alt.music.ontario-emperor

Long-form reference · 4,765 words · about a 21-minute read

Everything the record holds, which is nothing

The administrative record of alt.music.ontario-emperor, as it stood when this page was researched on 25 August 2026, consists of no documents at all. A finding of that kind is worth only as much as the searching behind it, so the searching is set out in full before anything else is said.

The Internet Systems Consortium maintains the public archive of Usenet control messages — the small administrative articles by which newsgroups are created and removed — filed one compressed file per group name, in subdirectories by top-level hierarchy. Its own policy document describes it as an archive of all newgroup, rmgroup and checkgroups control messages seen on Usenet since early 1991, records that a month or so of messages around August 1992 are missing, and states that no guarantee is made that it is full and complete. The alt.* subdirectory of that archive holds 36,208 per-group files, 1,013 of which are names beginning alt.music. There is no file for this group. In the alphabetical listing the sequence runs alt.music.one-minute-silence, then alt.music.operation-ivy, with nothing between them.

ISC also publishes the two reference files most people mean when they speak of a list of Usenet newsgroups: active, which names groups and their posting status, and newsgroups, which pairs each name with a one-line description. The policy document for that directory states that it updates hourly; the copies consulted for this page carry a modification time of 12:00 GMT on 25 August 2026. The newsgroups file has 45,003 entries, of which 20,466 begin with alt. and 864 with alt.music. The active file has the same 45,003 entries. Neither contains alt.music.ontario-emperor. In the description file the same two neighbours sit on consecutive lines, 13,718 and 13,719, with no gap where a third might once have been.

Two further bodies of evidence were searched and are equally empty. The archive of checkgroups messages — the third kind of control article, which carries an entire group list at once and asks servers to reconcile their own lists against it — is filed by year; all thirty-one files in it, spanning 1993 to 2026, were decompressed and searched, and the string does not occur in any of them. The ISC configuration logs, which record what the archiving host did with each control message it received and why any message was rejected, exist as 280 monthly files running from May 2003 to August 2026; all 280 were fetched and searched, and the only occurrences of the string Ontario belong to an unrelated binaries group whose name happens to contain it.

Finally, the place where a Big Eight group's paperwork would live does not apply here at all. ISC mirrors the archive of news.announce.newgroups, in which a Big Eight proposal's Request for Discussion, Call for Votes and result are filed by hierarchy. Its per-hierarchy subdirectories are comp, humanities, misc, news, rec, sci, soc and talk; administrative directories and two long-running list series sit alongside them, but there is no alt directory, because alt.* groups were never proposed, discussed or voted on there. The two list series draw the same boundary in their own words: the roster posted monthly from that archive is titled the List of Big Eight Newsgroups, and the older Bogus Usenet Groups postings state in their explanatory notes that they cover the eight hierarchies of comp, humanities, misc, news, rec, sci, soc and talk, together with INET groups. For a group in the alt hierarchy there was never a file of any of those kinds to lose.

Summarised, the searches performed for this page were these:

• ISC control-message archive, alt.* subdirectory (36,208 group files) — no file for this name.
• ISC newsgroups file of 25 August 2026 (45,003 entries; 20,466 alt.*) — absent.
• ISC active file of the same date (45,003 entries) — absent.
• ISC checkgroups archive, all thirty-one files, spanning 1993 to 2026 — no occurrence.
• ISC configuration logs, all 280 monthly files, May 2003 to August 2026 — no occurrence.
• news.announce.newgroups archive and its two Big Eight list series — no alt.* branch exists to search, and the lists exclude alt.* by their own stated scope.

What follows is therefore not the history of a newsgroup. It is an account of the kind of name this was, of the machinery that allowed such names, and of why the machinery kept so little.

A note on what this page is about

The last component of this group's name is a personal handle. This directory documents newsgroups; it does not document the people whose names, handles or personas appear in newsgroup names, and no attempt has been made here to identify whoever used this one, to describe them, or to find what was said about them anywhere.

