens whether editors are fully open about their identities, or do everything they can to conceal it. Even so, secrecy is not a sin in itself and, as I said at the beginning, there is nothing inherently immoral about anonymity.

[roger_pearse]

Why does a free and neutral encyclopedia that publicly logs all edits have "opponents" working on it? Perhaps that's the point you pro-secrecy advocates need to address?

Because human beings are human beings and everywhere human beings interact there is confrontation. Unless you plan to fix all the world's ills, you can't honestly demand people address the human tendency towards confrontation on Wikipedia alone. It is no different than suggesting Wikipedia should have some sort of permanent solution to incivility. People get angry and there is no real way to control that anywhere except with exceedingly invasive measures.

After which it seems almost a moral duty to troll...

Is there any evil that someone could not "defend" using this kind of argument? I rather doubt it. And it's all special pleading for anonymous admins anyway.

All the best,

Roger Pearse

[ErrantX]

Bottom line, anonymity is not that big a part of the problem, and removing probably makes things worse. One thing I could get behind is an attempt to ensure no person can ever get adminship via a sock account - require ID verification to the Foundation, maybe even a token $0.01 credit card deposit (the current system of emailed ID paper scans isn't exactly foolproof). Knowing that you can't just walk away from a bad admin rep to start again would be a much better defence against wantonly bad behaviour than public identification, I think.

The idea of the $0.01 credit card payment has always appealed to me as well.

Some of us don't use credit cards!

[Peter Damian]

Apply to the WMF of your own country, giving a postal address. They check that no one has registered from there before, and post back a password. Not completely foolproof but would eliminate 99.9% of known socks. Would give gainful employment to anyone incapable even of passing GCSEs.

They would reject this because (a) it would slow down the process of registration, and might deter editors. This, despite the fact that many more editors are deterred by the persistent sockpuppetry and chaos of the current system. And (b) there would be nothing for most of the admins to do. No checkuser, no blocking, no vandal patrolling and gaming.

It would never fly.

[Kelly Martin]

One of my hobbies, amateur radio, has dealt with this issue to some degree. Some hams are big into "contesting", which involves trying to make a radio contact with other hams in as many other countries (or various other divisions of the world or parts of it) as possible. The way you prove a contact is by exchanging a "QSL card" (QSL being a radio code for "confirm") with the other contact, usually by mail, but in recent years the Internet has become more common as well. Once you collect enough of these you can submit them to the contest bureau of whatever entity is running the contest in question, and once they validate them they'll issue an award certificate for whatever level of success you've attained.

Where this relates to Wikipedia's problem is that the contest bureaus have to have some way to validate that the logged contacts are valid. For paper QSLs, the fact that it was mailed from one party to the other is pretty strong proof; the card will have postmarks indicating where it was mailed from. For "electronic QSLs" (eQSLs) the bureaus typically require that the eQSL be cryptographically signed by keys that they themselves issue (I have one from ARRL's contest bureau, although I don't contest much), and they use a variety of strategies to ensure that they only issue a key to the actual amateur so that a eQSL signed by, say, me, was actually made by me. In most cases, this involves mailing a post card with a challenge code to the amateur's postal address, which is then used to log into the eQSL site and generate the signing key.

Peter's strategy mentioned above seems fairly similar. What's interesting to me, though, is the way culture affects the respective processes. In the QSL process, cheaters are excoriated: anyone caught cheating will be banned from any future awards for life, and may even risk losing their license for "conduct unbecoming an amateur radio operator". There is cheating, of course (people are people, after all), but by and large most of the people involve behave honorably and most of the erroneously claimed contacts that the QSL bureaus find are due to honest errors and misunderstandings, rather than deliberate attempts to cheat. The thing is, contesters (most of them, at least) have an interest in having winners be acknowledged as winners because they fairly and truly won, not because they cheated, and that includes ensuring that people can't submit fake QSLs, and so the QSL bureaus support that. What's Wikipedia's incentive to do the same? What incentive does Wikipedia have to care if people edit Wikipedia under assumed identities?

[The Devil's Advocate]

Apply to the WMF of your own country, giving a postal address. They check that no one has registered from there before, and post back a password. Not completely foolproof but would eliminate 99.9% of known socks. Would give gainful employment to anyone incapable even of passing GCSEs.

They would reject this because (a) it would slow down the process of registration, and might deter editors. This, despite the fact that many more editors are deterred by the persistent sockpuppetry and chaos of the current system. And (b) there would be nothing for most of the admins to do. No checkuser, no blocking, no vandal patrolling and gaming.

It would never fly.

You know, it just occurred to me that Wikipedia doesn't even require an e-mail address, which is actually pretty common and almost a formality on most websites as a way of insuring people do not use multiple accounts. I wonder how many socks are created by taking advantage of that rather simple loophole? Granted, someone could still create multiple e-mail accounts to dodge the restriction, but it would be a little more annoying and time-consuming than the current system where you can log out, create a new account, and then go right back to editing as someone else without ever going to another website.

[lilburne]

Since 2000 I have had unlimited email accounts via my ISP. I can have whatever@accountname.hostingsite.com I gather you can do something similar using periods on a gmail account. If you register a domain via Godaddy (<$10 a year), then you get up to 100 xxxx@domain.name accounts. You can set up a catchall forward for everything that isn't a specified XXX@domain.name which in my book means unlimited email address too. Over the last 12 years I've never been short of an alternative email account or two, just by typing it into the email field on the registration form.

