AFPCode [F] v0.2beta
CyberCat
mu...@lspace.org (Murky) wrote:
>Marriage
I didn't see anything about duels in the list. I saw somebody
suggesting "I've been dueled over" so there should be a I've dueled
one as well. As having a duel doesn't say anything about the number of
proposals you've done I'd suggest something like:
M-d+ (I rarely propose, but I've been involved in a challenge (the d
part) and actually fought the duel (the +). You could add a second +
for "and I won" which isn't entirely clear in my case.
>
>Education
>
Could somebody translate this? I have no idea how these things compare
to the weird system we Dutch use.
--
cybe...@introweb.nl | Yesterday my goldfish died.
| It happened sometime late last night.
"Ha. Aha. Ahahahaha....." | I just wanted a midnight snack,
Terry Pratchett | but now I want my goldfish back.
Nat
>
>mu...@lspace.org (Murky) wrote:
>
>>Marriage
>
How about using the : thing; the first half for proposing:the second half
for being proposed to?
Although I still score nil on both counts (:
On the book front, what about using B--- for "I beg/borrow/steal my books
from friends/relatives/the library" and probably B! or B? for "I don't
read Pratchett?"
Nat
***ever so slightly neurotic***
Flabbergast
Murky wrote:
<huge snippage>
> This gives
>
> ----- Begin AFPCodeblock -----------
> AFP Code 0.2beta AE/P d s+:+ a- UP+ R+ P->++ OSW3+: C+++ M-- L+++ c-
> B+ Cn+:++ PT+++(++) 5++ X+ e++ r++ y?
> ------ End AFPCodeblock ------------
<more snippage>
There's nothing wrong with the idea, OTOH I think it's great. So don't
take this as any critizm at all, but there's one thing that kind of
worries me with all of this.
Soon we'll see people add this to their sig., as well and proper it is
after all, IMHO, perfect sig. material, was thinking of doing it myself.
Here comes the cath, the Codeblock is four lines. IIRC, and I belive I do,
a sig. shouldn't exceed four lines. Which leaves no place for anything but
the AFPCodeblock, yesno, and how many will not want to keep atleast parts
of their old sig. Things like "to reply remowe XXX" or these on AFP so
much liked appointments (or what ever to call them) for example "keeper
of "this"" or "member of "that"" [1]. Which leads, for those who cares
about netiquett, to a somewhat difficult situation of wanting both. And
ofcourse to some who just won't care, therefore creating irritation,
giving the mood[2] we have seen to much of lately maybe even flaming.
It well may be that I just paranoid, in this case I hope I am, but
considering this, I feel that purrrhaps we should find a way to avoid
even the risk, of more of these sad things, if possible. In this case it
would be, if we could agree on another look on the Codeblock, that would
leave room for such as above mentioned thing.
My suggestion, surely worthless, would be to cut out the "decorations"
thus leaving it to be a two-liner.
Using Murky's example above it would give:
AFP Code 0.2beta AE/P d s+:+ a- UP+ R+ P->++ OSW3+: C+++ M-- L+++ c-
B+ Cn+:+ PT+++(++) 5++ X+ e++ r++ y? end
But that doesn't look very good, maybe if it was framed by something,
wiggely lines purrrhaps:
~AFP Code 0.2beta AE/P d s+:+ a- UP+ R+ P->++ OSW3+: C+++ ~
~M-- L+++ c- B+ Cn+:+ PT+++(++) 5++ X+ e++ r++ y? end ~
It still don't stand out enough, I am at loss.
Help please...
I'm just no good at these things.
Let's hope it's only paranoia.
TTFN
[1] why don't I have one of those. *sulk*
[2] I'm sure you all know exactly what I'm refering to
Iwan Lamble
In article <33CC15...@goteborg.mail.telia.com>, Flabbergast
<annika....@goteborg.mail.telia.com> writes <snip>
>Soon we'll see people add this to their sig., as well and proper it is
>after all, IMHO, perfect sig. material, was thinking of doing it myself.
>Here comes the cath, the Codeblock is four lines. IIRC, and I belive I do,
>a sig. shouldn't exceed four lines.
>My suggestion, surely worthless, would be to cut out the "decorations"
>thus leaving it to be a two-liner.
