alt.idiot.marsrules

Long-form reference · 5,880 words · about a 26-minute read

What the record holds

The administrative record of Usenet is kept in two plain text files, and this group appears in both. The Internet Systems Consortium mirrors the namespace at ftp.isc.org: a newsgroups file pairing every group name with a one-line description, and an active file recording which names a server is expected to carry. In the copy dated 25 August 2026 each of the two files holds 45,003 lines, of which 20,466 are in the alt.* hierarchy. One of those lines is this group's, and it is the whole of what the current listing says:

discussion of the usenet kook MarsRules

That sentence is doing several jobs at once, and it is worth separating them. It states a subject, and the subject is a person rather than a topic. It identifies that person by a Usenet handle rather than a name. And it applies the period's standard pejorative for a disruptive poster in the flat register of a catalogue entry, as though the label were a category of thing rather than a judgement about someone. Descriptions in alt.* are not written by any editor or registry; they are lifted verbatim from the For your newsgroups file: line that the group's own proposer puts in the message that creates it. The description is therefore not a neutral summary of the group. It is the proposal's own claim about its purpose, preserved by machinery that has no opinion.

The active-file entry adds nothing but a posting flag. The article-number watermarks in the consortium's copy are placeholders — every one of the 45,003 lines carries the same pair of numbers — and they are not a measure of traffic; nothing in either file records how many people read a group, how many articles it carried, or whether anybody ever posted to it at all. What the listing establishes is narrower and still worth having: that this name entered the namespace, that it was accepted widely enough to persist in the reference copy of it, and that it is still there a quarter of a century later.

The control archive, message by message

Behind the listing sits the second and larger part of the record. The consortium also archives the control messages themselves, one compressed file per group, and for this name the file holds fifteen distinct messages spanning 30 April 2000 to 30 December 2002: nine asking servers to create the group, six asking them to delete it. Read in order they do not describe a discussion in the group. They record an exchange between two senders conducted entirely in the administrative channel, and then, after a gap of two and a half years, a second exchange among three more senders with no evident connection to the group's subject.

The first message is dated 30 April 2000 at 10:51 GMT and follows the alt.* proposal form of the day. It is cross-posted to alt.config and to the group being created; its Summary header compresses three claims into a single line — that the name had been discussed in alt.config, that there was substantial interest, and that the name was appropriate; and its body supplies the newsgroups-file line quoted above, a charter, and a section headed Numerical justification. That last item is worth a note of its own. What it contains is a count of matching articles retrieved from the Deja archive — the Usenet search service launched as Deja News in March 1995 and rebranded deja.com in 1999 — across a stated range of dates, offered as evidence that a topic was already generating traffic. An unregulated hierarchy is here borrowing the vocabulary of evidence from the formal one. This page reproduces neither the search terms nor the figures, for reasons set out at the end.

Of the charter itself only two clauses are reproducible here without doing exactly what this page declines to do, and they are the housekeeping ones: the posting of binaries was prohibited, and unsolicited commercial advertising was discouraged. Both are boilerplate of the period, copied from proposal to proposal across the hierarchy, and their presence in a group of this kind is one of the odder features of the record — the form was followed scrupulously whatever the content.

Five hours and twenty-three minutes later, at 16:14 GMT on the same day, a second message asked every server on the network to remove the group again. It came from the standing alt.config removal account, it was PGP-signed under a key issued in the name of alt.config in April 1999, and its stated reason was a single line:

* Vanity Group

What followed was a slow exchange of the sort the hierarchy had no mechanism for resolving. Three further creation messages, labelled in their own bodies as boosters, went out on 7 May, 11 May and 28 May 2000, each repeating the original proposal text; each was answered by another signed removal from the same account, on 7 May, 11 May and 1 June, the first two of them within hours. A fourth and final booster is dated 25 July 2000, and no removal answered it. Then the file goes quiet for two and a half years. All five creation messages of 2000 carry the same Sender and Approved addresses and were injected through the same news provider, although the visible From address changes partway through; all four removals came from the one signed account. The group's recorded administrative life in its first year is a disagreement between two parties about whether a name should exist.

The twin in another hierarchy

On 1 May 2000, some twenty hours after the first removal, the same proposer created a group of the same name in the free.* hierarchy, with a description of its own in the same derisive register. The consortium's control archive holds that message too, and free.idiot.marsrules is in the current newsgroups file alongside its alt.* counterpart.