[The Devil's Advocate]

Since 2000 I have had unlimited email accounts via my ISP. I can have whatever@accountname.hostingsite.com I gather you can do something similar using periods on a gmail account. If you register a domain via Godaddy (<$10 a year), then you get up to 100 xxxx@domain.name accounts. You can set up a catchall forward for everything that isn't a specified XXX@domain.name which in my book means unlimited email address too. Over the last 12 years I've never been short of an alternative email account or two, just by typing it into the email field on the registration form.

Well, the registering a domain name solely for this use or the g-mail thing would probably be easy to accommodate for, though the ISP one is where I would need a little more detail. If, like the other work-arounds you suggest, the result is still a clearly identifiable link between the accounts that has no logical explanation outside use of multiple accounts by a single user it would still be a plausible thing to filter.

[roger_pearse]

You know, it just occurred to me that Wikipedia doesn't even require an e-mail address, which is actually pretty common and almost a formality on most websites as a way of insuring people do not use multiple accounts.

An interesting thought. But of course the basic thinking is "the more the merrier"; do nothing to exclude anyone from hitting "edit". Everything that modifies this is done reluctantly for powerful reasons. Some web users have no email (few, I know). Others might not like to give theirs (yes, we all use Google these days, but many do not).

So I can see the logic.

All the best,

Roger Pearse

[Kelly Martin]

Since 2000 I have had unlimited email accounts via my ISP. I can have whatever@accountname.hostingsite.com I gather you can do something similar using periods on a gmail account. If you register a domain via Godaddy (<$10 a year), then you get up to 100 xxxx@domain.name accounts. You can set up a catchall forward for everything that isn't a specified XXX@domain.name which in my book means unlimited email address too. Over the last 12 years I've never been short of an alternative email account or two, just by typing it into the email field on the registration form.

Any domain can be directed to Google Apps and hosted there for free (up to ten accounts, but each account can have an unlimited number of aliases, and you can specify a catch-all redirection as well).

[lilburne]

Since 2000 I have had unlimited email accounts via my ISP. I can have whatever@accountname.hostingsite.com I gather you can do something similar using periods on a gmail account. If you register a domain via Godaddy (<$10 a year), then you get up to 100 xxxx@domain.name accounts. You can set up a catchall forward for everything that isn't a specified XXX@domain.name which in my book means unlimited email address too. Over the last 12 years I've never been short of an alternative email account or two, just by typing it into the email field on the registration form.

Any domain can be directed to Google Apps and hosted there for free (up to ten accounts, but each account can have an unlimited number of aliases, and you can specify a catch-all redirection as well).

email account checks are no insurance against multiple account sign ups, for many the hardest part is deciding which name to use today. OK back in 2000 it was rare to get unlimited addies, but some ISPs have done that for a long long while, and it is becoming more common.

[HRIP7]

An interesting thought. But of course the basic thinking is "the more the merrier"; do nothing to exclude anyone from hitting "edit".

It's not just that. If you tell the press you have 30,000 (or whatever number) regular users and then it turns out that on average, each user has two sock accounts, then doing anything to prevent users using their socks would lose you 20,000 users in one fell swoop.

[Kelly Martin]

email account checks are no insurance against multiple account sign ups, for many the hardest part is deciding which name to use today. OK back in 2000 it was rare to get unlimited addies, but some ISPs have done that for a long long while, and it is becoming more common.

I tried to register for some online thing in 2003 or 2004 and their gatekeeper demanded that I list "all of my email addresses". At the time, I was hostmaster (at my employer) for something like 180 domains. I replied with "I have the means to use any email address I like at any of nearly 200 different domains. I'm not going to list all of them for you as you have no need to know that information." Ms. Officious refused to process my registration, but when I complained she was overridden.

The thing is, email address matching is still fairly effective, because most people simply don't know how to register a domain, even now, or are too cheap or lazy to do it solely for the purpose of concealing their identity from an online service.

The thing is, just because a countermeasure is not 100% effective does not mean that you should not do it. That sort of thinking is letting the perfect be the enemy of the good.

[Sweet Revenge]

email account checks are no insurance against multiple account sign ups, for many the hardest part is deciding which name to use today. OK back in 2000 it was rare to get unlimited addies, but some ISPs have done that for a long long while, and it is becoming more common.

I tried to register for some online thing in 2003 or 2004 and their gatekeeper demanded that I list "all of my email addresses". At the time, I was hostmaster (at my employer) for something like 180 domains. I replied with "I have the means to use any email address I like at any of nearly 200 different domains. I'm not going to list all of them for you as you have no need to know that information." Ms. Officious refused to process my registration, but when I complained she was overridden.

The thing is, email address matching is still fairly effective, because most people simply don't know how to register a domain, even now, or are too cheap or lazy to do it solely for the purpose of concealing their identity from an online service.

It's true, most people don't understand the possibilities at all. I have control over about 100 domains, and sometimes put notes in the subdomains, like

Code: Select all

sweet.revenge@why.in.hell.do.i.have.to.give.you.an.email.address.for.this.nonsense.example.com

and it freaks people out no end. I use the names of companies for subdomains to assist in spam filtering and have had customer service reps deny that the email address could work since their employer obviously owned it. Email account checks would definitely help against a lot of the troublemakers.

[HRIP7]

email account checks are no insurance against multiple account sign ups, for many the hardest part is deciding which name to use today. OK back in 2000 it was rare to get unlimited addies, but some ISPs have done that for a long long while, and it is becoming more common.

I tried to register for some online thing in 2003 or 2004 and their gatekeeper demanded that I list "all of my email addresses". At the time, I was hostmaster (at my employer) for something like 180 domains. I replied with "I have the means to use any email address I like at any of nearly 200 different domains. I'm not going to list all of them for you as you have no need to know that information." Ms. Officious refused to process my registration, but when I complained she was overridden.