>Using Murky's example above it would give:
>AFP Code 0.2beta AE/P d s+:+ a- UP+ R+ P->++ OSW3+: C+++ M-- L+++ c-
>B+ Cn+:+ PT+++(++) 5++ X+ e++ r++ y? end
>But that doesn't look very good, maybe if it was framed by something,
>wiggely lines purrrhaps:
> ~AFP Code 0.2beta AE/P d s+:+ a- UP+ R+ P->++ OSW3+: C+++ ~
> ~M-- L+++ c- B+ Cn+:+ PT+++(++) 5++ X+ e++ r++ y? end ~
Somewhere else in this (extremely long) thread someone said something
about an AFP code decoder. I've no idea how this would work but if it is
going to be automatic there will have to be something unique in the code
for the decoder to pick up on it. It's porbably best to put this in now
instead of everyone having to change theeir .sigs later.
Is there any real need to the ------- Begin AFPCodeblock ---------
bits? If people don't know what the code block is about then they will
not be able to understand it anyway.
The all on one line idea is good but unfortunately some people (like me)
have automatic wrapping of text (Turnpike - I think quite a few people
use that). The AFP code is too long to fit on one line. I suppose I
could try to turn the wrapping off but that would just be a nusance for
the rest of my post.
Therefore you must either forget the {enter} and just run on the line
even over wraps (would this work with a decoder?). This however may
cause the code to be split awkardly like half way through a code.
The new line aught to start at the same place on everyone's codeblock. I
suggest that since there are 20 codes (haven't looked at v0.3 yet so
this will have to change), if you have to have a split, it should come
after the M code as that is no. 10:
AFP Code 0.2beta AU d-( ) s-:+ a-- UP++ R+ P- OSW9:+ !C M---
L+++ c B+ Cn- PT--- 5 X e>e+ !r y? end
It still doesn't stand out too well though. A : after the 0.2beta wmay
help matters (example in my .sig).
I don't think that there is any real need for ~s to make the codeblock
stand out. The AFP Code is practicle rather than looking nice. You can't
really do anthing to make a string of characters look nice apart from
placing them all in a box which would take at least 3 lines, 4 if you
want to keep to the netiquette of wrapping the text.
Also, if you are going to have a decoder you need to make sure that
everyone's .sig is the same. I think that a 'coder' program would be a
good idea and then you could take the resulting text out of that and put
it in your .sig. I have no idea how you would make on though.
Sorry to change my .sig before any announcement but it was far too long
with the official version and there is no decoder as yet so it shouldn't
really matter. When an official verdict is reached on the size of the
codeblock I will change it asap.
--
Iwan Lamble
AFP Code 0.2beta: AU d-( ) s-:+ a-- UP++ R+ P- OSW9:+ !C M--- L+++ c B+ Cn- PT--
- 5 X e>e+ !r y? :end
Adam Jones
Recently, Murky amused an FBI wiretap squad thusly:
> Males use y, females x, if your gender is not for public knowledge use
> z.
Should I be worried that I parsed this as "pubic knowledge"?
--
Adam Jones (Ad...@yggdrasl.demon.co.uk) (http://www.warwick.ac.uk/~mauei/)
Darrell Ottery
In article <33d78a74...@news.ftech.net>, Murky <mu...@lspace.org>
writes
>In alt.fan.pratchett, Nat was seen to say...
>>On the book front, what about using B--- for "I beg/borrow/steal my books
>>from friends/relatives/the library" and probably B! or B? for "I don't
>>read Pratchett?"
>
>B--- I steal books from ralatives/the library when nobody is looking
B---- I am solely responsible for making Terry the most shoplifted
author.
--
Darrell [INTJ] - Dar...@lspace.org - http://www.toreador.demon.co.uk/
Never hit a man with glasses. Use a baseball bat.
M. Riosa
On Tue, 15 Jul 1997 07:58:35 GMT, mu...@lspace.org (Murky) wrote:
>In the style of geekcode and Gothcode, I have been working on AFPCode.
>
>Here it is, version 0.2beta
<snip>
>P--- Psions are a con perpetrated by the computer industry to take
> away money from people who can't afford a real laptop computer.
