Wikipedia talk:Requests for comment/Ixfd64

From Wikipedia, the free encyclopedia

As an aside, why is it that so many of these blocks are made with the almost deliberately unhelpful summary "user.."? Christopher Parham (talk) 21:43, 12 March 2006 (UTC)

I believe the "user..." block summary is the vandalbot run by Curps. Other than that, I have no explanation (or don't remember it when last I read it) for why it is only "user...". --Syrthiss 12:55, 13 March 2006 (UTC)

I understand that it's a bot, I'm just wondering what technical limitation prevents the bot from using a sensible summary. Christopher Parham (talk) 19:41, 13 March 2006 (UTC)

This is almost a FAQ; if I recall correctly, he said it's to avoid confusing AOL users caught by the autoblocker (which would otherwise say something like "but I didn't do that!"). (And it's not only the bot; he also does that when he blocks by hand.) --cesarb 11:03, 14 March 2006 (UTC)

It's fully explained on the text seen by blocked users editing a page, so no, "user..." is not a bad block reason. Stifle (talk) 13:35, 5 May 2006 (UTC)

John Reid (talk • contribs) made a very valid point. --Ixfd64 22:52, 14 March 2006 (UTC)

[edit] Comments

I would prefer not to further clutter the former page with ranting commentary, so I'll post it here instead. Prior to drafting this document, I sought advice from a couple of users (whom I'd prefer not to identify) who stated, in no uncertain terms, "go directly to arbcom, do not pass go", "zomg desysop", etc. I disagreed, replying that I did not feel that arbitration and/or stiff penalties were necessary, and that my primary objective was to raise awareness and get additional opinions on the matter.

So, they tell me RFCs aren't worth wasting time on, and for a moment, I started to agree. I've been questioned by uninvolved parties via IRC for "spamming" a link to this RFC to the talk pages of other administrators, but actual recipients of those messages thanked me, many of them stating that they unaware of the covert unblockings, others stating that they had never even heard of Ixfd64.

This realization underscores the importance of my original objective, raising awareness. Ixfd64, please be assured I'm not trying to "take your buttons away", but I really feel your activities in this area are counter-productive and your desire to listen to anything I say, and anything Curps says, is minimal. And that's why we're here.

Regarding the "(user...)" in-jokes, they've been discussed in several places ([1], [2], [3], [4], [5], [6]), with varying degrees of silliness, most humorously here. Curps explains it rather well at one of the previous links:

Many of the usernames that get blocked are just throwaway sockpuppets or vandals playing denial of service games or username creation performance art and watching the block log and never even triggering a block (for instance the "block me" accounts... but if we ignore them and don't block, they sometimes go ahead and vandalize anyway, for instance [7]). In other words, much of the time when "user..." is used, the user who actually created the username will never even read or see the block message.

Because of autoblocking, meaningful block summaries very often do more harm than good. In cases of vandalism, I do put "vandalism" in the block summary, but every single day I get e-mail from one or more collaterally damaged AOL users asking "why are you accusing me of vandalism?"

The way that the Mediawiki software handles autoblocking is entirely inadequate and often harmful. At the very least, "established users" (non-throwaway accounts) should be immune from all forms of IP-based blocking, and since autoblocks almost never make sense for AOL addresses, the software should internally detect AOL IP ranges and silently not autoblock them.

Perhaps it's time to get rid of autoblocking altogether. Half the time it doesn't work because the vandal can shift IP addresses, and most of the rest of the time it causes more harm than good because it blocks everyone stuck behind an ISP or school or corporation proxy IP. -- Curps 06:43, 28 February 2006 (UTC)

It would be quite beneficial for Curps to participate in this RFC, but he has thusfar declined to do so.

I agree with him on the issue of the IP autoblocks, to the extent that it would be great if the software would intuitively refrain from enforcing them in screwball-dynamic IP ranges, such as AOL's 152.163.0.0/16. Other AOL IPs, such as 172.128.0.0 - 172.216.255.255 (which I used to belong to) are assigned at the beginning of the dialup connection and remain static until the customer disconnects, or more frequently, gets dropped by the carrier. This mega-range consists of 5,767,168 individual addresses. I used to check, every time I connected, to see if the IP I was assigned to had ever been previously used for anonymous editing, and the contributions lists invariably appeared blank. Thus it would be perfectly reasonable to enforce IP blocks in this range for at least an hour (to prevent the quickfire registration of nine more software-allotted offensive usernames), but again, this would all have to be determined by the developers.

Another particular concern I have is the unblocking of usernames which, (more seriously than impersonating well-known Wikipedia users), impersonate notable living people, such as Bob Dole (talk • contribs). Leaving those names unblocked could result in more complaints to the office if the accounts are used maliciously. This is, in practice, more likely to occur with non-household names, which are less likely to be blocked on sight, as fewer administrators are capable of taking one glance and saying "yup, that's a celebrity name, we have [[NAME|an article]] about that person". Somebody could register a User:Jack Thompson (attorney) or somesuch, and cause a wave of new problems, due to the lack of communication, or whatever else to which one cares to attribute it.

As Wikipedia becomes more of a household name in itself, we need to be more proactive in nipping this sort of shit in the bud. If the username policy has this many perceived loopholes in it, we've got some serious patching to do. We need to take a good, hard look at it, and make policy match practice and avoid "misunderstandings" such as the ones being discussed in this RFC. Give me a few hours. Regards. — Mar. 15, '06 [03:13] <freakofnur_xture|talk>