The verifiable subject of this page is an address in a namespace and the administrative system that produced it, and on that subject the sources quoted below are precise. Everything else that could be written under this heading would be about a private individual, and would be neither verifiable to the standard this directory holds itself to nor any of its business.

The check an administrator was told to make

There is a reason the absence of a control message matters more than an ordinary archival gap, and it is written into the configuration file that ships with the standard Usenet server software. control.ctl is the file in which a news server records whose control messages it will honour for which hierarchies, and the alt.* section of the version ISC distributes is preceded by several paragraphs of advice to administrators. Among them is a recommendation of a specific test:

Quality, user-desirable new groups can often be discovered by a quick perusal of recent alt.* newgroup messages after discarding obvious junk groups. One good initial filter is to check the archive of control messages for a requested group to see if a syntactically valid newgroup message was issued. Many of the junk control messages are invalid and won't be archived, and many sites will only add alt.* groups with valid control messages.

The file then prints the exact archive address to look at, which is the archive searched above, and adds that the first and last few messages should be read to see whether the group ought to be moderated. The test recommended to administrators deciding whether a requested alt.* group was real is the test applied to this name, and this name fails it. It is worth being clear about what that is a statement of: not that the group never existed anywhere, but that the check the era's own tooling recommended returns nothing.

The rest of that stanza — the four active lines, which tell a server to honour alt.* newgroup messages from anyone except the two Big Eight control addresses that never issue them, and to ignore every alt.* rmgroup from everyone, along with the file's frank concession that no official, generally accepted alt.* policy exists — is set out in detail on this directory's page for alt.support.jock-strap, whose whole archived record is three control messages, and is not repeated here.

What a valid message would have left behind

The control archive is not the only place a properly formed newgroup message would have registered. ISC's group-creation policy document sets out how the two reference files are maintained, and it treats alt.* as a special case. In that hierarchy and in free.*, all properly formatted newgroup messages are executed automatically, all rmgroup messages are ignored, and no manual change is made except under very special circumstances and with substantial justification. The moderation status and the description of a group are simply set to whatever the last control message for it contained.

The consequences for a name are therefore asymmetric: the message that would put it into the published files is acted on without anyone's approval, and the message that would take it out again is discarded unread. The two sources still disagree about a great many names, and the arithmetic is worth stating rather than glossing. Of the alt.* names appearing in either, 20,350 appear both in the control archive and in the current list, 15,858 are archived but not listed, and 116 are listed with no archived control message. A file in the archive can exist because a removal message was sent for a name as readily as because a creation message was, so the two sets were never going to coincide. What can be said without straining is that alt.music.ontario-emperor belongs to neither of them.

The same document is specific about what a newgroup message had to contain to be acted on. It had to be a properly formatted Usenet article carrying an Approved header and a syntactically valid Control header; it had to be no larger than 64KB; and it had to carry a line in the body reading exactly For your newsgroups file:, with no leading whitespace, followed on the next line by a description for the group. The name was checked as well: a component could contain only lower-case letters, digits and a short list of permitted punctuation, had to contain at least one non-digit, could not be all or ctl, and the whole name could be no longer than eighty characters. Those criteria are said there to follow the rules in RFC 5536. They are the conditions a message had to survive in order to leave any of the traces looked for above, and a message that failed them left nothing at all.

Four ways a name reaches a roster without reaching the archive

A name can appear on a gateway's list of carried groups in the early 2000s and be absent from every administrative record in 2026 by at least four routes, and the honest position is that the evidence does not choose between them.

1. A valid newgroup message was sent and simply did not reach the host that was doing the archiving. Propagation on Usenet was normally fairly good and never guaranteed; the archive's maintainers say as much themselves, and add that the archiving software is not perfect either.
2. A newgroup message was sent and was malformed. Only valid control messages are filed, by the standard set out above, and a great deal of the control traffic aimed at alt.* failed one of those tests; what fails is discarded rather than kept.
3. No control message was ever sent. A news administrator could add any name to a server's own list by hand, on request or on whim, and nothing about that act travelled anywhere. A group carried by one provider and known to its users can leave no global trace whatever.
4. The name reached this directory by some route other than a live group — a mistranscription, a variant spelling, a name someone asked the gateway to carry. This is the least likely explanation and it cannot be excluded.

