I wanna write an RFC

I wanna write an RFC

[Lars Magne Ingebrigtsen]

No particular reason, really -- I just want to.

One thing that I think is severly underdocumented is the X-Face
header.  It's now getting a bit long in the tooth, and I think ten
years is way to long for an "X-" header to survive, so I want to write 
an RFC that would, like, standardize it and stuff.  "Face" is a nice
header name, isn't it?

Anyway, I just wanted to know whether anyone knows of any previous
attempts to formalize this header.  I've done Metacrawler, and I can't 
really find anything at all about the format.  I have the source to
un/compface (from '90; written by James Ashton), but that's it.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@ifi.uio.no * Lars Magne Ingebrigtsen

Re: I wanna write an RFC

[Jari Aalto+list.ding]

* 1999-03-14 Lars Magne Ingebrigtsen <larsi@ifi.uio.no> list.ding
* Message-Id: <m3g178xbq6.fsf@quimbies.gnus.org>
| No particular reason, really -- I just want to.
| 
| One thing that I think is severly underdocumented is the X-Face
| header.  It's now getting a bit long in the tooth, and I think ten
| years is way to long for an "X-" header to survive, so I want to write 
| an RFC that would, like, standardize it and stuff.  "Face" is a nice
| header name, isn't it?
| 
| Anyway, I just wanted to know whether anyone knows of any previous
| attempts to formalize this header.  I've done Metacrawler, and I can't 
| really find anything at all about the format.  I have the source to
| un/compface (from '90; written by James Ashton), but that's it.

Contact Marty Fouts <gnus@.fogey.com>, He's IETF USEFOR and can
provide all the information you need.

jari

Re: I wanna write an RFC

[David Hedbor]

Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

> No particular reason, really -- I just want to.
> 
> One thing that I think is severly underdocumented is the X-Face
> header.  It's now getting a bit long in the tooth, and I think ten
> years is way to long for an "X-" header to survive, so I want to write 
> an RFC that would, like, standardize it and stuff.  "Face" is a nice
> header name, isn't it?
> 
> Anyway, I just wanted to know whether anyone knows of any previous
> attempts to formalize this header.  I've done Metacrawler, and I can't 
> really find anything at all about the format.  I have the source to
> un/compface (from '90; written by James Ashton), but that's it.

What about creating a new, improved standard instead? The X-Face thing 
is rather dated. A picture with a couple of bitplanes would be nice
for example. Also all implementations of x-face are buggy. A friend
made a version which worked as the description said it should, and it
naturally couldn't decompress any other x-face headers.

I really think it's a rather bad standard. :)

-- 
[ Below is a random fortune, which is unrelated to the above message. ]
Sweater, n.:
	A garment worn by a child when its mother feels chilly.

Re: I wanna write an RFC

[Hrvoje Niksic]

David Hedbor <david@hedbor.org> writes:

> What about creating a new, improved standard instead?  (...)

...and here we have the reason why X-Face never got standardized.
Every time someone suggests implementing it, ten people jump and say
it should be replaced with something better.  Which never gets done,
partly because pictures with a couple of bitplanes would be slightly
too longish to attach to every message.

> Also all implementations of x-face are buggy.

I don't understand this.  All the implementations I know of are using
the "uncompface" code, either through the external program, or through
the library.  Is that code buggy, or do you claim that the
specification is buggy?

Re: I wanna write an RFC

[David Hedbor]

Hrvoje Niksic <hniksic@srce.hr> writes:

> David Hedbor <david@hedbor.org> writes:
>
> > Also all implementations of x-face are buggy.
> 
> I don't understand this.  All the implementations I know of are using
> the "uncompface" code, either through the external program, or through
> the library.  Is that code buggy, or do you claim that the
> specification is buggy?

That code doesn't follow the specification. I don't know exactly what
the error in that code is, but I can find out if anyone is interested.

-- 
[ Below is a random fortune, which is unrelated to the above message. ]
Underlying Principle of Socio-Genetics:
	Superiority is recessive.

Re: I wanna write an RFC

[Norman Walsh]

| What about creating a new, improved standard instead? The X-Face thing 
| is rather dated. A picture with a couple of bitplanes would be nice

In a web connected world, I think it would make a lot more sense to
put the face on a server somewhere and just point to it with the header.
Just my two cents, but I can't see the faces anyway, so nevermind ... ;-)

                                        Cheers,
                                          norm
-- 
Norman Walsh <ndw@nwalsh.com>      | A heavy snowfall disappears into
http://nwalsh.com/                 | the sea. What silence!

