2024 Moderation Election ***Update***

[chrisw]

I am not convinced that providing a real name is not a legitimate controller interest here. I would make the case that it is required for credibility of the content. TalkChess is not just a place where anonymous users could exchange chitchat; the sole purpose of its existence is for experts on computer chess to disseminate their knowledge.

There must be precedents for this in scientific publication. There are plenty of websites were you can view scientific papers (for payment), and these all have authors. Providing your name when submitting a paper is not voluntary; without a name the paper would simply be rejected for publication.

Agreed

[Ras]

Amusingly enough you appear to be arguing for real names here ....
https://www.talkchess.com/forum3/viewto ... =2&t=67411

Haha, good one! However, the server back then was not only in the US so that EU law was irrelevant, the GDPR also only came in force after that posting. Plus that I still don't mind using real names on the users' own decision.

[chrisw]

If bad things get said on the forum such that it gets to law, the forum management are going to have a problem, surely, if they made no attempt at knowing who the badsayer was, no?

That's not related to GDPR ofc, but criminal law and depends on the country. But in general, platforms are not responsible for what their users are doing. Otherwise, no platform such as Youtube could even exist because they'd be taken down for copyright infringement alone within 24h. There is only a duty that platforms must remove illegal content once they have gained knowledge of it.

In the EU, we have the new Digital Services Act, and while I havn't looked at it much, it would seem to me that Talkchess already complied with that, by and large. That is, with the charter forbidding illegal content, reporting mechanism, moderation with transparent guidelines (the charter), potential bans for repeated offences. Since Talkchess isn't a very large platform such as Facebook, the heavier requirements don't even apply anyway.

In any case, what would you even want to give to police? A "real name" that doesn't mean anything because it has not been verified in any way? That's not a lead.

I was thinking civil law, libel, reputation damage and so on. Sometimes here moderation can be part of the problem of course. Aren't we duty bound to at least make some primitive effort at establishing poster identity even if we recognise that as a small special interest group with no funding we're not going to be able to exhaustively verify?

[chrisw]

Amusingly enough you appear to be arguing for real names here ....
https://www.talkchess.com/forum3/viewto ... =2&t=67411

Haha, good one! However, the server back then was not only in the US so that EU law was irrelevant, the GDPR also only came in force after that posting. Plus that I still don't mind using real names on the users' own decision.

Okay; so how would you like to phrase the registration document (and a voter registration document if necessary) to allow users to make their own decision?

[hgm]

I would make the case that it is required for credibility of the content.

That goes also out of the window because you are perfectly fine with any random character sequence that looks like a real name, and you don't think that this hampers credibility.

But I am not. This is why we are having this discussion. If I was fine with everything as it was, nothing would have to be done. So that window seems firmly closed...

There must be precedents for this in scientific publication.

Not comparable because in "publish or perish" academics, it is of vital interests for the authors to get their name on as many papers as possible. You also can't just put in any fake real name because you will be enrolled in some academic program where you had to show your ID card at some point.

But there is no pressure to publish your papers on line.

This reminds me of a story a mathematical physicist once told me. A colleague of his had once written a paper of which he was the sole author. (For mathematicians such a thing is easy.) But in the manuscript he consistently used the word 'we'. The referee, who apparenly was not a native English speaker, and was unfamiliar with the 'pluralis modestiae' objected to this. Rather than have the manuscript retyped, the mathematician just added an author 'Felix Domesticus', (which is latin for house cat) to get his paper accepted. Now the funny thing is that a year later Mr. Domesticus was invited to present this paper in a conference. Under the reasoning that they had already seen the mathematician often enough, and wanted to invite his 'co-worker' for a change!

[Ras]

I was thinking civil law, libel, reputation damage and so on.

No. That is the platform privilege. A platform has some reactive duties, as in, not allowing such bad content online knowingly. That's what got the one-click-hosting platforms - they were made for distributing content in violation of copyright and didn't make any serious effort to curb that because, well, it was their business model. Think Megaupload. But luckily, Talkchess is nothing like that.

Aren't we duty bound to at least make some primitive effort at establishing poster identity

No. Just like in the offline world, a bar owner is not required to check the identity of every guest on the off chance that one guest might pickpocket someone and then run off.

Okay; so how would you like to phrase the registration document (and a voter registration document if necessary) to allow users to make their own decision?

The registration could clearly label the real name as optional, with the hint that it will be displayed on every post and can be changed in the profile later. But that also depends on whether you want to use the real name only in voting context, or also for preventing re-registering after getting banned.

How to do some sort of verification, that's more difficult. The only times I've ever used my ID card online was not for identity checks, but for age verification when ordering age restricted goods. Otherwise, that is totally uncommon on the web, and I don't think many users would be comfortable sending around ID card scans. That would pretty much kill Talkchess.