>P? What's a Psion?
>!P I absolutely refuse to discuss Psions.
P---- I own a Windows CE palmtop and was not threatened with a cabbage
to take ownershp!
Michelena (Pvelo1 owner)
--
"Smile and remember, if the sky falls,
have clouds for breafast!"
patrick(unless he nicked it)
Joann L Dominik
Darrell Ottery (dar...@lspace.org) wrote:
: In article <33d78a74...@news.ftech.net>, Murky <mu...@lspace.org>
: writes
: >In alt.fan.pratchett, Nat was seen to say...
: >>On the book front, what about using B--- for "I beg/borrow/steal my books
: >>from friends/relatives/the library" and probably B! or B? for "I don't
: >>read Pratchett?"
: >
: >B-- I read a little bit, and get the book if I like the look of it.
: >B--- I steal books from ralatives/the library when nobody is looking
:
: B---- I am solely responsible for making Terry the most shoplifted
: author.
I thought going from "I get the first books off the shelves" to "I
only buy paperbacks" was a bit of a jump. How about "I buy the
hardcovers when I can, and prefer not to wait for the paperback"
instead of "first ones off the shelf"?
--
This has been your daily Waste Of Time message from Joann.
We hope you enjoy the fine quality of service we prevent you
from finding elsewhere. If you have any complaints, we wouldn't
be a bit surprised. http://www.cloudnet.com/~jldomini
Void
Iwan Lamble <iw...@lamble.demon.co.uk> wrote:
>In article <33CC15...@goteborg.mail.telia.com>, Flabbergast
><annika....@goteborg.mail.telia.com> writes <snip>
>Somewhere else in this (extremely long) thread someone said something
>about an AFP code decoder.
Me. Murky suggested it and I've volunteered. It's going to be written in
C, and compiled first for DOS, then for Windows 3.x or 95 native (depends
which platform the compiler's for when it comes)...
> I've no idea how this would work but if it is
>going to be automatic there will have to be something unique in the code
>for the decoder to pick up on it. It's porbably best to put this in now
>instead of everyone having to change theeir .sigs later.
>Is there any real need to the ------- Begin AFPCodeblock ---------
>bits? If people don't know what the code block is about then they will
>not be able to understand it anyway.
These bits are traditional on codes, 'cause the original (geek code IIRC)
was produced as a PGP takeoff...
Quite apart from anything else, my decoder picks up on them to start
reading in the codeblock...
>The all on one line idea is good but unfortunately some people (like me)
>have automatic wrapping of text (Turnpike - I think quite a few people
>use that). The AFP code is too long to fit on one line. I suppose I
>could try to turn the wrapping off but that would just be a nusance for
>the rest of my post.
>Therefore you must either forget the {enter} and just run on the line
>even over wraps (would this work with a decoder?). This however may
>cause the code to be split awkardly like half way through a code.
Line breaks are no problem - the decoder will just strcat() the lines
together and do a "seek and destroy" on and \n's it finds. Basically, you
could write each code on a different line without too much trouble. I'm
not so sure about line breaks between the code and modifier, though[1].
I'll cross that bridge when I come to it[2]...
>The new line aught to start at the same place on everyone's codeblock. I
>suggest that since there are 20 codes (haven't looked at v0.3 yet so
>this will have to change), if you have to have a split, it should come
>after the M code as that is no. 10:
>
>AFP Code 0.2beta AU d-( ) s-:+ a-- UP++ R+ P- OSW9:+ !C M---
>L+++ c B+ Cn- PT--- 5 X e>e+ !r y? end
That's a problem, though. If we decide to "hard-code" where the linebreak
is expected, then any slight deviation would bugger up the decoder good 'n'
proper. If we leave it optional it's much simpler (the decoder is more
fault-tolerant, and it only takes a couple of extra lines of code).
>It still doesn't stand out too well though. A : after the 0.2beta wmay
>help matters (example in my .sig).
>I don't think that there is any real need for ~s to make the codeblock
>stand out. The AFP Code is practicle rather than looking nice. You can't
>really do anthing to make a string of characters look nice apart from
>placing them all in a box which would take at least 3 lines, 4 if you
>want to keep to the netiquette of wrapping the text.
Point. Why not allow someone to just keep a list with everyone's on (like
elan offered to do), or ask Leo an' the rest of the NELSC[3] if they could
put up an AFPCodes page (possibly linked to an on-line version of the
Dafptabank[4]).
>Also, if you are going to have a decoder you need to make sure that
>everyone's .sig is the same. I think that a 'coder' program would be a
>good idea and then you could take the resulting text out of that and put
>it in your .sig. I have no idea how you would make on though.
Ok. Anyone want to write one, or am I volunteered again?
Actually, I reckon I could get the encoder to use the same .INI file as the
decoder, and since they're both as flexible as I can make them, when later
revisions of the AFPCode come out, only the .INI file should need updating.
>Sorry to change my .sig before any announcement but it was far too long
>with the official version and there is no decoder as yet so it shouldn't
>really matter. When an official verdict is reached on the size of the
>codeblock I will change it asap.
I'm going, I'm going...
>--
>Iwan Lamble
>AFP Code 0.2beta: AU d-( ) s-:+ a-- UP++ R+ P- OSW9:+ !C M--- L+++ c B+ Cn- PT--
>- 5 X e>e+ !r y? :end
[1] Besides, surely most word-wrapping routines break the line at the last
space from the end, not at +, -, etc.
[2] Read "I'll extensively re-write the code when I come to it..."
[3] Non-Existent L-Space Cabal
[4] Well, Leo? Well? An answer? I asked /ages/ ago... :-)
Void
BTW, e-mail address is deliberately mangled to avoid spam bots
Write to userid "oyster" at domain "enterprise.net"
--
-----------------------<Void>--------------------------
AFPurity: 64% http://void.home.ml.org
"Nail it to the counter Lord Ferguson, and DAMN
the cheesemongers"
-------------------------------------------------------
Andy Fawcett
In message <wwv204r...@sfere.greenend.org.uk>
Richard Kettlewell <ric...@greenend.org.uk> wrote:
> l...@lspace.org (Leo Breebaart) writes:
>> please.ign...@this.time (Void) writes:
>>> Me. Murky suggested it and I've volunteered. It's going to be
>>> written in C, and compiled first for DOS, then for Windows 3.x or 95
>>> native (depends which platform the compiler's for when it comes)...
>> C is a language which allows you to write *portable* programs. Why on
>> *Earth* is it necessary for this program to be "compiled for DOS,
>> Windows 3.x or 95 native" or to have anything to do with .INI files?
> In any case C is quite clearly the wrong language for the job, Perl
> being rather more suitable.
Either would be suitable.
I'll have a go at a C version after my holidays, and I'll make the code
available for putting on the L-Space web.
I'll also have a go (once I see the finished code) of writing a
web-based encoder. The only real hard bit here would be the modifiers,
which will take some thought...
--
Andy Fawcett a.f.p. recipes (recip...@lspace.org)
http://www.afawcett.demon.co.uk/ send mail with subject 'send index'
--------------------------------------------------------------------------
The idle mind knows not what it is it wants.
Void
l...@lspace.org (Leo Breebaart) wrote:
>Arghl.
>I've tried e-mailing Void about this earlier, but got no answer. Perhaps my
>attempts at defusing the spamtrap were unsuccessful, or perhaps he/she just
>wasn't interested in what I had to say. But you can't get rid of me this
>easily!
Ummm, you were unsuccessful at defusing the spamtrap.
>C is a language which allows you to write *portable* programs. Why on
>*Earth* is it necessary for this program to be "compiled for DOS, Windows
>3.x or 95 native" or to have anything to do with .INI files?
Because many people who will want to use it may not have access to a C
compiler. Lots of people only use Windows/DOS, and you don't get a nice
freebie compiler with them, do you?
What I originally meant was that I'd post the source up on the web, then
post compiled versions for all the platforms I could. That way *everyone*
should be able to use it.
A configuration file would be used so that each time a new version of the
AFPCode comes out the actual program would need no modifications - just a
simply-formatted configuration file update (the way it's set out at the
moment, even the end user could update it him/her/itself).