Re: I wanna write an RFC

[David Kågedal]

David Hedbor <david@hedbor.org> writes:

> Hrvoje Niksic <hniksic@srce.hr> writes:
> 
> > David Hedbor <david@hedbor.org> writes:
> >
> > > Also all implementations of x-face are buggy.
> > 
> > I don't understand this.  All the implementations I know of are using
> > the "uncompface" code, either through the external program, or through
> > the library.  Is that code buggy, or do you claim that the
> > specification is buggy?
> 
> That code doesn't follow the specification. I don't know exactly what
> the error in that code is, but I can find out if anyone is interested.

As I understand Marcus (out mutual friend who investigated this), the
code uses arrays for some sort of prediction, but it indexes them at
least one index off, resulting in both that the predictions don't work
as expected, and that it reads memory outside the table.  This
out-of-bounds memory happens to be in another structure in the
compface code, but since everybody uses the same code, it always works
the same way.  However, if you want to write another implementation
you'll have to reimplement all the bugs.

-- 
David Kågedal        <davidk@lysator.liu.se> http://www.lysator.liu.se/~davidk/

Re: I wanna write an RFC

[Hrvoje Niksic]

davidk@lysator.liu.se (David Kĺgedal) writes:

> As I understand Marcus (out mutual friend who investigated this),
> the code uses arrays for some sort of prediction, but it indexes
> them at least one index off, resulting in both that the predictions
> don't work as expected, and that it reads memory outside the table.
> This out-of-bounds memory happens to be in another structure in the
> compface code, but since everybody uses the same code, it always
> works the same way.

How can this out-of-bound reads ever work cross-platform?  I mean,
X-Faces created on sparcs can be read on both sparcs, intels, and
alphas.  If their creation/rendition were truly the result of
out-of-bounds randomness, I don't see how it could work.

Re: I wanna write an RFC

[Laura Conrad]

Norman Walsh <ndw@nwalsh.com> writes:

> In a web connected world, I think it would make a lot more sense to
> put the face on a server somewhere and just point to it with the
> header.

Lots of us aren't reading mail in a web-connected world.

-- 
Laura 
lconrad@world.std.com

Re: I wanna write an RFC

[Lee Willis]

Laura Conrad <lconrad@concord.com> writes:

> Norman Walsh <ndw@nwalsh.com> writes:
> 
> > In a web connected world, I think it would make a lot more sense to
> > put the face on a server somewhere and just point to it with the
> > header.
> 
> Lots of us aren't reading mail in a web-connected world.

In which case you're gnus wouldn't try and go get it, and you've saved
the bandwidth difference between actually sending an image and sending
the image's location. 

Seems cruel, but I can see both sides of the argument. Personally I like
the way X-Face:s are included in the message since I don't really want
Gnus to go and get an image off the web everytime I open a message, but
I can also see how it would be nice in some cases.

Lee.
-- 
I was doing object-oriented assembly at 1 year old ...  
For some reason my mom insists on calling it "Playing with blocks"

Re: I wanna write an RFC

[Hrvoje Niksic]

Lee Willis <lee@gbdirect.co.uk> writes:

> In which case you're gnus wouldn't try and go get it,

...and you end up not seeing the image.  :-(

The nice thing about X-Face is that it's small (OK, relatively small)
so it can be shipped off with every message.

Not to mention the problem grabbing things via web to show a mail
message is slow, and that w3 itself is big, loads slowly and is
somewhat bulky (no offense to Bill who is doing everything in his
power to make it more usable.)  A hard-core Emacs groupie that I am, I
can't bring myself to use w3 seriously.

Re: I wanna write an RFC

[Didier Verna]

Hrvoje Niksic <hniksic@srce.hr> writes:

> Lee Willis <lee@gbdirect.co.uk> writes:
> 
> > In which case you're gnus wouldn't try and go get it,
> 
> ...and you end up not seeing the image.  :-(
> 
> The nice thing about X-Face is that it's small (OK, relatively small)
> so it can be shipped off with every message.

        And what's more, if you rilly want to use the net to retreive an
image, the picons are for you. Most people now have a picon face very similar
to their X-Face:, except that the image can be of a higher quality, in color,
and obviously bigger.