The two hierarchies are configured alike in the shipped defaults, which is what makes the manoeuvre available. The control.ctl file that ships with the INN news server — the file in which an administrator records whose control messages to honour for which hierarchies — annotates free.* as an open hierarchy where anyone can create a group. Its default policy for free.*, as for alt.*, is to accept all creations and silently discard all removals. A name removed from one open hierarchy could simply be asserted in the next one along, and the record shows that being done inside a day. This is the structural fact that most of the argument described below eventually runs into.

December 2002: the name as collateral

The last four dates in the archive have nothing to do with the group's subject. On 20 December 2002 a removal message for the group arrived from an entirely different sender, at a Dutch consumer provider, giving as its reason that the group was a stealth binaries group and a binaries group outside the alt.binaries hierarchy. The group was neither, by its own charter, which prohibited binaries; the reason line was boilerplate being applied indiscriminately, and the message's Summary header supplied the principle it was working from: “Every rmgroup deserves at least one repetition”.

Within the hour, a third sender, injecting through a Finnish university's news server, issued a creation message reversing it, describing itself as a booster sent to counteract what it called a rogue removal and pointing readers to the consortium's archive for the original proposal. The sequence repeated on 21 December. On 28 and 30 December two further creation messages arrived from a fourth sender, this one an openly automated tool whose headers advertised that it worked from a standing list of removal issuers whose messages it was configured to reverse on sight, that list being revised by vote every three months.

Three senders acted on the name that December, and none of them was the proposer or the alt.config account; neither of those two appears in the file after July 2000. The name had become a token in a general-purpose quarrel about who was entitled to issue removals in alt.*, of the kind that occupied the hierarchy for years. The archive ends there. Nothing has been added to it since 30 December 2002, and the name has sat undisturbed in the namespace ever since.

The word the description uses

The term in the listing was not a coinage of this group's proposer. It was established Usenet vocabulary with a settled sense, and the clearest period definition is in the Jargon File, the collaborative hacker glossary, which records the word as originally and more formally net.kook and defines it as a term used to describe a regular poster who continually posts messages with no apparent grounding in reality. The entry is careful to distinguish it from a neighbouring word: it is “different from a troll, which implies a sort of sly wink on the part of a poster who knows better”, the same sentence adding that the people the term describes really believe what they write. The distinction being drawn is one of sincerity. A troll was understood to be performing; the other label asserted that the person meant it.

That is a strong claim to encode in a newsgroup name, and the glossary's own entry qualifies it in passing, noting that such posters are usually filtered out by the other regular participants in a group and that the difficulty falls mainly on newcomers, to whom the necessity for the filtering is not immediately apparent. The same entry offers the era's practical argument for keeping track: it records instances of journalists writing stories with quotations from posters who had caught them unaware.

Around the word sat a small vocabulary of remedies, all of them individual rather than collective. The kill file — also called a bozo bin or twit list, and first implemented in Larry Wall's rn newsreader — let a reader tell their software to discard articles matching a pattern in the author, subject or other header lines. To announce that you had done so was itself a convention: the Jargon File dates plonk, which it glosses as the sound a newbie makes as he falls to the bottom of a kill file, to the newsgroup talk.bizarre, and records that as of 1994 it was widespread on Usenet as a form of public ridicule. The etiquette literature of the period recommended the filter; the announcement was a custom of the network's own making rather than anything the guidelines proposed.

alt.usenet.kooks, and what a tracking group said it was for

Groups devoted to cataloguing disruptive posters were not an invention of the late alt.* period. The genre had a recognised home, and its creation message survives with a date on it. alt.usenet.kooks was created by a control message dated 20 December 1993, sent after discussion in alt.config, with the newsgroups-file line “Discussion of the kooks on usenet.” The creating message is explicit about method, and because it describes the practice in general rather than any individual, it can be quoted here as the plainest statement of what such a group understood itself to be doing:

Examples of their work will be posted here and it will be a meeting place for kook-hunters to gather and swap stories and keep one another up to date as to where the new budding kooks can be found. It is hoped that through the careful use of the Followup line, kooks can be lured into crossposting to alt.usenet.kooks.

Three things are worth drawing out of that. The first is the term the message uses for the group's intended participants — kook-hunters — which frames the activity as pursuit rather than observation. The second is the archival ambition: examples of a person's posting were to be collected and republished in a place they had not chosen. The third is the last sentence, which proposes a technique, the careful setting of the Followup line so that a target's replies would be carried into the tracking group. That is not merely discussion of a person; it is a documented mechanism for moving their posting somewhere they had not chosen, described in the message that brought the group into existence.