>All that is needed (or so I would think) is a filter program that takes a
>string as input (the code), and gives a string as output (the decoding). If
>you must have it, command line arguments can be used to read/write from
>files instead, and the resulting program will still be portable enough so
That's what I'm writing now...
>that easily be recompiled and made available to people who are (gasp) *not*
>on a DOS/Windows platform.
Scarcasm. The lowest form of humour. (Barring most of *my* jokes).
>It is particularly important to have this filter running on Unix, so we can
>wrap a nice web page around it and allow people to decode their AFPCode on
>L-Space Web.
Agreed. Is someone working on this now, or are you waiting to modify my
code?[1]
>> >The new line aught to start at the same place on everyone's codeblock.
>I agree with Void that this is a completely unnecessary restriction.
>The answer's the same as it always has been in these cases: if someone has
>Pratchett- or a.f.p.-related web pages they want to bring under the L-Space
>roof, all they have to do is send mail to ca...@lspace.org and ask.
>Just don't expect us to write those pages ourselves :-)
Bugger, must have missed that bit. Sorry. I'll do that bit, then.
>--
>Leo Breebaart (l...@lspace.org)
[1] Read "Can I take it slowly, or should I be working my assorted
extremities off?".
Void (for future reference, Void is a "he")
Richard Kettlewell
al...@lspace.org (The Bellinghman) writes:
> Oh, indeed it is. Of course, you've just volunteered to port Perl to
> MS-DOS.
MS-DOS? What's that? l-)
--
Richard Kettlewell http://www.elmail.co.uk/~richard/
C Speed
al...@lspace.org (The Bellinghman) writes:
>>In any case C is quite clearly the wrong language for the job, Perl
>>being rather more suitable.
>Oh, indeed it is. Of course, you've just volunteered to port Perl to
>MS-DOS.
Erm, it already has been though. Unless you mean perl5 which hasn't yet
(afaik).
Claire
--
******************************************************************************
* Claire Speed [ENTX] * Network & Operations Unit, Manchester Computing *
* Dial-up, ISDN, TICTAC * C.S...@mcc.ac.uk http://www.mcc.ac.uk/Claire/ *
******************************************************************************
Tim Shuttleworth
In <5r7m52$i8l$1...@lurch.mcc.ac.uk>,
C Speed <per...@mcc.ac.uk> wrote:
>al...@lspace.org (The Bellinghman) writes:
>
>>>In any case C is quite clearly the wrong language for the job, Perl
>>>being rather more suitable.
>
>>Oh, indeed it is. Of course, you've just volunteered to port Perl to
>>MS-DOS.
>
>Erm, it already has been though. Unless you mean perl5 which hasn't yet
>(afaik).
Has. And my spies tell me there's even a rather nice one for Windows which
comes with its own editor.
Tim.
--
Dr Tim Shuttleworth
Network Support Programmer
Manchester Computing
email: Tim.Shut...@mcc.ac.uk
Mike Knell
In article <5r5kji$ged$1...@news.enterprise.net>,
Void <please.ign...@this.time> wrote:
>l...@lspace.org (Leo Breebaart) wrote:
>
>>I've tried e-mailing Void about this earlier, but got no answer. Perhaps my
>>attempts at defusing the spamtrap were unsuccessful, or perhaps he/she just
>>wasn't interested in what I had to say. But you can't get rid of me this
>>easily!
>
>Ummm, you were unsuccessful at defusing the spamtrap.
Hmm. When are people going to work out that spamtraps cause more problems
than they solve? I do not see *why* I should have to jump through hoops,
solve puzzles and wave dead chickens in order to mail people. And if that
mail then drops on the floor due to having missed it, the effort's wasted.
Hmm. Valid addresses *are* required in the From: line, y'know, folks.. [1]
Mike, who filters his spam quite effectively, thankyew, without the
need for spamtraps.
[1] RFC1036. sec 2.1.1. A cite I know backwards..
--
Mike Knell -- a Wholesale Alternative to Good, Safe Murder. ((c) jldomini)
Mr. Tony Blair! You are a thirsty gonk! --- http://www.cs.nott.ac.uk/~mpk/
Leo Breebaart
please.ign...@this.time (Void) writes:
> >I've tried e-mailing Void about this earlier, but got no answer. Perhaps my
> >attempts at defusing the spamtrap were unsuccessful, or perhaps he/she just
> >wasn't interested in what I had to say. But you can't get rid of me this
> >easily!