-- 
    /     /   _   _       Didier Verna        http://www.inf.enst.fr/~verna/
 - / / - / / /_/ /      E.N.S.T. INF C201.1      mailto:verna@inf.enst.fr
/_/ / /_/ / /__ /        46 rue Barrault        Tel.   (33) 01 45 81 73 46
                      75634 Paris  cedex 13     Fax.   (33) 01 45 81 31 19

Re: I wanna write an RFC

[François Pinard]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> The way w3 displays text/html (after turning off all features) in Gnus
> means that it almost doesn't bother me anymore.  I also use it for
> following links in Gnus articles.

It is a real blessing to have w3 around, and so well integrated with Gnus.
Replies work wonderfully, it is even fun to use.  All is nice, spring is
coming, flowers, bees, and the rest.  Thanks to all for the happiness! :-)

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: I wanna write an RFC

[William M. Perry]

Hrvoje Niksic <hniksic@srce.hr> writes:

> Lee Willis <lee@gbdirect.co.uk> writes:
> 
> > In which case you're gnus wouldn't try and go get it,
> 
> ...and you end up not seeing the image.  :-(
> 
> The nice thing about X-Face is that it's small (OK, relatively small) so
> it can be shipped off with every message.
> 
> Not to mention the problem grabbing things via web to show a mail message
> is slow, and that w3 itself is big, loads slowly and is somewhat bulky
> (no offense to Bill who is doing everything in his power to make it more
> usable.)

Emacs/W3 is probably going to die very soon, due to lack of time and
interest on my part.  I just can't afford the time any more.  I've got
funner things like digging up my passport and writing my presentation for
japan to worry about. :)

_But_ the image fetching stuff could easily be done just with the URL
package, which is much smaller than Emacs/W3.

You could asynchronously fetch the image and have some nice persistent
cache of them, etc.  If the fetch did not complete by the time you have
moved onto another article, you just don't see it, but it would still go
into the cache.  This would be 

> A hard-core Emacs groupie that I am, I can't bring myself to use w3
> seriously.

I'd be interested to see if anybody could plug the mozilla stuff into an
XEmacs frame once the dynamic loading stuff gets finished off.

-Bill P.

Re: I wanna write an RFC

[Per Abrahamsen]

wmperry@aventail.com (William M. Perry) writes:

> Emacs/W3 is probably going to die very soon, due to lack of time and
> interest on my part.  

Arrgghh!  It has just gotten useful.  The way w3 displays text/html
(after turning off all features) in Gnus means that it almost doesn't
bother me anymore.  I also use it for following links in Gnus
articles.

Netscape is still used as the "stand-alone" browser, for reading
/. and comics.

Re: I wanna write an RFC

[Harry Putnam]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> wmperry@aventail.com (William M. Perry) writes:
> 
> > Emacs/W3 is probably going to die very soon, due to lack of time and
> > interest on my part.  
> 
> Arrgghh!  It has just gotten useful.  The way w3 displays text/html
> (after turning off all features) in Gnus means that it almost doesn't
> bother me anymore.  I also use it for following links in Gnus
> articles.
> 
> Netscape is still used as the "stand-alone" browser, for reading
> /. and comics.

Arrgghh! Here too.  Just learned enough html to want to use w3 for
editing etc.  Link work-horse too.

But Per, how do you shut off that horrbile blockyness.  I run gnus in
a dark color  (grey20).  I'd rather not spend days slogging thru the
style sheet, learning new syntax etc.
But would like to see web stuff in the same bg and fg as gnus.

-- 
Harry Putnam reader@newsguy.com  
Running Redhat Linux-5.2
See http://www.jtan.com/~reader  for a brief pictorial 
saga of construction work in the trade of "Boilermaker"

Re: I wanna write an RFC

[William M. Perry]

Harry Putnam <reader@newsguy.com> writes:

> Per Abrahamsen <abraham@dina.kvl.dk> writes:
> 
> > wmperry@aventail.com (William M. Perry) writes:
> > 
> > > Emacs/W3 is probably going to die very soon, due to lack of time and
> > > interest on my part.  
> > 
> > Arrgghh!  It has just gotten useful.  The way w3 displays text/html
> > (after turning off all features) in Gnus means that it almost doesn't
> > bother me anymore.  I also use it for following links in Gnus
> > articles.
> > 
> > Netscape is still used as the "stand-alone" browser, for reading
> > /. and comics.
> 
> Arrgghh! Here too.  Just learned enough html to want to use w3 for
> editing etc.  Link work-horse too.
> 
> But Per, how do you shut off that horrbile blockyness.  I run gnus in a
> dark color (grey20).  I'd rather not spend days slogging thru the style
> sheet, learning new syntax etc.  But would like to see web stuff in the
> same bg and fg as gnus.

