response 26 of 128:

Mar 23 11:52 UTC 2000

I think what Eric is referring to is a command in the gate
editor to replace one piece of text by another.  But it only
applies to text that you haven't posted yet.

If by "self-censorship" you mean the right to remove one's
own words from the public record, I disagree quite strongly.

response 28 of 128:

Mar 23 16:18 UTC 2000

It *is* possible for a FW or the author of an item to kill an entire item.
I don't think the killed item goes into any logs.  So there is a way to
self-censor, if you don't mind blowing up other people's responses.

response 30 of 128:

Mar 24 23:48 UTC 2000

Re the podium analogy in #14:  The way things are is not necessarily the way
things ought to be.  Just because one can't self-censor in most forms of
communication, that doesn't mean one shouldn't be able to.  It doesn't
necessarily mean one should be able to either.  I just don't see that that
argument holds much weight in either direction.

response 32 of 128:

Mar 25 01:15 UTC 2000

Technically speaking, given grex not having the source code to picospan, could
the scribble command be made to actually scribble responses?  If the answer
is yes, then I'd be willing to put this to a membership vote.  Even given the
caveat of someone having made a copy of a posted response, I'm in favor of
the scribble command actually scribbling.  If it technically can be done, then
the membership should vote on it.

response 34 of 128:

Mar 25 04:31 UTC 2000

Scribble (and censor) can be disabled by making /bbs/censored something
that can't be written.  Scribble & censor will silently discard text if
/bbs/censored is linked to /dev/null.  Neither command tries to do a
chmod - making the file 600 would result in a file that can't be
publically read, but nothing stops people from saving their own copies
before the fact (and no doubt some clown would take to saving *all*
their bbs sessions for just this eventuality).

The real issue here is just what should scribble/censor do, and there's
quite clearly absolutely no consensus there, and more than two strongly
different theories on just how it should work.

As for what it does right now, that's in some part historical accident,
and some part "this is a reasonable solution".  The censor command
originally came about because the NETI people, in designing eForum,
thought that censorship was an essential part of putting Picospan into
"business" clothes.  Probably they were wrong, but in the process of
this, PicoSpan got slightly different flavours of the same logic --
"scribble" (which does what "censor" did in eForum), and "expurgate"
(which for a long time was called "censor" on M-net, and may have been
called other things elsewhere).  For quite some time, fair-witnesses
could also censor comments, and this produced some inevitable abuse
problems on m-net.  I did several "tweaks" to attempt to deal with this,
and the first was that (really from a very early stage of things),
censored things went into /usr/bbs/censored.  I believe this file was
publically readable for a long time, which is why most of the
"old-timers" here on grex just know that's where stuff ends up.  I
suspect most people who read through that file think that it would be
impolite to comment on what shows up there, or to take offense or
otherwise react to what the author clearly thought better of, after the
fact.  I did two more tweaks for grex; the first was to change "censor"
into "expurgate", because it's a less loaded word, and enough people had
convinced me it was a useful tool for dealing with stupid spelling
mistakes that I was convinced disabling censor was not best, and the
second was to entirely remove the ability of f-w's to delete responses.
I believe both changes have worked out very well for grex, so as an
"artist", I feel I did good.  So far as leaving /bbs/censored publically
readable, obviously that removes some of my reason for having it so in
the first place.  I can certainly see the "surprise" factor in not
having the help warn about this possibility; at the same time, I'm not
exactly sure I would want to encourage people to just go rummaging
through it.  There's probably some sort of middle ground of wording that
would accomplish this, but there's clearly some matters of socialization
involved here, and etiquette, and all that, and I haven't yet come up
with the one short sentence that will squish all that meaning into a few
well-chosen pithy words that will make the right thing obvious.

Now you know the historic reasoning, & my thinking, and the
possibilities.  I suspect a "True" consensus may not be possible here,
but it would be nice if this resolved itself in a way that was not as
decisive as the anonymous web conference reading fiasco.

response 36 of 128:

Mar 25 12:57 UTC 2000

Scribble should have a warning and folks should know there is 
no such thing as taking it all back here, because there isn't.

If we were to go to a system where someone could come back
and permanently remove text from a discussion thread then you'd
end up with USNET type posts, where everyone quotes the comment
they are responding to before or after their new comments.
I've often thought that was a cumbersome way of getting to 
new material.

I've used the scribble command when I've so mangled a response
that I simply want to have another go at it without all the
typos and grammatical errors.  That way I don't ask someone to
have to re-read the beta version too.  They can if they like,
no problem.  

response 38 of 128:

Mar 25 16:27 UTC 2000

<expurgated and scribbled>

response 40 of 128:

Mar 25 21:11 UTC 2000

I see several alternatives here:

  CURRENT STATUS - expurgate just hides a response, scribble removes it
    from the item and puts it in the log file which is publically
    readable.

    Nobody likes this much.

  SCRIBBLE WORKS - Same as current state, but censored log would not
    be publically readable.

    What was originally proposed here.  I dislike the idea, as do
    several other old-timers.

  SCRIBBLE WARNS - Same as current state, but a warning message tells
    you that there is still a readable copy of your message.

    OK with me, but a bit silly.

  ELIMINATE SCRIBBLE - The scribble command is disabled.  You can still
    expurgate responses.

    My personal preference.  As long as we can't really delete the
    response from the history of the universe, why even pretend to?

Backtalk can be adapted to any of these options, except that in the
SCRIBBLE WARNS case, I'd feel like, in all fairness, I'd have to provide
a way for Backtalk users to read the censored log.  I'm not enthusiastic
about this.

response 42 of 128:

Mar 26 03:51 UTC 2000

I think it's premature to think about going for a vote.

response 44 of 128:

Mar 26 05:46 UTC 2000

Expurgate hides a response?  Define "hides".  

response 46 of 128:

Mar 26 06:04 UTC 2000

An expurgated response is still stored in the item file, but has a flag
set indicating that it should not be displayed by default.

Suppose response 47 is expurgated.

To view it in picospan, do "set noforget" and then display the response
(for example, by doing "only 47" at the respond or pass prompt.

In backtalk, there will be a message displayed instead of the response
that says something like <click here to view hidden message>".

So it's not difficult to view in either system, but much less difficult
in Backtalk.