The group persisted, though not unopposed. Its own control archive holds an anonymous removal message from February 2000, sent through a mail-to-news gateway, giving as its reasons the disintegration of the group and its being a troll group and a flame group; and two later booster creation messages, in 2005 and 2006, asking sites that did not carry the group to add it. The current newsgroups file lists nineteen names beginning alt.usenet.kooks, most of them second-level entries naming an individual, and 103 alt.* names contain the era's term somewhere in them.

Awards, FAQs and the culture's own archives

What makes the practice an institution rather than a habit is that it kept records, ran ceremonies and appointed officers. By mid-1994 alt.usenet.kooks had a maintained FAQ and a monthly award. The evidence is in the administrative record again: a booster creation message for the group dated 25 June 1994, sent in reply to a reader who had asked whether the group could be refreshed because his site did not carry it, is signed by a poster describing himself as keeper of the group's FAQ and coordinator of its Kook of the Month Awards, with an organisation header reading “Kook-of-the-Month Awards HQ” and an FTP address at netcom.com where the award archives, the group FAQ and the net.legends FAQ were all kept together. The net.legends FAQ, compiled by someone else, records one such award made for May 1994, so the ceremony was running by the first half of that year.

Later variants of the award proliferated in the same fashion as the groups did: the current newsgroups file still carries a name for a century-scale version of the same joke. There was merchandise, too. Joel Furr, a Usenet personality of the early and mid-1990s who is credited by Brad Templeton as among the earliest to apply the word spam to network abuse, created a line of Usenet T-shirts on the theme, reported in Wired in 1995. By the middle of that decade the practice had a maintained FAQ, a named coordinator, an archive of past results and a line of shirts.

The net.legends tradition

Running alongside the tracking groups, and repeatedly confused with them, was a distinct and older practice: the recording of the network's notable characters as folklore. Its central document is the Net.Legends.FAQ, subtitled Noticeable Phenomena Of UseNet, compiled by David DeLaney, last modified 13 September 1994, issued in four parts with a stated posting frequency of every seventy-three days, and archived through news.answers at rtfm.mit.edu under the subdirectories for alt.folklore.computers and alt.usenet.kooks among others. It was later mirrored on killfile.org by Tim Skirvin, who noted on the page that he was not its author, was in no way affiliated with the author, and had put it up only because the document had become hard to find — after which, he wrote, his copy became its primary distribution method. The same name appears in the administrative record in 1996, as the proponent of the news.admin.net-abuse.* reorganisation described further down this page.

The FAQ is explicit that it is not the same thing as a tracking group. It states flatly that it is not the FAQ for alt.usenet.kooks, and that many of the people in it are there only because they were noticeable, adding that “some are not by any stretch of the imagination” members of the category the tracking groups existed for. Its own account of its origins makes the drift audible: it began in a thread on alt.folklore.computers entitled “Call For Loons”, and the compiler records that it became clear in the course of compiling it that “notoriety, not looniness, should be the distinguishing characteristic” of the people gathered in it. The list acquired an editorial policy while it was being assembled.

The document also states its own evidentiary standards. It warns that no real attempt has been made to verify that the people described hold the opinions attributed to them or exist at the addresses given; that much of the material consists of contributed recollections which are not guaranteed accurate but are only what the contributor remembered; and that the compilation is biased, in most cases towards the humorous side, “because they are ultimately written by people who are biased”. It offers a right of reply, inviting anyone described in it to correct their own entry, and records that one person did so and received a rewritten entry as a result.

The case the practice made for itself

The argument for cataloguing individuals was made in the period in three distinct forms, and it is fairer to the record to separate them than to treat them as one position.

• Warning newcomers. The regulars of a group could recognise a persistent disruptive poster within a line and filter accordingly; a newcomer could not, and might spend weeks in an argument that everyone else had stopped having years earlier. A public record shortened that learning period. The Jargon File makes this argument inside its own definition, noting that the necessity for the filtering is not immediately apparent to newcomers, and citing journalists caught into quoting posters as evidence that the problem reached beyond the network.
• Concentration. If a dispute was going to happen anyway, a dedicated address kept it out of the group whose subject was being displaced. This is the same reasoning that produced alt.flame — listed in the current newsgroups file as “Alternative, literate, pithy, succinct screaming”, with a hundred and twenty names beneath it — and it is an argument about traffic management rather than about people.
• Self-defence by a community with no other remedy. Usenet had no administrator with authority over its users. A site could disconnect its own customer and nothing else; there was no mechanism by which a group could exclude anyone. Publicity, in this account, was the only sanction available, and refusing to use it meant having no sanction at all.