None of the four can be preferred on the material available, and this page does not pretend otherwise. What can be said is that the first two would each have left a probability of a trace, and there is none; the third would have left exactly what is observed, which is a name in one operator's list and silence everywhere else.

The group named for a person as a documented form

The practice this group belongs to is documented even where the individual instances are not. Wikipedia's account of the alt.* hierarchy records that the alt.fan branch is mostly devoted to discussions of the work and life of famous people — writers, musicians, actors and athletes — and that the same sub-hierarchy has also been used for self-promotion by otherwise unknown people. That describes the form exactly: a naming slot designed for public figures, open to anyone who wanted to occupy it, and costing nothing to occupy.

The scale is measurable. Today's newsgroups file lists 1,984 names beginning alt.fan., and the control archive holds files for 2,765; comparing the two sets, 800 archived alt.fan names are absent from the current list, and 19 listed ones have no archived control message at all. The branch was never restricted to admiration; a name of the form alt.fan.somebody could be a fan club, a running joke, a piece of self-advertisement or an insult, and nothing in the namespace distinguished them. Neither did anything require the person named to be famous, to consent, or to know.

The uglier end of that practice is visible in the archive without being reproducible here. Among the alt.* control files are group names that consist of an individual's name followed by their street address and telephone number, spread across several branches of the hierarchy in near-identical variants. They are not quoted in this article and their subjects are not identified: the point is only that the naming mechanism placed no limit of any kind on what a component could contain beyond the character and length rules, and that the record of alt.* accordingly contains a good deal that a directory has no business republishing. The surrounding culture of pursuing named individuals through the namespace is the subject of this directory's page for alt.idiot.marsrules and is not retold here.

The name measured against the naming advice

The creation guide that circulated in alt.config devotes a section to newsgroup names, and two of its points can be applied to this one without any inference about the group or the person behind it. The first is a length rule. Under the heading Component Too Long the document warns that some systems cannot handle a name component longer than fourteen characters, offers alt.fan.bgcrisis in preference to alt.fan.bubblegum-crisis as its worked example, and notes that the restriction was becoming less critical as software such as INN handled it better, while C News continued to enforce it. The component ontario-emperor is fifteen characters long, one over that threshold. This is not a rare condition in the branch: of the 1,013 alt.music names in the control archive, 73 carry a component longer than fourteen characters, as do 49 of the 864 in the current list.

The second is a separator rule. The same section tells proposers not to use a dot as a word separator but a dash, on the ground that stripping components from the end of a name should not leave a directory unlikely to hold any other group. This name follows that advice. Nothing in ISC's own validity rules would have rejected it either: a fifteen-character component sits well inside the eighty-character limit on a whole name, and every character used is permitted. Whatever became of the name, syntax was not the obstacle.

The guidance is also explicit about what a newsgroup name is for. It tells those who want to read the group that the group is for them; it tells those who do not that it is not; it classifies similar groups together, so that they fall near one another in sorted listings and a small site can choose what to take a feed of. The document's related complaint concerns groups named after acronyms: an acronym, it argues, is not a good identifier of what a newsgroup is about, groups so named spend a significant part of their traffic answering the question of what they are for, and an abbreviation familiar in one country is not recognisable in the rest of the world. A name built from a personal handle is legible on the same terms as the handle itself — to readers who already know it, and not to readers who do not.

What the guidance documents said about groups of this kind

alt.* had no rules, which did not stop it from producing literature about how to behave. The standing document was So You Want to Create an Alt Newsgroup, posted to news.answers every fourteen days and pointing readers to a web copy at Pennsylvania State University. It concedes its own status in the first line — there are no guidelines or rules for creating alt groups, and there is nobody in charge of the alt hierarchy — and describes itself as a guide based on common sense and experience, and a reflection of established procedure, rather than an attempt to codify anything.