What might work for many (but not all) users here is some sort of website check. Most engine devs have a Github repo or website so that upon registration, a challenge code could be given, and they are expected to provide a link on their website to a text file containing that exact same code.

We'd need some alternative for some of the people, sure, but we could also get rid of some annoying posters who aren't devs or useful testers of anything and only spam the forum with low-key noise. IMO, that's a much more important issue than "credibility". That's a different discussion, though.

But I am not. This is why we are having this discussion. If I was fine with everything as it was, nothing would have to be done.

Maybe I missed that, but my impression so far wasn't that you proposed some sort of verification?

But there is no pressure to publish your papers on line.

Ofc there is massive pressure because with "publish or perish", you have to be found as easily as possible so that as many other authors as possible can reference you and hence increase your ref count. That's an attention economy, and yes, it does damage to science.

This reminds me of a story a mathematical physicist once told me.

LOL, brilliant! It would have been even more hilarious had he actually followed the invitation with a cat.

[chrisw]

ChatGPT:

Can a EU based message board lawfully ask users to provide their real name on registration (under GRPR).

Under the General Data Protection Regulation (GDPR), a message board based in the EU can lawfully ask users to provide their real names during registration, but certain conditions must be met to ensure compliance:

1. Legal Basis for Processing

The message board must have a valid legal basis under Article 6 of the GDPR. Potential bases include:
• Consent: The user voluntarily agrees to provide their real name.
• Contract: If the provision of a real name is necessary to fulfill a service.
• Legitimate Interests: If the platform can demonstrate a legitimate interest, such as preventing abusive behavior or ensuring accountability.

2. Data Minimization (Article 5(1)(c))

• The platform should collect only the data necessary for its stated purposes. If collecting real names isn’t essential, this could violate the GDPR’s principle of data minimization.

3. Transparency (Articles 12-14)

• The platform must inform users why their real name is being collected and how it will be used, typically through a clear privacy policy.

4. User Rights

• Users have the right to access, rectify, or even request the deletion of their data. If users refuse to provide their real names and there’s no compelling legal basis, they might still be able to use the service under pseudonyms.

5. Security and Confidentiality (Article 32)

• The message board must ensure that the collected personal data (real names) are securely stored and not misused.

Conclusion

Yes, an EU-based message board can lawfully ask for real names if:
• It has a valid legal basis.
• It complies with data minimization.
• It provides clear justification to users.

Failure to meet these requirements could result in non-compliance with GDPR, exposing the platform to significant fines.

The board periodically has membership voting for moderators. Can it exclude voters who declined to guve their real name?

Yes, an EU-based message board can lawfully exclude users who decline to use their real names from participating in membership voting for moderators, provided certain conditions under the GDPR and other relevant legal frameworks are met. Here’s how:

1. Legitimate Interest

• Legitimate interest (Article 6(1)(f) of the GDPR) could justify the exclusion. If using real names is essential to ensure transparency, fairness, and accountability in the voting process (e.g., preventing fraud or abuse), this could be a valid legal basis.

2. Conditions for Participation

• If the voting process is part of a contractual obligation (e.g., terms of service specify that voting requires real-name participation), users who decline to provide their real names might not fulfill the conditions required to participate.

3. Data Minimization and Purpose Limitation

• The platform must demonstrate that collecting real names is necessary for voting purposes. It should ensure that no excessive data is collected beyond what is strictly required for this purpose.

4. Transparency and Consent

• Users must be informed clearly (e.g., in the privacy policy or terms of service) that participation in voting requires the use of real names.
• Participation in the platform or voting should be voluntary, meaning users are free to decline providing their real names but would then forfeit the ability to vote.

5. Non-Discrimination and Fair Practices

• Exclusion of users who prefer pseudonyms must align with the platform’s stated rules. As long as this is applied fairly and consistently and users are informed beforehand, the exclusion would likely be lawful.

Conclusion

The message board can exclude users who refuse to provide their real names from voting if:
• The requirement for real names is clearly justified by the need for transparency or integrity in voting.
• Users are informed of this requirement in advance.
• There’s a valid legal basis (e.g., legitimate interest or contractual necessity).

This approach ensures compliance with the GDPR while maintaining the integrity of the voting process.

[chrisw]

I forgot to add, ChatGPT and Ras are pretty much in agreement.

[towforce]

Talk about making a mountain out of a molehill.

+1

[LazySMP]

I think all members should be able to vote. People who have real names have a vote factor of 3, and other people have a factor of 1.

This is formula: Σ(V) = 3*R+F

R:Real name
F:Fake name
V:Vote

We can also define that members who have a real name but were not active in the last year can vote with a factor of 2.