Re: How best to implement HashCons in user-level code within CL?

Below is a response I posted several days ago but I can't find it
in Google Groups, so I presume something went wrong at our NNTP
server, so I'm posting again. If anybody sees this as a duplicate,
blame it on Google, the giant who is not always our friend, for
refusing to show me an article that **was** actually posted.

If this is not a duplicate, if my first attempt really didn't work,
then blame it on ispnews.usenetserver.com, where I posted it the
first time (and will post this time also) but where it didn't go
through.

From: p...@xxxxxxxxxxxxxxxxx (Pascal J. Bourguignon)
From: p...@xxxxxxxxxxxxxxxxx (Pascal J. Bourguignon)
Use an up-to-date newsreader!
Which one do you recommend that runs on a VT100 Unix term?
gnus on emacs runs perfectly well on unix with a VT100 terminal.

Oops, for some reason I thought you asked me to use an up-to-date
Web browser. Sorry. Anyway, given that I use Google Groups search
engine to find articles that mention 'uh3t' hence might be replies
to something I previously wrote, I have quick question: Google
Groups provides me with the message-ID (as part of the original
full text of the newsgroup article). Given that message-ID copied
from Google Groups, how hard is it to direclty select that
particular article in "gnus on emacs", and how would that be better
than TELNETting to NNTP server and manually pasting in the HEAD and
BODY commands with the <messageID> in each command?
(Before you ask: I don't use ARTICLE command to get both head and
body in one gulp, because that increases chance that serial-output
buffers to modem will overflow and data will be lost. On the other
hand, using the print-to-file command in Lynx then using 'more' to
actually download the article, avoids *all* possibility of buffer
overflow, and of course avoids the nuisance of needing to switch
from Google Groups to TELNET, which is why I usually do that.)

Hmm, I'm going to try TELNET to NNTP server now on that article
with the funny (non-USASCII) characters in what was supposed to
have been ANSI CommonLisp code (which per ANSI specification is
*supposed* to be nothing but USASCII) and see what comes out ...
Message-ID: <4f27a265-da55-419f-8def-5a6ca536940b@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
OK, going into TELNET to NNTP server and entering this command:
BODY <4f27a265-da55-419f-8def-5a6ca536940b@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
produces this line:
(if (/=3D counter *gensym-counter*) ; X or Y new =3D> (X . Y) new
so that didn't help resolve the problem.

Using telnet with 'tee' is a royal pain, so let me see if wget will
allow me to get the text into a file so that I can see what's
really there...
"It supports HTTP, HTTPS, and FTP protocols, ..."
Nope, no mention of NNTP. So let me go through the pain of tee...

% telnet nntp2.tsoft.net 119 | tee tmp-tee
BODY <4f27a265-da55-419f-8def-5a6ca536940b@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
QUIT
% more tmp-tee
...
(if (/=3D counter *gensym-counter*) ; X or Y new =3D> (X . Y) new
...
% emacs tmp-tee
(if (/=3D counter *gensym-counter*) ; X or Y new =3D> (X . Y) new
It really is three different characters, equalsign three dee, not
some non-USASCII character u+003D = (code-char (parse-integer "3D" 16))
being expanded by 'more' et al, because emacs expands non-USASCII
characters a completely different way using backslant that would
show there.

Recently there have been several lisp job offers:
Requires =E2=80=9CTop Secret Clearance with Full Lifestyle Polygraph.=E2=80=9D

All my life I've suffered from social phobia, or social anxiety
disorder as it's now called, whereupon embarrassing or stressful
things cause me great anxiety, and since a polygraph measures
anxiety directly, I could never pass a polygraph test, and it would
cause extreme stress if somebody accused me of lying just because I
suffered anxiety, so to protect me from being harassed and demeaned
in that way to the point where I have to murder the person calling
me a liar, thereby committing my first act of violence against
another human being in my entire pacifist life, it's best I not
even apply for such a job.

Besides polygraph tests have no scientific validity even for
"normal" people who don't suffer SP/SAD, so really it shouldn't
even be legal to require them for security clearances.

Now if they had a test that measured truth/lying rather than
anxiety, that would be nice. Maybe <http://TinyURL/TruFut> will
provide a way that anybody who suspects somebody has told a lie can
post the possible lie and get a market response whether it was true
or not. If somebody testing me for security clearance believes I
told a lie, and posts to TruFut, and turns out indeed I lied, then
I don't deserve the security clearance. On the other hand, if it
turns out I told the truth, then the person accusing me of lying
should be fired from his job as security-clearance tester.

Accenture has several full-time positions open for experienced
Common Lisp programmers, based in Sydney, Australia.

As you are most certainly aware already, many times, I have no
money even for food, much less airline travel or luggage or even a
passport, so I'll need to check whether they accept telecommute
from California to Australia.

Please direct all enquiries to Claudio.Nat...@xxxxxxxxxxxxxx

Damned Google Groups for munging the e-mail address.
Here's where TELNET to NNTP wins:

% telnet nntp2.tsoft.net 119
HEAD <7cab0tizt8.fsf@xxxxxxxxxx>
BODY <7cab0tizt8.fsf@xxxxxxxxxx>
< Please direct all enquiries to Claudio.Natoli@xxxxxxxxxxxxxx
There we go!

Composing my e-mail now ...

Google search for [Accenture digital optimization product] turns up
<http://www.wikio.co.uk/economy/accounting_and_auditing/accenture>
which has a copy of this job-ad, but still no explanation of what
"digital optimization product" means.

I'm asking in that e-mail what they're talking about.

OK, I've finished composing the e-mail inquiring about that Lisp
programming job, and I've sent it, and their SMTP server accepted
it.

Now composing the e-mail to hiring@xxxxxxxxxxxxxxx for the other
reasonable job ad ... done.

So clearly, there is a market for such skills.

Well, we'll see if either of those jobs will allow telecommuting
from here. (I have no money even for food, much less luggage and
travel expenses.)
.

Relevant Pages

  • Re: kokoro ga komotte
    ... >Any storage or forwarding of this message except through a NNTP ... >server is strictly forbidden. ... >copied or quoted in any manner by Google Groups, ... If you want a reply by e-mail, don't write to my Yahoo address! ...
    (sci.lang.japan)
  • Re: How best to implement HashCons in user-level code within CL?
    ... in Google Groups, so I presume something went wrong at our NNTP ... server, so I'm posting again. ... Common Lisp programmers, based in Sydney, Australia. ...
    (comp.lang.lisp)
  • Re: Tricky umpiring conundrum (UK)
    ... Kit wrote: ... Use a real NNTP newsreader, ... Google Groups is an awful, awful way to read & reply to newsgroups. ... RSR is my only use of Usenet and Usenet access alone ...
    (rec.sport.rowing)
  • Re: kokoro ga komotte
    ... >Any storage or forwarding of this message except through a NNTP ... >server is strictly forbidden. ... Google Groups removes it from their archives after 7 days. ... that message could be all over the place in archived posts. ...
    (sci.lang.japan)
  • Re: I cant post to this group
    ... You cannot use e-mail protocols, like SMTP, to connect to NNTP servers used to manage Usenet posts. ... If you want to use webnews-for-dummies interfaces, like Google Groups, you'll have to figure out how to script your client to interface with their webnews interface or find a different client with that support already built in. ... If you are attempting to hide behind an converter gateway service where you send using e-mail to newsgroups then you need to find a mail-to-news service. ...
    (alt.computer.security)