>
> Ummm, you were unsuccessful at defusing the spamtrap.
I wasn't. I've checked my logfiles, and this is the relevant entry (edited
for readability, and with your address manually mangled):
Jul 19 10:35:30 rama sendmail[12682]:
KAA12680: to=oes...@boobyprise.net, ctladdr=leo (38/1001), delay=00:00:04,
xdelay=00:00:03, mailer=esmtp, relay=mail.enterprise.net. [194.72.192.20],
stat=Sent (JAA26659 Message accepted for delivery)
In other words: you better start bugging you sysadmins, because they're not
delivering your mail like they should. And they aren't bouncing it either,
because I received nothing.
(I am posting the above, rather than e-mailing it, for obvious reasons. On
to the [F] stuff.)
> >C is a language which allows you to write *portable* programs. Why on
> >*Earth* is it necessary for this program to be "compiled for DOS, Windows
> >3.x or 95 native" or to have anything to do with .INI files?
>
> Because many people who will want to use it may not have access to a C
> compiler.
My point was that if the *code* gets too Windows-specific (which the use of
the word .INI definitely suggested to me -- I now understand that it is
simply some sort of configuration text file, yesno?), it will be difficult
to port at all. Sorry if I was mistaken about the Windows-specificness of
.INI files.
> A configuration file would be used so that each time a new version of the
> AFPCode comes out the actual program would need no modifications - just a
> simply-formatted configuration file update (the way it's set out at the
> moment, even the end user could update it him/her/itself).
The best option would be, IMHO, to contact Murky about this, and work
things out so that his own description of the AFPcode (i.e. that what he
posts) can serve directly as input for the decoder without further
modification. With the use of a few unobtrusively placed markers this
should be quite easy to do, and would greatly simplify things.
> >It is particularly important to have this filter running on Unix, so we can
> >wrap a nice web page around it and allow people to decode their AFPCode on
> >L-Space Web.
>
> Agreed. Is someone working on this now, or are you waiting to modify my
> code?[1]
>
> [1] Read "Can I take it slowly, or should I be working my assorted
> extremities off?".
Nobody's working on it yet, and even if we were: just take your time. The
world won't collapse if we go without an afpcodedecoder for a while
longer...
--
Leo Breebaart (l...@lspace.org)
Void
mu...@lspace.org (Murky) wrote:
>In alt.fan.pratchett, Leo Breebaart was seen to say...
>>The best option would be, IMHO, to contact Murky about this, and work
>>things out so that his own description of the AFPcode (i.e. that what he
>>posts) can serve directly as input for the decoder without further
>>modification. With the use of a few unobtrusively placed markers this
>>should be quite easy to do, and would greatly simplify things.
>Quite happy to do this... however I only want to do it ONCE, so can
>the people who are going to have a go come to some sort of standard?
The only probs are that
1) This would involve much processing and some deep magic to get the
program to distinguish between, for instance, the actual specs and
examples, leading to a *much* larger and more unweildy source.
2) The file that holds the AFPCode specs could very easily get overlarge,
quite unnecessarily.
The way I've got the file formatted at the moment[1] is something like
this:
[Category (Eg Chocolate Covered Coffee Beans)]
one or two letter code (eg PU or C)
++ description of value
+ description of value
description of value
- description of value
-- description of value
[Next Category]
etc...
(With a little header at the top for "special" characters, like &>(), etc.)
Is it that onerous a task for someone to just render Murky's postings into
this format? I'll do it if no-one else wants to (I've *got* to stop saying
that...)
>Also, I'd recommend a look at the geek code decoder. This doesn't
>require delimiters on the code, it can either print codes with highest
>confidence, or only print codes it is SURE about.
Ok - shouldn't be a problem. Something for version 2 though, I think.
>(I'd also have an indication for undecodable codes, if poss)
Again - no probs. What I'll do is get it to look at the version number as
well, and warn if the AFPCodeblock is a different version to the
AFPDecoder's config file.
>Murky
[1] But subject to sudden and drastic change.
Void