(setq w3-user-colors-take-precedence t)

-Bill P.

Re: I wanna write an RFC

[William M. Perry]

[-- Attachment #1: Type: text/plain, Size: 2110 bytes --]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> wmperry@aventail.com (William M. Perry) writes:
> 
> > Emacs/W3 is probably going to die very soon, due to lack of time and
> > interest on my part.
> 
> Arrgghh!  It has just gotten useful.  The way w3 displays text/html
> (after turning off all features) in Gnus means that it almost doesn't
> bother me anymore.  I also use it for following links in Gnus articles.

I agree that the integration with Gnus kicks ass, but there is no way that
I am going to bother implementing some of the stupid shit coming out of the 
W3C (CSS2 + CSS3).

I think a few things need to happen:

1) URL loading gets completely separated.  This is mostly done.  It could
   do with more of a rewrite to be even more asynch though.
2) XML parser as a c-level plugin or using the current external program +
   xml.el that I have attached.
3) A semi-minimal CSS2 implementation that would be enough to render simple 
   XML documents and tables correctly.
4) Perhaps use DSSSL in light of the stupid microsoft patent claiming to
   cover CSS.
5) More people than me and Thierry doing substantial work.

The current HTML parser and display engine could then suffer from bitrot
and not many people would care.

Between work, family, and trying to have more of a life, I just don't have
the time to devote to Emacs/W3 any more.  I'm going to talk to hrvoje,
steve, et al about some of this breakout and work while I am in japan.
I'll likely draft something that goes out to w3-beta and w3-dev as a plea
for help on the plan outlined above on the plane ride over.

If only I could convince Aventail to pay me to write elisp instead of
(shudder) C/C++.[1]

> Netscape is still used as the "stand-alone" browser, for reading
> /. and comics.

More sites tend more and more to be unusable without javascript and other
stupid shit.  Truly depressing.  I avoid the web like the plague it has
become. :)

-Bill P.

[1] I did at one point have a pretty kick-ass Emacs interface to our server
config file that we actually sent to some customers that wanted a TTY
version of the config tool. :)


[-- Attachment #2: sample expat-based xml->elisp parser --]
[-- Type: text/plain, Size: 4349 bytes --]

/*
The contents of this file are subject to the Mozilla Public License
Version 1.0 (the "License"); you may not use this file except in
compliance with the License. You may obtain a copy of the License at
http://www.mozilla.org/MPL/

Software distributed under the License is distributed on an "AS IS"
basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
License for the specific language governing rights and limitations
under the License.

The Original Code is expat.

The Initial Developer of the Original Code is James Clark.
Portions created by James Clark are Copyright (C) 1998
James Clark. All Rights Reserved.

Contributor(s):
*/

#include "xmlparse.h"

#include <stdio.h>
#include <stdlib.h>
#include <stddef.h>
#include <string.h>
#include <fcntl.h>

#ifdef _MSC_VER
#include <io.h>
#endif

#ifndef O_BINARY
#ifdef _O_BINARY
#define O_BINARY _O_BINARY
#else
#define O_BINARY 0
#endif
#endif

#ifdef _MSC_VER
#include <crtdbg.h>
#endif

#ifdef _DEBUG
#define READ_SIZE 16
#else
#define READ_SIZE (1024*8)
#endif

static void characterData(void *userData, const char *s, int len)
{
	FILE *fp = userData;
	putc('"',fp);
	for (; len > 0; --len, ++s) {
		putc(*s,fp);
	}
	putc('"',fp);
}