The thing is, email address matching is still fairly effective, because most people simply don't know how to register a domain, even now, or are too cheap or lazy to do it solely for the purpose of concealing their identity from an online service.

It's true, most people don't understand the possibilities at all. I have control over about 100 domains, and sometimes put notes in the subdomains, like

Code: Select all

sweet.revenge@why.in.hell.do.i.have.to.give.you.an.email.address.for.this.nonsense.example.com

and it freaks people out no end. I use the names of companies for subdomains to assist in spam filtering and have had customer service reps deny that the email address could work since their employer obviously owned it. Email account checks would definitely help against a lot of the troublemakers.

The Wikipedia Review always had this "no throwaway e-mail accounts for registration" policy. One possible approach would be to allow basic editing and account registration by anyone, but make any further flags beyond basic editing dependent on e-mail registration.

[lilburne]

email account checks are no insurance against multiple account sign ups, for many the hardest part is deciding which name to use today. OK back in 2000 it was rare to get unlimited addies, but some ISPs have done that for a long long while, and it is becoming more common.

I tried to register for some online thing in 2003 or 2004 and their gatekeeper demanded that I list "all of my email addresses". At the time, I was hostmaster (at my employer) for something like 180 domains. I replied with "I have the means to use any email address I like at any of nearly 200 different domains. I'm not going to list all of them for you as you have no need to know that information." Ms. Officious refused to process my registration, but when I complained she was overridden.

The thing is, email address matching is still fairly effective, because most people simply don't know how to register a domain, even now, or are too cheap or lazy to do it solely for the purpose of concealing their identity from an online service.

The thing is, just because a countermeasure is not 100% effective does not mean that you should not do it. That sort of thinking is letting the perfect be the enemy of the good.

Most likely true. Social networks rely on people using their primary address to sign up, as a way of tracking user's social networks. I've always recommended that people use a dedicated email addresses for SN signups, and if they haven't to do so as soon as possible.

The first FB account I registered, in the early days of FB I forgot about until someone I knew contacted me by searching using that old email addy from the early 1990s. If I use the address book for the account I was using on USENET, the 100s of automatic collected emails from USENET posts, if handed over to FB will give me a large number of FB accounts, and if stuffed into flickr will give me access to photos of those users at home, with their kids, and grandkids.

Back when people were posting to USENET in the early 1990s they would not have thought that by 2000 one would be able to search up all their ill considered posts, that those posting would have been spread across multiple websites, and that given the information available someone would be able to obtain a photo of their front yard. Anyone that signed up for flickr in the mid 2000s would not have realized that by the end of the decade anyone with their email would be able to browse their photos.

Use throwaway addresses, everywhere. Arseholes track ex-partners through the social networks by using their email addresses.

That said, until people realize what their primary email address reveals about them, email matching will be mostly effective.

[Ngor]

What an eye opener.

[ErrantX]

Apply to the WMF of your own country, giving a postal address. They check that no one has registered from there before, and post back a password. Not completely foolproof but would eliminate 99.9% of known socks. Would give gainful employment to anyone incapable even of passing GCSEs.

They would reject this because (a) it would slow down the process of registration, and might deter editors. This, despite the fact that many more editors are deterred by the persistent sockpuppetry and chaos of the current system. And (b) there would be nothing for most of the admins to do. No checkuser, no blocking, no vandal patrolling and gaming.

It would never fly.

Not a bad idea; but hindered by a) the lack of a "WMF" in most countries and b) usually the storage of things like postal address means more regulatory issues (especially in the EU). I doubt it would pass muster to store postal addresses long term purely for the purpose of editor verification.

[Randy from Boise]

Honestly, I think full editorship in Wikipedia should be afforded only to members in a dues-paying organization. $10 to $20 a year seems reasonable. Editors who appear able to contribute but cannot afford to pay dues can petition for a waiver of dues. Full membership would be required, at a minimum, to be an administrator or any other role of responsibility beyond casual contributor.

That's an interesting concept.

RfB

[Peter Damian]

Apply to the WMF of your own country, giving a postal address. They check that no one has registered from there before, and post back a password. Not completely foolproof but would eliminate 99.9% of known socks. Would give gainful employment to anyone incapable even of passing GCSEs.

They would reject this because (a) it would slow down the process of registration, and might deter editors. This, despite the fact that many more editors are deterred by the persistent sockpuppetry and chaos of the current system. And (b) there would be nothing for most of the admins to do. No checkuser, no blocking, no vandal patrolling and gaming.

It would never fly.

Not a bad idea; but hindered by a) the lack of a "WMF" in most countries and b) usually the storage of things like postal address means more regulatory issues (especially in the EU). I doubt it would pass muster to store postal addresses long term purely for the purpose of editor verification.

No problem. (a) The main WMF would do. (b) UK data protection act does not set out any specific minimum or maximum periods for retaining personal data. It says "Personal data processed for any purpose or purposes shall not be kept for longer than is necessary for that purpose or those purposes." Since the purpose is to ensure that there is no more than one account per named person, that would not be a problem. Clearly the information should not be available except to those involved in that purpose, but that would be a good thing too. Preferably someone actually employed to do so, rather than some 'volunteer' with some ulterior motive.

This would have immeasurable benefits to the project.

1. It would weed out some of those rather odd 'volunteers' for checkuser who are simply interested in spying on editors, and who in one known case identified a user's workplace address and got in touch with employers, plus other unsavoury things. These people have no place in a project to build a comprehensive and reliable reference work.

