Beefy Boxes and Bandwidth Generously Provided by pair Networks
No such thing as a small change

Anonymous Identifier

by pryrt (Abbot)
on Feb 23, 2021 at 19:32 UTC ( #11128706=monkdiscuss: print w/replies, xml ) Need Help??

Recently, the anonymous account is mostly one infamous user, one detractor who seems to think that every anonymous post (other than his) are by that infamous, and an increasingly-rare useful post.

I understand the desire for an anonymous account -- it lowers the barrier of entry for one-off questions, and it probably helps with GPDR.

But often, especially in the last year or so, it seems to me to be more trouble than it is worth. It allows infamous monks to hide behind a cloak that sometimes (but not always) masks who they are and how dangerous their "advice" is. But it also allows angry monks to carry out vendettas against the infamous monks any time there's a non-zero probability that an anonymous post might be from that infamous monk. And, on those rare occasions when the AM isn't one of those two, it makes it hard to follow questions, "no, I'm not that AM, I am a different one, the one from id://...."

Most forums I've visited don't allow any anonymous posts. Do the negatives here outweigh the positives? And if TPTB don't want to disable AM, could we at least add a non-identifying identifier to AM posts?

Something I've thought of before, I finally suggested in CB after today's anonymous-edit, and am now reiterating here: I would suggest a one-way hash on the IP address -- so it wouldn't tell us who or where the person is, but it would say "this is likely the same AM as from that other post". For the useful AM conversation, it would help other readers follow which AM said what; and for the infamous and his detractor(s), it would make it easier to confirm or deny whether it is likely the same monk or not. Both seem like "wins" to me. I understand that IP addresses can change or be behind big corporate blocks, so it's not a foolproof identifier in either direction (same IP might feasibly be used by good AM and bad AM, or a single AM's IP might change between posts) ... but it might help some. As long as the particular hash is not also applied to logged-in posts, I wouldn't think it would run afoul of GPDR (but, IANAL, so take it for what it's not worth).

Anyway, just an idea I've had, and since there was some support in CB, I finally decided to suggest it officially.

Replies are listed 'Best First'.
Re: Anonymous Identifier
by LanX (Saint) on Feb 23, 2021 at 19:49 UTC
      ++ to judge posts by quality no by (presumed) author -- I definitely agree with that sentiment.

      To me, RFC: Hide Very Bad Answers From Visitors doesn't seem to have been as effective as it was intended... or monks are considering posts for reaping without allowing the hide-very-bad-answers to take effect.

      I think we have enough measures to handle this.

      The need for your plea in the first case, and the number of considerations of bad anonymous posts rather than letting them get de-emphasized properly, leads me to believe that the measures aren't sufficient, or aren't being used properly. Hence the impetus for my PMD.

      Further, that doesn't address the other aspect: the "no, I'm the other anonymous monk" issue for following discussions. And a non-reversible, non-personal, but somewhat-uniquely-identifying hash seems a good way to solve that problem.

      The latter problem has actually bothered me almost since I first arrived here, but it was never strong enough to be worth suggesting. But now that I've realized that the same solution will also help with the other issues I brought up, I thought it was worth discussing.

        Techies tend to look for technological solutions for problems, but IMHO this is rather a social one.

        Though discussing an identifier per IP may be worth it in general, I'm sceptique here.

        IMHO evolutionary pressure will only push abuse into other patterns, like VPN tunnels or throw away accounts.

        So please forgive me if I'm not deepening this discussion any further.

        For me it's just another variant of feeding in this asymmetric war. :)

        Cheers Rolf
        (addicted to the Perl Programming Language :)
        Wikisyntax for the Monastery

        HINT: You needed to log in and vote to collect those votes. The gods can identify you now ...

        Good night! :)

        Cheers Rolf
        (addicted to the Perl Programming Language :)
        Wikisyntax for the Monastery

Re: Anonymous Identifier
by pryrt (Abbot) on Feb 23, 2021 at 19:47 UTC
    Similarly, I feel the prevalence of anonymous nodes considered because they are probably by the infamous monk is getting pretty high. I understand that the infamously bad advice needs to be dealt with, but knee-jerking a Consideration just because it's anonymous and not overly helpful seems overkill to me. Having the hashed-identifier might make it easier to decide whether to KEEP, REAP, or to abstain (as I usually do in all but the most egregious instances of anonymous bad advice)

      Note that Janitors can unconsider a node... I do feel your pain though, but I'm not at the point where I feel we should eliminate Anonymonk account... yet.

Re: Anonymous Identifier
by aitap (Curate) on Feb 25, 2021 at 19:46 UTC
    I understand the desire for an anonymous account -- it lowers the barrier of entry for one-off questions, and it probably helps with GPDR.

    FWIW, here are two related discussions 1 2 from the SQLite forum and the forum of its related project, the Fossil SCM. Ever since they moved from the mailing list to a forum hosted and developed in-house, anonymous posting was allowed, subject to pre-moderation. Some people did't like anonymous posts and replies, but so far the ability to post anonymously has been judged useful enough to stay. I realise that the more "democratic" PerlMonks as a whole has to pay somewhat bigger price for that, since the SQLite community never sees what their moderators don't approve; on the other hand, every time I post anonymously, an SQLite forum moderator gets an extra piece of work.

    I post both here and over at the SQLite forum when I have a spare minute or two and want to feel helpful. This total absence of barriers between me and helping someone is great; I would like to preserve it, if possible.

    P.S. Yes, the "tough beans" anti-spam feature is annoying, but it does beat the alternatives.

Re: Anonymous Identifier
by Fletch (Bishop) on Feb 23, 2021 at 19:52 UTC

    A hash of the (say) /24 part of their address (and/or the IPv6 equivalent) would probably be a decent compromise being fuzzy enough to reasonably smear things likely from the same source. I'd be most surprised if there were someone commenting from the same block as the infamous one. If the hash were stable and made it into the CSS somehow as an attribute or class that would be even nicer.

    The cake is a lie.
    The cake is a lie.
    The cake is a lie.

      I ran an ISP as the senior global network engineer for a decade, and I am against blocking based on IP block or in most cases address level. I've always believed that any risk of collateral damage is too much risk.

      IPs can be spoofed (although that's admittedly become much harder in today's age), people forget to remove the restriction, admins can mistype the prefix and wipe out an address space way larger than they intended (yep, I've done that one... blocked most of Australia once by blackholing way too large a block, and took me a while to sort out WTF because the problem didn't crop up for a few days after the change).

        Oh no I didn't mean block as in block the IPs, but as in hash every IP in the same /24 net block to one identifier so that e.g. someone bouncing their cable modem probably would still hash to the same token.

        The cake is a lie.
        The cake is a lie.
        The cake is a lie.

Re: Anonymous Identifier
by Anonymous Monk on Feb 23, 2021 at 19:38 UTC
    the anonymous account is mostly one infamous user

    are you sure?

      are you sure?

      No. And that's part of my point.

      During my drafts of that post, I thought I had said "it seemed" or "it feels like"; I should have left that in. But the argument between infamous-or-maybe-some-other-anonymous-poster and the AM whose gut reaction is to label many posts as belonging to the infamous (even when, IMO, the post doesn't carry enough infamous DNA to decide one way or the other) feels to me like it dominates the anonymous scene, to the point of annoyance.

Re: Anonymous Identifier
by Anonymous Monk on Mar 02, 2021 at 03:14 UTC

Log In?

What's my password?
Create A New User
Domain Nodelet?
Node Status?
node history
Node Type: monkdiscuss [id://11128706]
Approved by LanX
and the web crawler heard nothing...

How do I use this?Last hourOther CB clients
Other Users?
Others lurking in the Monastery: (6)
As of 2023-12-03 06:55 GMT
Find Nodes?
    Voting Booth?
    What's your preferred 'use VERSION' for new CPAN modules in 2023?

    Results (20 votes). Check out past polls.