/* Lexicographically comparing UTF-8 encoded attribute values,
is equivalent to lexicographically comparing based on the character number. */

static int attcmp(const void *att1, const void *att2)
{
	return strcmp(*(const char **)att1, *(const char **)att2);
}

static void startElement(void *userData, const char *name, const char **atts)
{
	int nAtts;
	const char **p;
	FILE *fp = userData;
	putc('\n',fp);
	putc('(', fp);
	fputs(name, fp);
	putc('\n',fp);

	p = atts;
	while (*p)
		++p;
	nAtts = (p - atts) >> 1;
	if (nAtts > 1)
		qsort((void *)atts, nAtts, sizeof(char *) * 2, attcmp);

	putc('(',fp);
	putc('\n',fp);
	while (*atts) {
		putc('(', fp);
		fputs(*atts++, fp);		/* attribute */
		fputs(" . ",fp);
		characterData(userData, *atts, strlen(*atts));
		fputs(")\n",fp);
		atts++;
	}
	putc(')',fp);
}

static void endElement(void *userData, const char *name)
{
	FILE *fp = userData;
	fputs("\n); close tag ",fp);
	fputs(name, fp);
	putc('\n', fp);
}

static
void processingInstruction(void *userData, const char *target, const char *data)
{
  FILE *fp = userData;
  fprintf(fp, "<?%s %s?>", target, data);
}

static
void reportError(XML_Parser parser, const char *filename)
{
	int code = XML_GetErrorCode(parser);
	const char *message = XML_ErrorString(code);
	if (message)
		fprintf(stdout, "%s:%d:%ld: %s\n",
				filename,
				XML_GetErrorLineNumber(parser),
				XML_GetErrorColumnNumber(parser),
				message);
	else
		fprintf(stderr, "%s: (unknown message %d)\n", filename, code);
}

static
int processStream(const char *filename, XML_Parser parser)
{
	int fd;

	if (!strcmp(filename,"-"))
		fd = fileno(stdin);
	else
		fd = open(filename, O_BINARY|O_RDONLY);
	if (fd < 0) {
		perror(filename);
		return 0;
	}
	for (;;) {
		int nread;
		char *buf = XML_GetBuffer(parser, READ_SIZE);
		if (!buf) {
			close(fd);
			fprintf(stderr, "%s: out of memory\n", filename);
			return 0;
		}
		nread = read(fd, buf, READ_SIZE);
		if (nread < 0) {
			perror(filename);
			close(fd);
			return 0;
		}
		if (!XML_ParseBuffer(parser, nread, nread == 0)) {
			reportError(parser, filename);
			close(fd);
			return 0;
		}
		if (nread == 0) {
			close(fd);
			break;;
		}
	}
	return 1;
}

static
void usage(const char *prog)
{
	fprintf(stderr, "usage: %s [-e encoding] file ...\n", prog);
	exit(1);
}

int main(int argc, char **argv)
{
	int i;
	const char *encoding = NULL;

#ifdef _MSC_VER
	_CrtSetDbgFlag(_CRTDBG_ALLOC_MEM_DF|_CRTDBG_LEAK_CHECK_DF);
#endif

	if (argc < 2)
		usage(argv[0]);
	
	for (i = 1; i < argc; i++) {
		if ((argv[i][0] == '-') && argv[i][1]) {
			switch (argv[i][1]) {
			case 'e':
				encoding = argv[++i];
				break;
			default:
				usage(argv[0]);
				break;
			}
		} else {
			FILE *fp = NULL;
			char *outName = 0;
			int result;
			XML_Parser parser = XML_ParserCreate(encoding);
			fp = stdout;

			XML_SetUserData(parser, fp);
			XML_SetElementHandler(parser, startElement, endElement);
			XML_SetCharacterDataHandler(parser, characterData);
			XML_SetProcessingInstructionHandler(parser, processingInstruction);

			result = processStream(argv[i], parser);

			XML_ParserFree(parser);
		}
	}
	return 0;
}

Re: I wanna write an RFC

[Per Abrahamsen]

Harry Putnam <reader@newsguy.com> writes:

> But Per, how do you shut off that horrbile blockyness.  I run gnus
> in a dark color (grey20).  I'd rather not spend days slogging thru
> the style sheet, learning new syntax etc.  But would like to see web
> stuff in the same bg and fg as gnus.

Here are the settings needed to make Emacs/W3 useful:

 (setq w3-user-fonts-take-precedence t)  ; Use _my_ font.
 (setq w3-user-colors-take-precedence t) ; Use _my_ colors.
 (setq w3-honor-stylesheets nil)         ; No, just do it..
 (setq w3-use-terminal-characters nil)   ; No weird characters.
 (setq w3-horizontal-rule-char 45)       ; I said: no weird characters.