2. It would give ordinary editors confidence that they are not dealing with socks. Accounts could still be anonymous (that's a separate question) but there would not be all that stupidity of arguing with teams of socks, socks arguing with each other etc. Nor the constant complaining and system-gaming that goes with it.

3. On paid editors – separate question. If the purpose of the check is simply to ensure one person, one account, that is separate from a check to ensure that the postal address used is not a corporate one. Of course, it would reduce the ability of paid editors to spread edits across different accounts. But equally, it would reduce the ability for activists, cranks etc to do that also.

4. It would make editors much more careful about getting banned. Currently, the only incentive for getting unbanned is to edit under your original account name. Otherwise you can simply open a sock. If banning were to mean 'the true death', that would focus a lot of minds. Would it give administrators more power? Yes it would, and it would increase the scope for abuse. Hence a better appeals process should be part and parcel of this. All the current drama connected with checkuser and SPIs could be better directed towards something useful and meaningful, i.e. treating users fairly. One step would be to reduce all blocks to maximum 6 months, i.e. all blocks preventative, none punitive. Someone returning from a 6 month block is likely to be much more careful, knowing that all eyes are upon them.

5. It would make the voting sytem much more logical. Currently there is all manner of abuse, as we know.

[ErrantX]

Apply to the WMF of your own country, giving a postal address. They check that no one has registered from there before, and post back a password. Not completely foolproof but would eliminate 99.9% of known socks. Would give gainful employment to anyone incapable even of passing GCSEs.

They would reject this because (a) it would slow down the process of registration, and might deter editors. This, despite the fact that many more editors are deterred by the persistent sockpuppetry and chaos of the current system. And (b) there would be nothing for most of the admins to do. No checkuser, no blocking, no vandal patrolling and gaming.

It would never fly.

Not a bad idea; but hindered by a) the lack of a "WMF" in most countries and b) usually the storage of things like postal address means more regulatory issues (especially in the EU). I doubt it would pass muster to store postal addresses long term purely for the purpose of editor verification.

No problem. (a) The main WMF would do. (b) UK data protection act does not set out any specific minimum or maximum periods for retaining personal data. It says "Personal data processed for any purpose or purposes shall not be kept for longer than is necessary for that purpose or those purposes." Since the purpose is to ensure that there is no more than one account per named person, that would not be a problem. Clearly the information should not be available except to those involved in that purpose, but that would be a good thing too. Preferably someone actually employed to do so, rather than some 'volunteer' with some ulterior motive.

This would have immeasurable benefits to the project.

1. It would weed out some of those rather odd 'volunteers' for checkuser who are simply interested in spying on editors, and who in one known case identified a user's workplace address and got in touch with employers, plus other unsavoury things. These people have no place in a project to build a comprehensive and reliable reference work.

2. It would give ordinary editors confidence that they are not dealing with socks. Accounts could still be anonymous (that's a separate question) but there would not be all that stupidity of arguing with teams of socks, socks arguing with each other etc. Nor the constant complaining and system-gaming that goes with it.

3. On paid editors – separate question. If the purpose of the check is simply to ensure one person, one account, that is separate from a check to ensure that the postal address used is not a corporate one. Of course, it would reduce the ability of paid editors to spread edits across different accounts. But equally, it would reduce the ability for activists, cranks etc to do that also.

4. It would make editors much more careful about getting banned. Currently, the only incentive for getting unbanned is to edit under your original account name. Otherwise you can simply open a sock. If banning were to mean 'the true death', that would focus a lot of minds. Would it give administrators more power? Yes it would, and it would increase the scope for abuse. Hence a better appeals process should be part and parcel of this. All the current drama connected with checkuser and SPIs could be better directed towards something useful and meaningful, i.e. treating users fairly. One step would be to reduce all blocks to maximum 6 months, i.e. all blocks preventative, none punitive. Someone returning from a 6 month block is likely to be much more careful, knowing that all eyes are upon them.

5. It would make the voting sytem much more logical. Currently there is all manner of abuse, as we know.

But surely it is still predicated on honesty; how would the WMF validate that address is where you live?

[lilburne]

1. It would weed out some of those rather odd 'volunteers' for checkuser who are simply interested in spying on editors, and who in one known case identified a user's workplace address and got in touch with employers, plus other unsavoury things. These people have no place in a project to build a comprehensive and reliable reference work.

Well of course currently there is no reason why say the Chinese or Iranian or whatever government of your choice is, couldn't insert one of there own into a CU position. Its not as if any of the kids would know whether any of them wasn't employed by some government or other.

[lilburne]

But surely it is still predicated on honesty; how would the WMF validate that address is where you live?

It makes it more difficult, for casual abuse. But you can never be sure. Example a friend needed a reference for a landlord when he was renting up in Manchester, after just getting out of jail. Said I was his old landlord for the previous 3 years. Guy phoned me up for the reference and laughing said "sometimes they just write their own references, so I like to phone to keep one step ahead of them". I thought some are a 100 yards ahead of you mate.

You won't stop the planned out deception, you'll deter the ad hoc stuff.

[Peter Damian]

But surely it is still predicated on honesty; how would the WMF validate that address is where you live?

The point is that they mail a password back to you – by postal mail. So you can’t make up any old address except one you have access to. Of course it is possible to subvert this. (Parents do it to get into school catchment areas). But it drastically limits your sock creation to the number of addresses which you physically have access to.

See also Lilburne's point.

A further check would be sending a utility bill for that address which was in your name. This is how our local council make sure that I don’t apply for more than one parking place in different names. This is how ‘old tech’ has worked for years. You will object that this can be subverted. Yes, but with an enormous amount of effort. The ‘old tech’ system works on the principle that you reduce the number of exceptions to manageable amounts.

[DanMurphy]

They are corrupt creeps, and so-called anonymity is frequently a weapon. Look at the behavior of that sniveling hypocrite Chris Owen (aka Chriso/Prioryman), who has rounded up the posse for a good old fashioned lynching today (mentioned also in the "youreallycan thread, but I think this comment belongs here). Why? Because youreallycan etc... wrote about "Cirt" (long term likely paid editor and defamer of Scientologists and Rick Santorum) here:

"No one should be able to hide behind a pseudonym and violate the projects goals and neutrality - Its not outing in any way to expose a project violator, its good for the project to expose such violators. Who is he? If anyone knows, please expose him."

Horrors! A senior writer at Wikipedia might be identified. Mr. Owen doesn't really care about any of this. It is all about destroying the "enemy" (for added fun, it appears that frequent liar is going to the well again, claiming he doesn't have an account here). For the rest? A lot of that too. The response is about half burn the witch so far, half "how do we know if 'youreallycan' is the witch in question?"

Doug Weller, who appears to be a bit of a self-styled Judge Roy Bean kook but is at least self identified writes: "That's completely unacceptable. If no one else does it soon, I think I will indefinitely block him. Dougweller (talk) 11:47, 29 August 2012 (UTC)"

"Soliciting outing is clearly wrong," writes editor "Worm," who is clearly a paid up cult member ("clearly wrong?" Do they know how insane that sounds). etc... etc...

What a pathetic, dysfunctional cesspool. The best that can be said is that many of them think they're doing the right thing. But Owen is as repulsive a four-flushing manipulator as I've come across on the internet, which is saying something.

Anonymity as a default position is evidence of a sick culture. In healthy ones, it should be granted only in rare cases where it's deemed necessary. If you join a church, or a cooking club, or a sports team, or a real world academic project, etc. etc... you are expected to identify yourself.

[ErrantX]

But surely it is still predicated on honesty; how would the WMF validate that address is where you live?

The point is that they mail a password back to you – by postal mail. So you can’t make up any old address except one you have access to. Of course it is possible to subvert this. (Parents do it to get into school catchment areas). But it drastically limits your sock creation to the number of addresses which you physically have access to.

See also Lilburne's point.

A further check would be sending a utility bill for that address which was in your name. This is how our local council make sure that I don’t apply for more than one parking place in different names. This is how ‘old tech’ has worked for years. You will object that this can be subverted. Yes, but with an enormous amount of effort. The ‘old tech’ system works on the principle that you reduce the number of exceptions to manageable amounts.

Given the speed of overseas post this means that signing up might take weeks in many parts of the world. Addresses are trivial to come by (mechanical turk style). The utility bill thing is dead simple to subvert. Especially if you're sending from a country where the WMF is unfamiliar with bill formats etc. (in fact; I don't think I could meet this requirement now, because everything is paperless!).

But the core of my objection is that its an example of security through obscurity - which is not good.

You would get a much reduced regular editing pool of people who are bothered enough to go through the rigmarole. And I'd suggest a *much smaller* reduction in persistent spammers/abusers who want to take advantage of the system.

And even *then* it still hasn't directly addressed the problem at hand.

Accountability is a cultural thing. It needs a cultural change.

[Kelly Martin]

But the core of my objection is that its an example of security through obscurity - which is not good.

That's a trite objection. The whole "security through obscurity is bad" trope that I often hear cited by people who don't understand security irritates me. There is nothing wrong with security through obscurity, and in fact you generally want to obscure your security measures because that alone will be enough to divert or thwart many casual attackers. The issue with security by obscurity is that it must not be your only line of defense, but that's true of any security system.

I see your comments as another form of the "perfect is the enemy of the good". No security system is perfect, so objecting to a security system because it is not perfect is pointless.

Your point about the problem being cultural, though, is spot on. And no "authentication" proposal will change that, so really this discussion is largely pissing into the wind, at least as regards reforming the existing Wikipedia. At best, it serves to inform successors to the Wikipedia project, once it has finally crashed and burned.

[rd232]

The following comments may be increasingly pissing in the wind, given that a few people seem to have settled on some inappropriate comparisons between offline contexts and online ones, and somehow loaded much of Wikipedia's problems onto the fact that its admins are (mostly) not publicly self-identified. (I can't help wondering how much of that is due to personal animosity against admins who rightly or wrongly have sanctioned people contributing here, but anyway...)

But let's try one more time. I've already pointed out that public identification for admins alone doesn't do much to change bad behaviour for the better as long as you can't effectively prevent socking and clean-starting - so we have to look at all editors, not just admins. So let's postulate an attempt to set the whole thing up on the basis that all editors must be reliably and publicly identified, to a reasonable yet practical standard.

1. What would the mechanism be?
2. What would the effects be? How might a decade-old WP look with such a mechanism if it had been born with it?
3. What would the effects be of inventing such a mechanism now or in the future? Which editors and problems would go away, and which wouldn't?

Of course without a good answer to question 1 (which no-one has come up with yet), questions 2 and 3 are moot. Bottom line, unless you have a good mechanism for authentication that can be applied to everyone, then any collaborative project of the scale of Wikipedia must accept pseudonymity for at least a proportion of its contributors. Maybe that's absolutely and always terrible, maybe it's occasionally a problem, maybe it's mostly harmless: whatever your opinion, if you can't explain how to do universal authentication in this global internet context to a reasonable standard, then demanding its complete abolition is pointless, and any attempts to outlaw pseudonymity will merely drive it underground. And of course once it's underground (let's say you manage to make it culturally unacceptable to be pseudonymous), it will be mostly the people you really don't want to be pseudonymous who are (but pretend not to be), whilst those who are mostly behaving well enough are identified.

So here's my conclusion: you can't do mandatory public identification for all accounts - it's not practical. What would be desirable would be
1. mandatory private authentication to a reasonable standard of all accounts with certain higher rights (certainly admins and above, maybe some rights below that too). I'd use a cc/debit card payment of $0.01 - that works for many, and we'd have to manage some backup means that are worth the paper they're written on for those who can't do it. Stressing that this authentication is private and intended to guard against easy clean-starting after bad reputations in positions of responsibility have accrued, and in this way to have some positive effects on behaviour.
2. voluntary public authentication by whatever means seem reliable enough to be worth doing (and maybe if the means vary a lot, the chosen means can be declared publicly somewhere, so people can to some degree evaluate the security of authentication), so that in cases where we have experts contributing, we can give more weight to their opinions about NPOV, sourcing, etc (without permitting original research or mere argument from authority of course). If authenticated Professor X says Y is a crappy source because so-and-so, that ought to carry more weight in a discussion than if PseudonymyBla says it (especially if it's within Professor X's expertise). Wikipedia's institutional claim of the opposite is it's real problem. It's developed all sorts of mechanisms for assessing reliability of sources and whether contributors are constructive enough to contribute - but any claims of external expertise are rejected because they're not authenticated. The answer, of course, is to authenticate them. In many cases that won't be possible - but it many cases, it will.

[Kelly Martin]

I actually care somewhat less if "admins" are identified, because, in theory at least, admins only have the obligation to maintain order, which means who they are, and more importantly, what their knowledge and competencies are outside of Wikipedia, is not that important. The people who need to be identified are the content arbitrators. But Wikipedia doesn't have content arbitrators, and so admins are pressed into that role (protests to the contrary notwithstanding).

If Wikipedia had functioning content dispute resolution by duly empaneled and credentialed experts, the lack of identification of its administrators would be far less of a problem.

[ErrantX]

But the core of my objection is that its an example of security through obscurity - which is not good.

That's a trite objection. The whole "security through obscurity is bad" trope that I often hear cited by people who don't understand security irritates me. There is nothing wrong with security through obscurity, and in fact you generally want to obscure your security measures because that alone will be enough to divert or thwart many casual attackers. The issue with security by obscurity is that it must not be your only line of defense, but that's true of any security system.

Hmm, well I'd tend to disagree there. Not to get too off-topic.

When we work with clients on security review we often talk about security through obscurity - and how it usually the benefit is not worth the cost. We always cite a (real) example of a jewellery broker who took pains not to place advertising/signace on his office (in a block of them) to avoid being targeted. But was the subject of two "drive-by" robberies anyway.

Same thing with Captcha's; they don't deter spammers because you can pay for captchas to be solved by the thousand on Amazon Turk. But they *do* deter a recognisable proportion of normal visitors.

Or the idiotic advice I see, regularly, to make software use non-standard ports to avoid attack. Usually given by people who have never heard of a port scanner

Security via obscurity is rarely worth the cost. The other lines of defence are more effective.

But anyway.

Your point about the problem being cultural, though, is spot on. And no "authentication" proposal will change that, so really this discussion is largely pissing into the wind, at least as regards reforming the existing Wikipedia. At best, it serves to inform successors to the Wikipedia project, once it has finally crashed and burned.

I've been working on ideas to address the cultural problem; perhaps now is the time to air them..

The key thing to address is the position of admins - we mix authority with tool access, and don't properly judge competency in both. Content arbitration needs an overhaul - the focus on requiring admins (or at least them having more weight) to close discussions is one important area to rethink.

To pick up one idea offhand; we should separate tool use from decision making. So AFD closes should be open to a wider group of competent editors - and any deletion closes should be placed in a queue for *a different person with admin tools* to clean up.

This might sound more long winded but it is what we do with speedies anyway - non-admin tags, anyone can reject it, or an admin deletes it.

[roger_pearse]

To pick up one idea offhand; we should separate tool use from decision making. So AFD closes should be open to a wider group of competent editors - and any deletion closes should be placed in a queue for *a different person with admin tools* to clean up.

Excellent idea. Authorities should be different to tool users. Then it would be possible to decide what made a good authority.

All the best,

Roger Pearse

[Vigilant]

http://en.wikipedia.org/wiki/Port_knocking

Is an example of security through obscurity than can be useful.

[Peter Damian]

Given the speed of overseas post this means that signing up might take weeks in many parts of the world. Addresses are trivial to come by (mechanical turk style). The utility bill thing is dead simple to subvert. Especially if you're sending from a country where the WMF is unfamiliar with bill formats etc. (in fact; I don't think I could meet this requirement now, because everything is paperless!).

But the core of my objection is that its an example of security through obscurity - which is not good.

You would get a much reduced regular editing pool of people who are bothered enough to go through the rigmarole. And I'd suggest a *much smaller* reduction in persistent spammers/abusers who want to take advantage of the system.

And even *then* it still hasn't directly addressed the problem at hand.

Accountability is a cultural thing. It needs a cultural change.

Cultural change won't solve the sock puppet problem. On utility bills and the like, pretty much all institutions require this, e.g. the British Museum http://www.bl.uk/reshelp/inrrooms/stp/r ... ntify.html

I don't see how it is that 'easy to subvert'. The main problem is sock accounts which spring up around a single contentious article, or set of articles. Currently you can set up a sock in minutes. How quickly would you set up one if you had to send copies of bills?

On the objection that this would reduce the number of editors, well, good.

[ErrantX]

Cultural change won't solve the sock puppet problem. On utility bills and the like, pretty much all institutions require this, e.g. the British Museum http://www.bl.uk/reshelp/inrrooms/stp/r ... ntify.html

Depends what we are talking about; the original topic is the problem is anonymity and "moral bankruptcy". Which I think is a more pressing problem (in general) than the fracas over a portion of the articles.

I don't see how it is that 'easy to subvert'. The main problem is sock accounts which spring up around a single contentious article, or set of articles. Currently you can set up a sock in minutes. How quickly would you set up one if you had to send copies of bills?

I agree it would certainly slow things down. Which would be no bad thing.

On the objection that this would reduce the number of editors, well, good

I'd suggest the biggest impact would be on the sort of editors you'd not want to see go; the quiet content contributors.

Those treating it like a game will simply meet the requirements.

[Peter Damian]

Depends what we are talking about; the original topic is the problem is anonymity and "moral bankruptcy". Which I think is a more pressing problem (in general) than the fracas over a portion of the articles.

Let's go back to my main thesis, which is that the ease of editing Wikipedia makes it like a large railway terminus or rush hour traffic. Therefore people on Wikipedia behave like people in large crowds – often hostile to each other, or suspicious, or unfriendly. Admins do know one another, more or less, but this makes it worse. They are like a small group of people trying to police a large crowd of strangers. End result is inevitable and predictable. An insular group, hostile to outsiders and often brutal to random strangers who were well-intentioned but rubbed them up the wrong way. A further process of Darwinian selection means that only the thickest-skinned of admins will survive the ordeal. Think about the kind of people that are employed at the doors of busy pubs or clubs. Got it?

Hence there is a trade-off between ease of access to Wikipedia, and its apparent 'moral bankruptcy'.

I'm reading a very good book at the moment by an evolutionary biologist who wants to explain how human society turned from hunter-gatherer about 60,000 years ago, when people only knew a handful of other people, and today, when we interact with strangers drawn from over a billion people. His argument is that we are still fundamentally attuned to relationships in small groups. We deal with large groups by all sorts of workarounds. His example is going to the station and buying a ticket. Would you give something very valuable to a total stranger? Not without qualification. You give it to the person on the other side of the counter because of all sorts of background knowledge. It's a counter, they have a uniform, the railway company has been there for a long time, there are methods of recourse etc etc.

The idea that 'anyone can edit', without any kind of qualification, is complete madness. Note that the much-derided idea of credentials is simply another workaround. You don't know whether you can trust this person's work, but someone else who you do trust has vouched for it. So OK.

All these workarounds – to get over the problem that modern civilisation emerged from the DNA of hunter-gatherers – have themselves evolved over thousands of years. There may be something better, but it's difficult to devise something at short notice that works. Wikipedia, by contrast, threw away all of this background knowledge and proclaimed techno-utopia with the MediaWiki software. Jimmy Wales proclaimed that people are fundamentally good.

Well of course people are fundamentally good, but only when they all know and trust each other. They aren't fundamentally good in large groups without these slowly evolved workarounds.

[Kelly Martin]

I'm reading a very good book at the moment by an evolutionary biologist who wants to explain how human society turned from hunter-gatherer about 60,000 years ago, when people only knew a handful of other people, and today, when we interact with strangers drawn from over a billion people. His argument is that we are still fundamentally attuned to relationships in small groups. We deal with large groups by all sorts of workarounds. His example is going to the station and buying a ticket. Would you give something very valuable to a total stranger? Not without qualification. You give it to the person on the other side of the counter because of all sorts of background knowledge. It's a counter, they have a uniform, the railway company has been there for a long time, there are methods of recourse etc etc.

The idea that 'anyone can edit', without any kind of qualification, is complete madness. Note that the much-derided idea of credentials is simply another workaround. You don't know whether you can trust this person's work, but someone else who you do trust has vouched for it. So OK.

All these workarounds – to get over the problem that modern civilisation emerged from the DNA of hunter-gatherers – have themselves evolved over thousands of years. There may be something better, but it's difficult to devise something at short notice that works. Wikipedia, by contrast, threw away all of this background knowledge and proclaimed techno-utopia with the MediaWiki software. Jimmy Wales proclaimed that people are fundamentally good.

Well of course people are fundamentally good, but only when they all know and trust each other. They aren't fundamentally good in large groups without these slowly evolved workarounds.

This thesis sounds fundamentally correct to me, and is one I've seen reflected in countless analyses of various modern human ventures. Care to share the book? Sounds like something I'd be interested in reading.

[Peter Damian]

I'm reading a very good book at the moment by an evolutionary biologist who wants to explain how human society turned from hunter-gatherer about 60,000 years ago, when people only knew a handful of other people, and today, when we interact with strangers drawn from over a billion people. His argument is that we are still fundamentally attuned to relationships in small groups. We deal with large groups by all sorts of workarounds. His example is going to the station and buying a ticket. Would you give something very valuable to a total stranger? Not without qualification. You give it to the person on the other side of the counter because of all sorts of background knowledge. It's a counter, they have a uniform, the railway company has been there for a long time, there are methods of recourse etc etc.

The idea that 'anyone can edit', without any kind of qualification, is complete madness. Note that the much-derided idea of credentials is simply another workaround. You don't know whether you can trust this person's work, but someone else who you do trust has vouched for it. So OK.

All these workarounds – to get over the problem that modern civilisation emerged from the DNA of hunter-gatherers – have themselves evolved over thousands of years. There may be something better, but it's difficult to devise something at short notice that works. Wikipedia, by contrast, threw away all of this background knowledge and proclaimed techno-utopia with the MediaWiki software. Jimmy Wales proclaimed that people are fundamentally good.