The third form is the one the objections below address most directly, and the one that the formal machinery described later in this article did least to answer.

The objection from inside the practice

One contemporaneous statement of the risk was written by a compiler of one of these lists, in the list itself, in 1994. The Net.Legends.FAQ carries a disclaimer warning readers that any attempt to e-mail the addresses it prints may produce results ranging from complete silence to mailbombing to complaints to the reader's own system administrator, and then turns the point around on the reader:

consider what it might be like to have *your* name appear on one of these things... and wake up with 1,000 pieces of mail in your mailbx

The asterisks and the misspelling are the original's. The passage sits in the document's own front matter, addressed to the readers it is inviting to enjoy the compilation. The same FAQ addresses the consent question in a single phrase: a contributor who would rather their own emailed contribution not appear is invited to say so, but the compiler adds, excluding emailed legal threats, that he is “counting posts as fair game”. That distinction — between what a person sent privately and what they had posted — was the working doctrine of the practice, and no forum existed in which it could be settled.

The answer made to that doctrine at the time is recorded in the administrative channel rather than in prose, and this group's own file is one place to read it. A public posting is a thing that can be quoted and replied to; a standing group named after a person is a permanent entry in a configuration file, replicated to every server that accepts the hierarchy and answerable in no forum at all. That is the distinction the vanity-group objection turned on, and it is why the same signed account removed such groups without reference to the merits of the particular case. Neither position prevailed over the other, because there was no body competent to decide between them.

Where the formal machinery drew its line

Usenet did build a formal apparatus for handling abuse, and it wrote personal conflict out of its own remit in as many words. Two net-abuse groups already existed, created after votes reported on 4 April 1995 — the misc group by 442 to 91, the announce group by 391 to 182. In 1996 the hierarchy was extended and reorganised through the full Big-8 procedure, in order to spread the load those two were carrying: a Request for Discussion was posted on 9 July 1996 by the same person who would later mirror the Net.Legends.FAQ, and a result was reported in news.announce.newgroups on 9 November 1996 in which each proposed group passed by a wide margin — the e-mail group by 451 votes to 28, the Usenet-abuse group by 447 to 32, the policy group by 441 to 36. The charter that passed with those votes says this, spelling as in the original:

The news.admin.net-abuse.* hierarchy is for discussion of abuse of Usenet and/or the Internet; it is not for discussion of abuse of groups or indiviuals on such networks, such as flames, personal attacks, or off-topic messages.

The hierarchy that existed to police the network's abuse thus declined, by an explicit and voted-upon clause, to be the venue for disputes between people. Its topic list is a list of system-level offences — excessive multi-posting, excessive cross-posting, chain letters, forged or third-party cancellations, broken gateways, forgery of another user, forged approvals in moderated groups — and it includes, in the same breath, “abuse of the newsgroup creation system”. Both halves of the argument about groups like this one can be read out of that single charter: the formal system refused to handle personal conflict, which is one reason the informal groups existed at all, and the formal system regarded misuse of the creation machinery as an offence in its own right, which is one reason such groups were removed on sight.

Where a formal sanction against an individual was attempted, it was itself contested. The Usenet Death Penalty — an agreement among administrators to cancel or refuse a source's traffic, announced in advance and applied after the operator of the offending server has been given chances to correct the problem — had been developed against providers that would not control their own outbound abuse, and was used in that form against several large ones in the late 1990s. During the newsgroup conflict described in the next section, a despammer with the ability to block posters from multiple servers declared a limited version of it against individual posting addresses; the measure had previously been reserved for servers. Other posters criticised the move on the ground that consensus limited the penalty to originating servers rather than users, and as targets altered their addresses and handles he ran into technical obstacles in applying it to individuals at all. Both objections are recorded in the same place as the action.

The era's etiquette literature took the narrower and quieter line throughout. RFC 1855, the Netiquette Guidelines published as FYI 28 in October 1995, tells its readers twice in the same words not to get involved in flame wars and neither to post nor to respond to incendiary material; on personal attack specifically it observes that sexually and racially harassing messages may also have legal implications, and the practical suggestion attached to that observation is filtering software rather than any collective response. That document, and the satirical tradition that inverted it, are treated at length on this directory's page for alt.complainers.bitch-n-moan and are not retold here.

When the tracking became the thing complained of

An adjacent episode of the same years gave both sides of the argument a concrete thing to point at, and both sides pointed at it. The Meow Wars, which began in 1996 and ran for more than a year, started as a quarrel between the occupants of two newsgroups and escalated into a campaign of crapflooding — posting a large volume of nonsense messages to swamp on-topic conversation in other groups. When targets used kill files, participants repeatedly altered their message headers and text so that their messages would bypass the filters, a practice known as morphing. When posters criticised them, participants used Deja News to find those critics' other newsgroups and moved the campaign there. Mail systems at more than one university were rendered inoperable by floods sent, typically, through anonymous remailers.

Two consequences of that episode bear on this group's file. The first is that floods of forged control messages in the same period caused hundreds of oddly named newsgroups to appear at many sites, and one recorded response was that some users adopted digital signatures to verify the authenticity of such messages — the same measure the removals in this group's archive rely on a few years later. The asymmetry is visible in the archive: the four removals of 2000 arrive PGP-signed under a key issued in the name of alt.config, and the two of 2002 under a personal key, while none of the nine creation messages carries a signature at all. The second is that the episode was readable in two directions: as evidence that organised group-based targeting escalates once it starts, or as evidence that the network needed better records of who was doing what to whom.

This page reaches no verdict on that, and the participants did not reach one either. The news.admin.net-abuse groups, the alt.config regulars and the tracking groups argued the question for the better part of a decade without any of them being in a position to bind the others.

alt.idiot.* as a naming genre

The name at the top of this page belongs to a recognised alt.* form, and its scale is measurable in the current record rather than estimated. The newsgroups file dated 25 August 2026 lists seventy names beginning alt.idiot., another eight in the variant alt.idiot-savant branch, a hundred and twenty beginning alt.flame., nineteen beginning alt.usenet.kooks, and 103 alt.* names containing the era's term somewhere in them. Some of these are aimed at organisations, at software, at nothing in particular. A large proportion are second-level names consisting of a person's handle or name.

The genre worked by placing the assertion in the name rather than in the content, and that is the feature that made it attractive and the feature that made administrators dislike it. A group's name is not a posting: it is a permanent entry in a configuration file, replicated to every server that accepts the hierarchy, printed in every listing of the namespace, and impossible to answer. A posting can be rebutted in the thread below it. A name has no thread below it. For a great many of these groups the record supports the reading that the creation message is the entire event and that nothing else survives.

Not every mocking name in the hierarchy points at a person. The same construction was applied constantly to fictional characters, corporations and products — this directory carries alt.fan.starwars.jar-jar-binks.die, whose target cannot be harassed, and alt.netscape.sucks, whose target was a browser — and the current file holds 105 alt.* names containing sucks and twenty-eight ending in die. The distinction between a mocking name and a name aimed at a private individual is the only distinction that matters for the purposes of this page, and the namespace itself does not record it. Both look identical in the file.

What administrators did about such names

The response recorded in this group's own archive is the standard one, and it is a statement about the category rather than about the merits: a signed removal giving as its sole reason that the name was a vanity group. The vanity-group objection in alt.* was not primarily about vanity in the ordinary sense — it covered any group whose name was a proper noun attached to a person, whether flattering or not, on the reasoning that the namespace was a shared resource and personal names in it were an appropriation of it. This directory's page for alt.music.ontario-emperor takes up the self-named version of the same form; the mechanics of alt.config, and of why a removal message is only ever a request, belong to alt.support.jock-strap and to the alt.* hierarchy page, and are not repeated here.

The general administrative view of the naming traffic is stated in the shipped configuration itself. The control.ctl file distributed with INN advises the administrator of a small site to drop alt.* control messages altogether and add groups only on user request, on the ground that, in its own words, “Tons of alt.* newgroups are sent out regularly with the intent more to create nonsense entries in active files than to actually create a useable newsgroup.” The same comments propose a test for the ones that are not nonsense: check whether a syntactically valid newgroup message for the name exists in the consortium's control archive, since many junk control messages are invalid and are never archived at all. That is the archive this page has been reading, and it is why the fifteen messages described above exist to be read.

The same file is candid about the limits of its own authority: “there is no official, generally accepted alt.* policy and all information about alt.* groups available is essentially someone's opinion, including these comments”. It adds that there are nearly as many policies on alt.* groups as there are Usenet sites. That comment, shipped in INN's own default configuration, is as close as alt.* came to a statement of governance.

Why the names outlive the argument

The asymmetry that decided this group's fate is structural and has nothing to do with its subject. Creating a name in an open hierarchy takes one message that most servers accept by default. Removing it takes a message that most servers, by the same default, discard unread. Six removal requests were issued for this group by two unrelated parties over two and a half years; the name is in the reference copy of the namespace today, and so is its twin in free.*.

The consequence is that the namespace preserves accusations better than it preserves anything else about them. A thread decays: articles expire from servers after weeks, archives are partial, and the arguments that produced a group are mostly gone. The name does not decay, because it is not content — it is configuration, and configuration is copied forward indefinitely by machines that do not read it. Whatever this group's proposer intended in April 2000, the durable result was not a discussion. It was a line in a file, and the line is the part that survived.

Why this directory keeps the address

Somewhere in the referring literature — the sort of link list, course page or bibliography that once mapped Usenet for people who had to navigate it — a reference resolves to /alt/idiot/marsrules.html, and a reference that resolves to nothing is a small piece of damage done to the record. Keeping the address alive is cheap; letting it break is not reversible.

What a preserved address owes its readers is an honest account of what was at it. For most groups in this directory that means describing what the group discussed. For this one it cannot, because what the group set out to discuss was a person, and describing that in any detail would mean republishing material about a private individual for no purpose except completeness. The alternative taken here is to document the record that exists, and to treat the group as an instance of a practice that is historical, documented, and was argued about at the time by the people conducting it. A reader arriving from a citation gets the administrative facts, the institutional context, and both sides of the contemporaneous argument. They do not get the contents of the group, and they are not meant to.

What the record does not show

The gaps here are large and worth listing, because the temptation with a group like this is to fill them from imagination.

• Whether anybody posted. The consortium's files record names, not traffic, and no article archive available to this directory holds messages from this group. One of the later boosters asserts a message count for the group's first full month, but a booster is an interested party arguing for its own group's continued existence, and the assertion cannot be checked against anything.
• Who read it. There is no subscriber figure for any Usenet group, then or now. The readership reports circulated during the 1990s were estimates extrapolated from a sample of participating sites, and they are not a headcount for any individual group.
• What was said in alt.config. The proposal claims prior discussion there. The discussion itself is not in the control archive, which holds only control messages, and this directory has not located an archive of alt.config for April 2000 that it can verify.
• Whether the removals were coordinated with anyone. The signed removals give a one-line reason and no discussion. Whether the objection was contested at the time, and where, the archive does not say.
• When the group actually stopped. Groups in alt.* do not close; they simply stop receiving articles, on a date nobody records. The last control message is dated 30 December 2002. That is not the same thing as an end date, and this page does not present it as one.

Scope and limits

The group's own material amounts to one listing line, one active-file line and fifteen control messages. Everything else above is the documented history of the practice that produced such groups, and the article stops where that documentation stops. The one direction in which this page could have been extended indefinitely — the individual named in the group's name — is closed on purpose, and closed at every point where the option arises.

Specifically, and deliberately: no attempt has been made to identify the individual behind the handle in the group's name, no real name appears here, no allegation or anecdote from the group or its neighbours is reproduced or summarised, no example of the individual's posting is quoted or characterised, and no figure of any kind accompanies this article. The description line is quoted once because it is the record's own words and the page would be dishonest without it; the era's term is otherwise reported as a term, in the mouths of the documents that used it, and not adopted here.

Everything factual above is drawn from sources that can be checked: the Internet Systems Consortium's newsgroups and active files as retrieved on 25 August 2026, and the control.ctl file distributed alongside them, last modified 5 August 2023; the consortium's control-message archives for alt.idiot.marsrules, free.idiot.marsrules and alt.usenet.kooks; the archived Big-8 creation messages for the news.admin.net-abuse.* groups, which carry the vote figures and the charter verbatim; the Net.Legends.FAQ of 13 September 1994 and its mirror page; the Jargon File; RFC 1855; and the encyclopaedia articles linked above. Where a figure could not be verified it has been left out rather than approximated, and where the record is silent this page says so instead of reasoning about what was probably the case.