Re: I wanna write an RFC

[Norman Walsh]

| Seems cruel, but I can see both sides of the argument. Personally I like
| the way X-Face:s are included in the message since I don't really want
| Gnus to go and get an image off the web everytime I open a message, but
| I can also see how it would be nice in some cases.

I'd expect gnus to cache them locally, so you only had to go out and get
new ones.

                                        Cheers,
                                          norm
-- 
Norman Walsh <ndw@nwalsh.com>      | My problems start when the smarter
http://nwalsh.com/                 | bears and the dumber visitors
                                   | intersect.--Steve Thompson,
                                   | wildlife biologist at Yosemite
                                   | National Park

Re: I wanna write an RFC

[William M. Perry]

But these things can tend to change... you would need to make sure your cache
was not stale.  Either honoring the 'Expires' header in HTTP, or a 'keep for n
days minimum' kind of thing.

-bp

Norman Walsh wrote:

> | Seems cruel, but I can see both sides of the argument. Personally I like
> | the way X-Face:s are included in the message since I don't really want
> | Gnus to go and get an image off the web everytime I open a message, but
> | I can also see how it would be nice in some cases.
>
> I'd expect gnus to cache them locally, so you only had to go out and get
> new ones.
>
>                                         Cheers,
>                                           norm
> --
> Norman Walsh <ndw@nwalsh.com>      | My problems start when the smarter
> http://nwalsh.com/                 | bears and the dumber visitors
>                                    | intersect.--Steve Thompson,
>                                    | wildlife biologist at Yosemite
>                                    | National Park

Re: I wanna write an RFC

[Robert Bihlmeyer]

Hi,

>>>>> On Thu, 18 Mar 1999 04:03:37 -0500
>>>>> "William M. Perry" <wmperry@aventail.com> said:

 Bill> But these things can tend to change... you would need to make
 Bill> sure your cache was not stale.

Hmm, the X-Faces of people here on ding, haven't changed for aeons. It 
would be more than enough to check once a month.

No, no, stopit! Let's just put into the standard that you should not
reuse face URLs, i.e. use <http://.../bla-1.jpg> if bla.jpg is
outdated. Or put a version number, timestamp, whatever in front of the 
URL.

	Robbe

-- 
Robert Bihlmeyer	reads: Deutsch, English, MIME, Latin-1, NO SPAM!
<robbe@orcus.priv.at>	<http://stud2.tuwien.ac.at/~e9426626/sig.html>

Re: I wanna write an RFC

[Hrvoje Niksic]

Robert Bihlmeyer <e9426626@stud2.tuwien.ac.at> writes:

> Hi,
> 
> >>>>> On Thu, 18 Mar 1999 04:03:37 -0500
> >>>>> "William M. Perry" <wmperry@aventail.com> said:
> 
>  Bill> But these things can tend to change... you would need to make
>  Bill> sure your cache was not stale.
> 
> Hmm, the X-Faces of people here on ding, haven't changed for aeons. It 
> would be more than enough to check once a month.

Whatever.  A friend of mine has five or ten X-Faces, which he changes
according to the nature of the message he's writing.

Re: I wanna write an RFC

[David Kågedal]

Hrvoje Niksic <hniksic@srce.hr> writes:

> Robert Bihlmeyer <e9426626@stud2.tuwien.ac.at> writes:
> 
> > Hi,
> > 
> > >>>>> On Thu, 18 Mar 1999 04:03:37 -0500
> > >>>>> "William M. Perry" <wmperry@aventail.com> said:
> > 
> >  Bill> But these things can tend to change... you would need to make
> >  Bill> sure your cache was not stale.
> > 
> > Hmm, the X-Faces of people here on ding, haven't changed for aeons. It 
> > would be more than enough to check once a month.
> 
> Whatever.  A friend of mine has five or ten X-Faces, which he changes
> according to the nature of the message he's writing.

So if we changed to use URLs instead he would alternate between
different URLs, which would not cause any caching problems.

-- 
David Kågedal        <davidk@lysator.liu.se> http://www.lysator.liu.se/~davidk/

Re: I wanna write an RFC

[Hrvoje Niksic]

davidk@lysator.liu.se (David Kĺgedal) writes:

> > Whatever.  A friend of mine has five or ten X-Faces, which he
> > changes according to the nature of the message he's writing.
> 
> So if we changed to use URLs instead he would alternate between
> different URLs, which would not cause any caching problems.