Well of course people are fundamentally good, but only when they all know and trust each other. They aren't fundamentally good in large groups without these slowly evolved workarounds.

This thesis sounds fundamentally correct to me, and is one I've seen reflected in countless analyses of various modern human ventures. Care to share the book? Sounds like something I'd be interested in reading.

Wired for Culture by Mark Pagel. Still in hardback.

[Vigilant]

I'm reading a very good book at the moment by an evolutionary biologist who wants to explain how human society turned from hunter-gatherer about 60,000 years ago, when people only knew a handful of other people, and today, when we interact with strangers drawn from over a billion people. His argument is that we are still fundamentally attuned to relationships in small groups. We deal with large groups by all sorts of workarounds. His example is going to the station and buying a ticket. Would you give something very valuable to a total stranger? Not without qualification. You give it to the person on the other side of the counter because of all sorts of background knowledge. It's a counter, they have a uniform, the railway company has been there for a long time, there are methods of recourse etc etc.

The idea that 'anyone can edit', without any kind of qualification, is complete madness. Note that the much-derided idea of credentials is simply another workaround. You don't know whether you can trust this person's work, but someone else who you do trust has vouched for it. So OK.

All these workarounds – to get over the problem that modern civilisation emerged from the DNA of hunter-gatherers – have themselves evolved over thousands of years. There may be something better, but it's difficult to devise something at short notice that works. Wikipedia, by contrast, threw away all of this background knowledge and proclaimed techno-utopia with the MediaWiki software. Jimmy Wales proclaimed that people are fundamentally good.

Well of course people are fundamentally good, but only when they all know and trust each other. They aren't fundamentally good in large groups without these slowly evolved workarounds.

This thesis sounds fundamentally correct to me, and is one I've seen reflected in countless analyses of various modern human ventures. Care to share the book? Sounds like something I'd be interested in reading.

Wired for Culture by Mark Pagel. Still in hardback.

Excellent.
His amazon page is chock full of interesting titles.

http://www.amazon.com/s/ref=ntt_athr_dp ... evancerank

[greyed.out.fields]

Is this along the lines of theory that the reason Jesus had 12 apostles, a petit jury has 12 members and that football teams (in their initial historical forms) had 11 to 15 members is because these are the numbers that made up prehistorical human bands?
I must admit I have core group of about that number that of other editors I interact with (are "wiki-friends" with, if you must) on a long term basis. (I'm comfortably in the top 1000 - 2000 for both edits and articles, by the way.)

[HRIP7]

Depends what we are talking about; the original topic is the problem is anonymity and "moral bankruptcy". Which I think is a more pressing problem (in general) than the fracas over a portion of the articles.

Let's go back to my main thesis, which is that the ease of editing Wikipedia makes it like a large railway terminus or rush hour traffic. Therefore people on Wikipedia behave like people in large crowds – often hostile to each other, or suspicious, or unfriendly. Admins do know one another, more or less, but this makes it worse. They are like a small group of people trying to police a large crowd of strangers. End result is inevitable and predictable. An insular group, hostile to outsiders and often brutal to random strangers who were well-intentioned but rubbed them up the wrong way. A further process of Darwinian selection means that only the thickest-skinned of admins will survive the ordeal. Think about the kind of people that are employed at the doors of busy pubs or clubs. Got it?

Hence there is a trade-off between ease of access to Wikipedia, and its apparent 'moral bankruptcy'.

I'm reading a very good book at the moment by an evolutionary biologist who wants to explain how human society turned from hunter-gatherer about 60,000 years ago, when people only knew a handful of other people, and today, when we interact with strangers drawn from over a billion people. His argument is that we are still fundamentally attuned to relationships in small groups. We deal with large groups by all sorts of workarounds. His example is going to the station and buying a ticket. Would you give something very valuable to a total stranger? Not without qualification. You give it to the person on the other side of the counter because of all sorts of background knowledge. It's a counter, they have a uniform, the railway company has been there for a long time, there are methods of recourse etc etc.

The idea that 'anyone can edit', without any kind of qualification, is complete madness. Note that the much-derided idea of credentials is simply another workaround. You don't know whether you can trust this person's work, but someone else who you do trust has vouched for it. So OK.

All these workarounds – to get over the problem that modern civilisation emerged from the DNA of hunter-gatherers – have themselves evolved over thousands of years. There may be something better, but it's difficult to devise something at short notice that works. Wikipedia, by contrast, threw away all of this background knowledge and proclaimed techno-utopia with the MediaWiki software. Jimmy Wales proclaimed that people are fundamentally good.

Well of course people are fundamentally good, but only when they all know and trust each other. They aren't fundamentally good in large groups without these slowly evolved workarounds.

Yes. Put that in the book. It's good.

[dogbiscuit]

Having seen 15 minutes of The Tube this week, I thought that there is an analogy there. The theme was fare dodging on the London Underground, and there was the typical shrug of the shoulders "It's London, everyone does it." from someone caught not paying.

It strikes me that there is a similar crowd activity at play - a perceived anonymity, a lack of interaction with the majority of fellow travellers, a sense of right regardless of the real moral position. It is repeated around the world, the default position for paying a fare on a train of a human being is not fundamentally good, but, certainly amongst a wide swathe of society, the assumption appears to be that if they believe they will not get caught, they will try not paying - and don't have a problem with it, more are even aggrieved at being caught. Any train system that relies on honesty is backed up by a punishment system and it is the punishment system that enforces honesty, not human nature.

Anyone who relies on honesty needs to understand what the trade-off is and there are no real downsides to Wikipedia dishonesty - even if you are a named and shamed real life person except in the case of public figures.