Its second section, a list of the common reasons proposed groups were rejected, contains an entry that describes the expected life cycle of a group created around a single person:

Because of newsgroup longevity (see below), newsgroups which are started just to get people to laugh at the name, and/or to get revenge on some specific person, and/or to shock people, are discouraged. They tend to generate a flurry of articles (sometimes) for a maybe even a month or two, but quickly die.

The longevity it refers to is set out in the same document. Some people, it says, insist that once an alt newsgroup is created it can never be destroyed, and re-create it whenever anyone tries to remove it; even without them, mistakes at new sites can revive almost-dead groups everywhere; and so alt groups are effectively immortal, unable to be removed or even renamed. Its summary of the resulting condition is that alt groups never die, they just fade away, though some fade away faster than others.

Those two passages between them describe a short flurry of articles followed by silence, with the name outlasting the flurry indefinitely. The record here does not show the second half. This name is absent from the lists that preserve tens of thousands of alt.* names for which no activity of any kind is visible, which places it outside the category of the fading group the document had in mind rather than inside it.

The guide is candid, too, about how thin the machinery of opposition was. Groups of extremely limited or regional interest might be opposed by some news administrators, it says, and groups not aired in alt.config might not be created by others; but every one of those sentences is about what some administrators will do, never about what any of them must do. Opposition in alt.* was a private decision taken several thousand times over, and it produced no document.

alt.music.*: a branch of proper nouns

The branch this name sits in is largely a directory of proper nouns. Today's newsgroups file lists 864 names under alt.music., and the control archive holds files for 1,013; the great majority are the name of a band, a performer or a genre, and the descriptions are correspondingly terse. Of the 864 listed names, 312 — better than a third — carry no description at all, only the placeholder text the reference file uses where none was ever supplied.

Placing a persona there rather than directly under alt. was, by the conventions of the time, the indicated move. The creation guidance lists .music among the sub-hierarchies it calls fairly well accepted, and the worked example it gives for why a group should sit inside a classification rather than at the top level is alt.music.monkees rather than alt.monkees — a group which does exist, and whose one-line description in today's file reads, in full, that hey, hey, they were the Monkees. A name is expected to inherit meaning from the branch above it. The difficulty with a branch made of proper nouns is that it can supply only a category, never a warrant: alt.music.x asserts that x is musical, and nothing in the file has ever been able to say whether x is a band with three albums or a handle with an opinion. No authority existed to check a description against a group's contents, and for better than a third of the branch there is no description to check.

This directory holds pages for two of the branch's other inhabitants, alt.music.bluegrass and alt.music.country.classic, alongside this one. In the group list the three names are indistinguishable in form, which is the condition the whole hierarchy lived in.

Who decided what existed

Nobody, is the short answer, and the mechanism is worth stating once because everything peculiar about this page follows from it. A control message is an ordinary Usenet article carrying a Control header; a newgroup message names a group to be created and an rmgroup message names one to be deleted. Sending either creates and deletes nothing. The message propagates like any other article, and at each server the software acts on it or files it away according to how that server's administrator configured the software. Creation was a request put to the thousands of independent news administrators the creation guide addresses, each of whom answered privately and none of whom reported the answer. The hierarchy's origins and the reasons it was built that way are told on the alt.* hierarchy page.

The convention that stood in for a procedure was discussion in alt.config before sending anything, and nothing treats that convention as binding. The creation guide records that some news administrators will not create any group that has not been discussed there and will wait several days for responses; Wikipedia's account of the hierarchy says the same, and adds that groups which skip the step are usually not well propagated. Against that, the default configuration distributed with the server software honours alt.* newgroup messages from anyone. Where the convention broke down — a proposer reporting truthfully that a discussion had occurred, an objector reporting truthfully that it had gone against him, and no third party able to adjudicate — is examined at length on the alt.support.jock-strap page, using a case where both control messages survive in the archive.

Applied to a name built from a handle, the absence of an authority has a particular consequence. There was no body that could have refused the name for being a person's handle, no charter requirement it could have failed to meet, and no vote it could have lost. Equally, there was no body that could have approved it, and no record of approval to point at afterwards. The only sanction the hierarchy possessed was propagation: a group few administrators wanted was carried by few servers and left to sit in a scattering of configuration files as a name. The only reward was the same thing in reverse. Both outcomes are silent, which is why the archive of a hierarchy that ran on argument for decades preserves so little of the arguing.

Why this address survives here at all

news2mail.com was a mail-to-news gateway that carried Usenet to e-mail between 2000 and 2004, and this directory is what its per-group pages have become. Its alt.* section preserves 47 group names, which are collected with the rest on the directory's list of groups. The presence of a name in that roster is evidence of one thing and no more: that the gateway's own list of carried groups contained it at some point during those years, and that somebody thought it worth giving a page to.

That is a genuine piece of evidence, and it is the only evidence this page has. It is not evidence that the group had readers, that anyone posted to it, that it propagated beyond the gateway's upstream feed, or that a control message ever created it. A gateway's roster could be assembled from whatever group list its provider supplied, and a provider's list could contain names its own users had asked for by hand as easily as names the network at large had agreed on. A roster records carriage, not existence, and in alt.* the two were never the same thing.

The page therefore stays, and states the basis on which it stays. What can be said is that in the early 2000s a mail-to-news gateway listed this name among the alt.* groups it would deliver to a subscriber's inbox, and that the administrative record of Usenet has nothing to add to that.

What the record does not show

No newgroup message survives. There is no archived control message of any kind for this name, and consequently no creation date, no sender, no originating site, no proposed description and no statement of purpose. Nothing on this page should be read as implying a date; none has been established, and the era referred to throughout is the era of the directory that lists the name, not of the group's creation.

No charter exists, and there is no reason to expect one to. alt.* groups were never required to produce the formal statement of purpose a Big Eight proposal had to publish, and a great many never did.

No discussion has been located. Whether the name was ever aired in alt.config is unknown; the archive that would show a control message posted there shows nothing, and the discussion itself, if it happened, was ordinary Usenet traffic rather than an administrative record.

No traffic is documented. This page makes no claim that anyone ever posted to the group and no claim that nobody did. Usenet had no subscription registry of any kind, so no figure of that sort exists for this group or for its busiest contemporaries; what estimates the network produced were sampled from sites that chose to report, and were never fine-grained enough to say anything about a name at this scale.

No archive holding has been confirmed. Whether a searchable web archive of Usenet contains anything filed under this name could not be established while writing this page: the attempt to check returned a rate-limiting response rather than an answer, both for this name and for a control name known to be a real group, and an unverified pointer is worse than none. The access box above should be treated as a set of addresses worth trying, not as an assertion that something waits at the end of them.

The name itself is not explained anywhere in the record. Its first element is a place-name carried by a Canadian province, by cities in California and Oregon, by a town in New York and by other places besides; its second is an ordinary English noun. No surviving document says what the pairing was meant to convey, and this page does not guess.

Scope and limits

The group-specific part of this article is short, and its shortness is the finding rather than a failure of effort. Six separate bodies of primary record were searched for the name — the alt.* control archive, the current active and newsgroups files, thirty-four years of checkgroups messages, twenty-three years of configuration logs, and the news.announce.newgroups archive with its two Big Eight list series — and every one of them is empty for it. The remainder of what is written above is what is documented about the naming practice the group belonged to, the branch it sat in, the guidance that existed for people creating such groups, and the machinery that produced all three. When that material ran out, the article stopped.

Two limits of principle apply as well. The first is the one stated near the top: this page is about a newsgroup name and an administrative system, not about the person whose handle supplied part of that name, and no research was conducted into that person. The second is that the long argument alt.* had with itself — whether a hierarchy founded on the proposition that anyone may create anything can coherently object to what gets created — is reported here and not adjudicated. It was not settled at the time, and the hierarchy had deliberately built itself with no place where such a thing could be settled.

The figures in this article are counts taken from the ISC reference files of 25 August 2026 and from the control archive listing of the same date, and they will drift as those files are regenerated. The counts are reproducible: anyone with the two published files can repeat every one of them, which is the only kind of authority a page like this one is entitled to.