No, he would likely still use the X-Faces, as would I.  :-)

(And I think I'd hate my mail reader to fetch things from the web when 
I'm reading mail, but I guess it's a matter of taste.)

Re: I wanna write an RFC

[Didier Verna]

Hrvoje Niksic writes:

Hrvoje> (And I think I'd hate my mail reader to fetch things from the web when
Hrvoje> I'm reading mail, but I guess it's a matter of taste.)

        It really depends on the connection you have. For instance, I retrieve
and display picons from the network. Things usually go rather fast for me.
Only sometimes. around 17h00, the connection might slow down and it becomes
annoying, so I eventually turn off the picons. Another thing to consider is
that my XEmacs sessions usually lasts for several weeks, so I end up having
many of the picons in the cache (and no more swap ;-).

-- 
    /     /   _   _       Didier Verna        http://www.inf.enst.fr/~verna/
 - / / - / / /_/ /      E.N.S.T. INF C201.1      mailto:verna@inf.enst.fr
/_/ / /_/ / /__ /        46 rue Barrault        Tel.   (33) 01 45 81 73 46
                      75634 Paris  cedex 13     Fax.   (33) 01 45 81 31 19

Re: I wanna write an RFC

[Harry Putnam]

wmperry@aventail.com (William M. Perry) writes:


> (setq w3-user-colors-take-precedence t)
> 
> -Bill P.
> 

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> 
> Here are the settings needed to make Emacs/W3 useful:
> 
>  (setq w3-user-fonts-take-precedence t)  ; Use _my_ font.
>  (setq w3-user-colors-take-precedence t) ; Use _my_ colors.
>  (setq w3-honor-stylesheets nil)         ; No, just do it..
>  (setq w3-use-terminal-characters nil)   ; No weird characters.
>  (setq w3-horizontal-rule-char 45)       ; I said: no weird characters.

Oh boy! Now we're talking ... Thanks fellas

-- 
Harry Putnam reader@newsguy.com  
Running Redhat Linux-5.2
See http://www.jtan.com/~reader  for a brief pictorial 
saga of construction work in the trade of "Boilermaker"

Re: I wanna write an RFC

[Kim-Minh Kaplan]

Didier Verna writes:

> I retrieve and display picons from the network. Things usually go
> rather fast for me.  Only sometimes. around 17h00, the connection
> might slow down and it becomes annoying

Mmm, that's strange: picons are supposed to be fetched as a background 
job.  It should not interfer with your ability to read messages.  Or
is it just the picons that appear while you are reading that annoy you ?


Kim-Minh.
.

Re: I wanna write an RFC

[Zlatko Calusic]

Hrvoje Niksic <hniksic@srce.hr> writes:

> davidk@lysator.liu.se (David Kĺgedal) writes:
> 
> > > Whatever.  A friend of mine has five or ten X-Faces, which he
> > > changes according to the nature of the message he's writing.
> > 
> > So if we changed to use URLs instead he would alternate between
> > different URLs, which would not cause any caching problems.
> 
> No, he would likely still use the X-Faces, as would I.  :-)
> 
> (And I think I'd hate my mail reader to fetch things from the web when 
> I'm reading mail, but I guess it's a matter of taste.)
> 

Yes, matter of taste definitely, but I really can't recall anybody who
would be willing to wait indefinite time just to get small b/w
picture, before reading his mail. And I'm speaking of people who
actually like x-faces, and use them (like me).

I wait for web pages, I wait for articles from the slow news server, I
wait for files to come when FTP-ing, but I DON'T want to wait to read
my email comfortably. Email is asynchronous in its nature, and I
believe it should stay that way.

So, I wan't to see those entertaining small pictures, and I want to
see them *right away*! :)
Few lines in (not) every article is small price to pay for that.

-- 
Zlatko

Re: I wanna write an RFC

[Didier Verna]

Kim-Minh Kaplan writes:

Kim-Minh> Didier Verna writes:

>> I retrieve and display picons from the network. Things usually go rather
>> fast for me. Only sometimes. around 17h00, the connection might slow down
>> and it becomes annoying

Kim-Minh> Mmm, that's strange: picons are supposed to be fetched as a
Kim-Minh> background job. It should not interfer with your ability to read
Kim-Minh> messages. Or is it just the picons that appear while you are reading
Kim-Minh> that annoy you ?

        Picons appearing while I'm reading is sometimes annoying because it
changes the buffer layout. But I think there are still bugs in the retrieving
routines. Sometimes for instance, gnus seems to wait for the first picon to be
retrieved, and then subsequent fetches are done in the background. I'm not
sure though, these days, I don't have time to investigate when the problem
occurs.

-- 
    /     /   _   _       Didier Verna        http://www.inf.enst.fr/~verna/
 - / / - / / /_/ /      E.N.S.T. INF C201.1      mailto:verna@inf.enst.fr
/_/ / /_/ / /__ /        46 rue Barrault        Tel.   (33) 01 45 81 73 46
                      75634 Paris  cedex 13     Fax.   (33) 01 45 81 31 19

Re: I wanna write an RFC

[Gerald Preissler]

Robert Bihlmeyer <e9426626@stud2.tuwien.ac.at> writes:

> No, no, stopit! Let's just put into the standard that you should not
> reuse face URLs, i.e. use <http://.../bla-1.jpg> if bla.jpg is
> outdated. Or put a version number, timestamp, whatever in front of the 
> URL.

Why not include some kind of hash-code of the actual pic in the
header. That way you can check that the pic you have is not stale.

just my DM 0.02

                Jerry

-- 
     |||       ... and now for something completely different ...
     o o
 --m--!--m--

Re: I wanna write an RFC

[Kim-Minh Kaplan]

Didier Verna writes:

> Sometimes for instance, gnus seems to wait for the first picon to be
> retrieved, and then subsequent fetches are done in the background.

Mmm, may be putting a '(sit-for 0)' in `gnus-picons-url-retrieve'
before the call to `url-retrieve' help aleviate this problem ?

Kim-Minh.
.

Re: I wanna write an RFC

[Lars Magne Ingebrigtsen]

davidk@lysator.liu.se (David Kågedal) writes:

> However, if you want to write another implementation you'll have to
> reimplement all the bugs.

Oh, wow.  Huh.  Hm.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: I wanna write an RFC

[Lars Magne Ingebrigtsen]

Norman Walsh <ndw@nwalsh.com> writes:

> | Seems cruel, but I can see both sides of the argument. Personally I like
> | the way X-Face:s are included in the message since I don't really want
> | Gnus to go and get an image off the web everytime I open a message, but
> | I can also see how it would be nice in some cases.
> 
> I'd expect gnus to cache them locally, so you only had to go out and get
> new ones.

Anyway, since if an RFC is written, the name of the header would
change from X-Face to Face, perhaps the format could be extended
beyond just having an X-Face Face in there.  Er.  Uhm.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: I wanna write an RFC

[François Pinard]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Anyway, since if an RFC is written, the name of the header would change
> from X-Face to Face, perhaps the format could be extended beyond just
> having an X-Face Face in there.  Er.  Uhm.

If a revision is going to happen in that area, could we also have `sound'
faces?  Maybe X-Hello or Hello?  So, when receiving a messages from Lars,
one could hear his voice saying "Lars Mgng'brtsn' here" :-).

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: I wanna write an RFC

[Lars Magne Ingebrigtsen]

François Pinard <pinard@iro.umontreal.ca> writes:

> If a revision is going to happen in that area, could we also have `sound'
> faces?  Maybe X-Hello or Hello?  So, when receiving a messages from Lars,
> one could hear his voice saying "Lars Mgng'brtsn' here" :-).

I don't think the world is quite ready for that one.  :-)

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: I wanna write an RFC

[François Pinard]

Lars Magne Ingebrigtsen <larsi@gnus.org> écrit:

> François Pinard <pinard@iro.umontreal.ca> writes:

> > If a revision is going to happen in that area, could we also have `sound'
> > faces?  Maybe X-Hello or Hello?  So, when receiving a messages from Lars,
> > one could hear his voice saying "Lars Mgng'brtsn' here" :-).

> I don't think the world is quite ready for that one.  :-)

You do not feel like pushing on the world a bit?  Writing an RFC might be
a proper way to define how to do a new thing which many people should do at
once for inter-operability, and I guess I would like that the world become
as ready for little greeting sounds than it is for little greeting faces.

Since X-Face and X-Hello might well use similar encoding and presentation
in message headers, it looks to me like a unique opportunity to address
both problems at once, which we should try to not miss if it can be done...

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard