Talk: Should I html'ize the FAQ?

   Many people, including my sysop, have asked if I'm going to make the 
FAQ an actual web document, not just a text file the web can access, 
display and download.

   Well, turning the FAQ into an actual web doc is not that hard, but it 
would be time consuming.  And the FAQ changes, more or less, each week.

   I've had people tell me that they'd be happy to html'ize it (html 
being the name of the command tags used in web docs) until I mentioned 
that the FAQ changes all the time.

   So, would you like to see a special version of the FAQ specifically 
made for the web?

--
         .:ooooo.
      .d88888888"
     .88P"   *"
     '88b:.        ..     ..     ..    ..     ..    ..     ..    ..  ..   ...
       '""*88b. .d888  .d888b ,Bo88r .8PY8 .o888 ,Bo88r ,d888b ;8@8 o8  o8P"
            X88.8P'*'.d8".q8" 8")8" ,8be* ,8P'*' 8"X8" ,88"*@8"88" 88' d8P
     .oooood88"!8. .o88 x*8".Z" db,dX8'  ,X8.  ,Z" db z#8. .8P 8bd88bd88P 
  .d8888888#"  '*8P" "8P" YwP"  *8" '*88#' '888P"  *8"  *88*'  #8P  *8P

In article <33tklt$i5...@gagme.wwa.com>, Bob Allison <bob...@wwa.com> wrote:
>   So, would you like to see a special version of the FAQ specifically
>made for the web?

Here's what I do:

Maintain the HTML version *only*. When it is time to post the
plain text version, do:

lynx -dump faq.html > faq.txt

Then post faq.txt. Myself, I have an automated utility that does
this for the Star Trek FAQs I maintain, but it really is that
simple. Lynx treats ASCII art inside <PRE></PRE> blocks kindly,
so it should work well for the ASCII art FAQs - I have an ASCII
graph in one of my FAQs.

However!

Be sure to go through the faq.txt file before posting; Lynx does
a pretty good job of rendering things, but if you have some
(mostly illegal) HTML combinations, it can look really gross. But
if the FAQ looks good with Lynx, the -dumped version should look
good as well. Once you've got the bugs out, the process is just
mechanical.

Joshua
--
"There's far too much to take in here, | MIME: jsb...@acs.ucalgary.ca
 more to find than can ever be found." | WWW:  http://www.ucalgary.ca/~jsbell
      - Tim Rice, "The Circle of Life" |

Joshua Bell <jsb...@acs.ucalgary.ca> wrote:
>
>In article <33tklt$i5...@gagme.wwa.com>, Bob Allison <bob...@wwa.com> wrote:
>>   So, would you like to see a special version of the FAQ specifically
>>made for the web?
>
>Here's what I do:
>
>Maintain the HTML version *only*. When it is time to post the
>plain text version, do:
>
>lynx -dump faq.html > faq.txt
>
>Then post faq.txt. Myself, I have an automated utility that does
>this for the Star Trek FAQs I maintain, but it really is that
>simple. Lynx treats ASCII art inside <PRE></PRE> blocks kindly,
>so it should work well for the ASCII art FAQs - I have an ASCII
>graph in one of my FAQs.
>
>However!
>
>Be sure to go through the faq.txt file before posting; Lynx does
>a pretty good job of rendering things, but if you have some
>(mostly illegal) HTML combinations, it can look really gross. But
>if the FAQ looks good with Lynx, the -dumped version should look
>good as well. Once you've got the bugs out, the process is just
>mechanical.


   The only problem is that I would probably HTML'ize the FAQ portion (or
maybe all of the) ASCII Art Resources file.  This file has a lot of ASCII
art included, some of which is wrapped with text.

   This type of file needs to be HTML'ized after it is changed.  It would
be very hard to change it once it has embedded HTML tags.  I suppose I
could take the previous week's text version, make the changes I want, and
then transfer those changes to the HTML version.

--

       |      |        |        |      |      |        |        |    |   |   |
    ___|    __|    _   |    _   |    __|    __|    _   |    _   |    |   |   |
   |       |      |    |   |    |   |      |      |    |   |    |    |   |   |
       |   |           |        |      |   |           |   |    |    |   |   |
____   |   |       _   |       _|    __|   |          _|   |    |    |   |   |
       |   |      |    |   |        |      |      |        |    |    |   |   |
       |      |   |    |   |    |      |      |   |    |        |            |
_______| _____|___| ___|___| ___|______| _____|___| ___| _______| ___________|

In article <340ai5$9k...@gagme.wwa.com>, Bob Allison <bob...@wwa.com> wrote:
>
>
>   The only problem is that I would probably HTML'ize the FAQ portion (or
>maybe all of the) ASCII Art Resources file.  This file has a lot of ASCII
>art included, some of which is wrapped with text.

>   This type of file needs to be HTML'ized after it is changed.  It would
>be very hard to change it once it has embedded HTML tags.

Why? The HTML file itself can look like shit and be a total mess,
since HTML is a *markup language*. As long as the ASCII stuff
looks okay and is wrapped in <PRE></PRE> the results of the
"lynx -dump" will look fine.

Try it. The nice thing about HTML (and *any* markup language,
like LaTeX) is that the source file doesn't have to be
pretty. For example:

......................................................................
<HTML>
<HEAD>
<TITLE>Example HTML Document</TITLE>
</HEAD>
<BODY>

<H1>
This is a sample
of
an
<a href="http://foo.bar.com/~bubba">
ASCII
Art
FAQ</a>
!!!!!</H1>

<dl>
<dt>Line<dd>Using mostly
<code>_-~|/\()&gt;&lt;.,'`"O</code>
to draw things.
<dt>
Greyscale
<dd>
Using things like
<code>.,"-oad*8M$&amp;@#</code> to represent pixels
</dl>

<!-- The above is the index! Really! -->

<hr>

By Joshua Bell, as an example of poor internal layout for an HTML file. This line, for example, is much wider than 80 columns, but it doesn't really matter.
<p>
<address> Joshua Bell
&lt;jsb...@acs.ucalgary.ca&gt; </address>

<BODY>
</HTML>
......................................................................

That, passed through lynx -dump, yields:

......................................................................

                 THIS IS A SAMPLE OF AN ASCII ART FAQ !!!!!

   Line
          Using mostly _-~|/\()><.,'`"O to draw things.

   Greyscale
          Using things like .,"-oad*8M$&@# to represent pixels


     _________________________________________________________________

   By Joshua Bell, as an example of poor internal layout for an HTML
   file. This line, for example, is much wider than 80 columns, but it
   doesn't really matter.

    Joshua Bell <jsb...@acs.ucalgary.ca>
......................................................................

Now usually the HTML file isn't that bad, and if you use
something like Emacs html-helper-mode (see
http://www.reed.edu/~nelson/tools/) it'll help format the HTML
nicely for you. But you don't *have* to keep it looking pretty
inside the HTML file, except for your own convenience. No-one
sees it unless they specifically ask for it, and they should be
expecting a mess.

If you want to add a new paragraph somewhere, just rememeber to
add in the <p> tags in the appropriate places (or <p>...</p> if
you want to be HTML+ compliant).

Maintaining HTML documents is MUCH simpler than maintaining plain
text documents; no worries about formatting! No need to juggle
word wrapping! The drudgery of adding in new hyperlinks is about
the only downer.

Joshua

Joshua Bell (jsb...@acs.ucalgary.ca) beweerde hier:
>In article <33tklt$i5...@gagme.wwa.com>, Bob Allison <bob...@wwa.com> wrote:
>>   So, would you like to see a special version of the FAQ specifically 
>>made for the web?

>Here's what I do:

>Maintain the HTML version *only*. When it is time to post the
>plain text version, do:

>lynx -dump faq.html > faq.txt

I do the same for the European Union Basics (FAQ) list, but I use the CERN
LineMode Browser for my screendumps: it renders my HTMLplus documents
better (although this may be a question of taste) but more importantly: it
keeps the links by inserting [1] type footnote numbers in the text and
giving a list of URLs at the end of the document. This is done with the
command:

cat faq.html | www -listrefs > faq.txt

You may see a demo of the CERN LineModeBrowser by telnet to info.cern.ch
(no login).
--
Roland Siebelink <rcs...@vub.ac.be> -- <rcs...@is3e.vub.ac.be> for finger

         + Democracy arises out of the renunciation of certainty,
       or at least the renunciation that one man (m/f) could impose
       his certainty on another man. ; -- adapted from The Economist

Joshua Bell (jsb...@acs.ucalgary.ca) wrote:

: Maintaining HTML documents is MUCH simpler than maintaining plain
: text documents; no worries about formatting! No need to juggle
: word wrapping! The drudgery of adding in new hyperlinks is about
: the only downer.

Yeah, but in your "awful html example" you conveniently "hid" the
&lt; &gt; escapes needed to pass "<" and ">" through the html parser,
yes, even in <pre></pre> structures.

It's no big deal usually, but in an ASCII ART FAQ, at least half the
pictures will contain a "<" or ">", sometimes in large quantities.  It
sounds like too much bother.

Now if you were to convert the art into GIFs, and inline them into the
html faq--THAT would be something to behold.  :)

--
/'''  Collin Forbes
c oo  col...@uidaho.edu 
`` > 
 `'`  http://gopher.uidaho.edu/1/users/collinf/

Collin Forbes <col...@crow.csrv.uidaho.edu> wrote:
>
>Joshua Bell (jsb...@acs.ucalgary.ca) wrote:
>
>: Maintaining HTML documents is MUCH simpler than maintaining plain
>: text documents; no worries about formatting! No need to juggle
>: word wrapping! The drudgery of adding in new hyperlinks is about
>: the only downer.
>
>Yeah, but in your "awful html example" you conveniently "hid" the
>&lt; &gt; escapes needed to pass "<" and ">" through the html parser,
>yes, even in <pre></pre> structures.
>
>It's no big deal usually, but in an ASCII ART FAQ, at least half the
>pictures will contain a "<" or ">", sometimes in large quantities.  It
>sounds like too much bother.

   This could be worked around by doing a search and replace with either
( and ) or with the escape codes.  If escape codes are put in, it makes
dealing with the art more difficult.

>Now if you were to convert the art into GIFs, and inline them into the
>html faq--THAT would be something to behold.  :)

   The big problem, as I've mentioned before, is that I would probably 
use the FAQ portion of the ASCII Art Resources file.  It has a lot of 
ASCII art in it.  And some of that art is wrapped with text, like:

   The first four use the standard printable set, and can be viewed anywhere,
anytime, on any equipment.  If you can view text, you can view art.  They 
are:
 
   o  Line drawing - Such as the 3-D snowman         ..:::::::::..
      below.  Line art is made using              ..:::::::::::::::..
      characters for their shapes.              ..:::::: ====== :::::..
                                               ..::::: ========== ::::..
      In the case of the candle at right,     ..::::: ==== |\ ==== ::::..
      characters were chosen for their        .::::: ==== /%#| ==== :::..
      shape AND brightness.                   .::::: === |&@/ ==== :::::.
                                              ..::::: ===_\|_==== :::::..
      For how to make ASCII art, see           ..::::: /| .:. | :::::...
      Ouestion 9 below.                         ..:::: || .:. | ::::...
                                                 ...:: (| .:. | ::....
      Candle Line Drawing by The Scarecrow.        .... | .:. | ....
 
   o  Lettering - Large and styled like the title "ASCII ART RESOURCES" above,
      ond like the lettering below.  It can be made by hand, or it can be
                                                               generated by
       _    ___  _____  _____  ___   ___   _____  _  _   ___   using special
      | |  | __!|_   _||_   _|| __! | _ ) |_   _|| \| | / __!  software.  For
      | |_ | __!  | |    | |  | __! |   \  _| |_ |    || \_ \  more info on
      !___!!___!  !_!    !_!  !___! !_!\_\!_____!!_!\_| \___/  auto-lettering
                                                               programs for
      use on some systems, see Question 13.  This is a 'Geometric Article'.
      It is used as a wrap-around for the hand drawn word "LETTERING" above.
 
   o  Gray Scale picture - These                  :<!!:
      create the illusion of gray shades     ..:::tNi@Nin!+: :::xr
      by using letters for their light     '!)Q88i@8888NNidbo@Ni8ir
      emitting value (assuming you are    .s:S988TR88RX*###RB8888i:x
      viewing light letters on a dark    x+RRX88888NL#?!< :<?#RR888bX
      background).  Below is a chart    !MN*#M#*BRR#8iXxi@NUX!MN88888:
      showing the light values of     <x@T!M!::(R8H@888888888xX88888#=
      various characters.  To the     x8W8Ui@!~!XMMM9M8RRRRBMMM!T588.
      right is an example of gray     .N8888M<<<~!#MXMHM8N888RMR<9R888N!.
      scale art.  This is an image    XQ888M!<:::sxXU!?MTT222#MM<?M8888~!
      of Einstein I made from a       988NMMf<!!~`"!Rk:X!"""#*N!:~!T588 !
      GIF, using Gifscii for the      "R(#MF `<      "!!~     ~<9XN#88XX<:
      Mac.  There are some emakples    "%/9" x~       :L       XH?R?M888x%
      at the end of this file.          xi/ ~M:u::~  <88>x<:<uuUF<~*"<?8!~
      ________________________         X38M::#M88R::d<88k9bxH888f4r o<<Ri
     |                        |       '!88N!!X6?M&i8#<88Rt88NMR9%!!'8:<?8!
     |  Lighter   *WMB8&%$#@  |         #R?<:~B:M88~ 488H~"88XM8~ ~HRtxH#
     |    /\      oahkbdpqwm  |      +++x8!:<'8!@8!.  ""!:.#8M8# ~<!@!M?::
     |   /||\     LCJUYXZO0Q  |      -<!#\x~:.?<M~ <<::;;>> ?%8xx!:i:`MHbLX
     |    ||      rcvunxzjft  |       :*:!!:nM.~!~`<<! <`~" d<?98NHRNi!?888eu.
     |   \||/     /\|()1{}[]  |       ` `xH98MN.      '~~:. ?:x888!888xd88888f
     |    \/      -_+<>i!lI?  |        :@8R?R?88ex  .boba94  WM888M88(W888888f
     |  Darker    .'`,^:";~   |      .oX"#MMX!!R88b ~"##**".d8f?RRX8PW8888888f
     !________________________!    .u8888iXMMnWU7T#@s.     @88kX888RN88888888f
       Scale from Jorn Barger.    .@888888NK9*MR88N86Q:.. x8T888R#88888888888f
                                 x8888888888@iX#M@8888888N88888#X@88888888888f
      Einstein GIF conversion   d88888888888kM888NN868RRRRRRR5b@8888888888888f
          by The Scarecrow.    '8888888888888!M888888888888888888888888888888!
 
   o  3-D image - They can be viewed by people with similar vision in both
      eyes.  You try to focus as if you are looking at the back of the
      monitor.  The image should pop into focus and create a 3-D illusion.
      Other 3-D images are viewed by putting your nose on the monitor glass.
 
       .-~~\           ::::::    .-~~\        ::::::   For automated help in
      /     \  _       |::::|   /     \ _     |::::|   making 3-D images,
      ~x    .-~_)_     l~~~~!   ~x   .-~_)_   l~~~~!   you can use a program
        ~>x".-~   ~-.   \RC/      ~x".-~   ~-. \RC/    called 'sirdvu11'.
     _   ( /         \   ||    _  ( /         \ ||     For red-blue 3-D,
     ||   T  o  o     Y  ||    ||  T o  o      Y||     you can use DTA
   ==:l   l   <       !  (3  ==:l  l  <        !(3     (Dave's Targa
      \\   \  .__/   /  /||     \\  \  ._/    / ||     Animation), and for
       \\ ,r"-,___.-'r.//||      \\,r"-,___.-'r/||     stereograms you need
        }^ \.( )   _.'//.||      }^\. ( )  _.-//||     a program called STAREO.
       /    }~Xi--~  //  ||     /   }~Xi--~  // ||\    And there's also
      Y    Y I\ \    "   ||    Y   Y I\ \    "  || Y   "mk3d.lha" for the
      |    | |o\ \       ||    |   | |o\ \      || |   Amiga.  For more
      |    l_l  Y T      ||    |   l_l  Y T     || |   information, see
      l      "o l_j      |!    l     "o l_j     || !   the alt.3d group.
       \                 ||     \               ||/
     .--^.     o  -Row .^||.  .--^.     o -Row  ||--.  3-D snowman by
          "           ~  `'        "           ~`'     Rowan Crawford.


   Once the art-text is html'ized, it's real hard to work with, unless 
you strip the html out of it.  The html code throws offs the alignment of 
the art-text when it's not being viewed in a web browser.

   So when you are working on a portion of art-text that contains html 
code, the code has to be stripped, the art-text modified, and the code 
put back in.

--
.eeeeee..eeeee..eeeeee..eeeeeee..eeeeee..eeeee.eeeeeee...eeeeee..eee..eee..eee
@@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@::@@@::@@@
%%%-----%%%----%%%--%%%-%%%--%%%-%%%----%%%----%%%--%%%-%%%--%%%-%%%--%%%--%%%
&&&&&&++&&&++++&&&&&&&&+&&&&&&&++&&&&&++&&&++++&&&&&&&++&&&++&&&+&&&++&&&++&&&
*||||||*|||****||||||||*||||||***|||||**|||****||||||***|||**|||*|||**|||**|||
====!!!=!!!====!!!==!!!=!!!=!!!==!!!====!!!====!!!=!!!==!!!==!!!=!!!==!!!==!!!
:::::::#::::::#:::##:::#:::##:::#::::::#::::::#:::##:::#::::::::#:::::::::::::
......@@@.....@...@@...@...@@...@......@@.....@...@@...@@......@@@...........@

i distinctly remember Bob Allison (bob...@wwa.com) saying...
:=    Once the art-text is html'ized, it's real hard to work with, unless 
:= you strip the html out of it.  The html code throws offs the alignment of 
:= the art-text when it's not being viewed in a web browser.

you  needn't  make a  separate  html file  for the
faq... you could always   just send it   through a
perl script  to make the  translations.  (nice and
easy: everything  into  &entities  and then  parse
"&lt;a href" back into "<a   href", et cetera.   i
implement this on a few pages of my own.)

but personally, i like the FAQ just the way it is.

--
									<sigh>
		     my handwriting's pretty bad... sorry about the signature.
		                  <a href="http://acs2.bu.edu:3000/"> . . </a>

In article <34af74$5n...@gagme.wwa.com>, Bob Allison <bob...@wwa.com> wrote:
>
>Collin Forbes <col...@crow.csrv.uidaho.edu> wrote:
>>
>>Joshua Bell (jsb...@acs.ucalgary.ca) wrote:
>>
>>: Maintaining HTML documents is MUCH simpler than maintaining plain
>>: text documents; no worries about formatting! No need to juggle
>>: word wrapping! The drudgery of adding in new hyperlinks is about
>>: the only downer.
>>
>>Yeah, but in your "awful html example" you conveniently "hid" the
>>&lt; &gt; escapes needed to pass "<" and ">" through the html parser,
>>yes, even in <pre></pre> structures.
>>
>>It's no big deal usually, but in an ASCII ART FAQ, at least half the
>>pictures will contain a "<" or ">", sometimes in large quantities.  It
>>sounds like too much bother.

If your editor doesn't have search & replace, and you aren't
using an operating system with commands that can do that to a
text file, you're seriously screwed anyway.

>   This could be worked around by doing a search and replace with either
>( and ) or with the escape codes.  If escape codes are put in, it makes
>dealing with the art more difficult.

They're called "elements", actually, but anyway...

>>Now if you were to convert the art into GIFs, and inline them into the
>>html faq--THAT would be something to behold.  :)

Aieee! :)

>   The big problem, as I've mentioned before, is that I would probably 
>use the FAQ portion of the ASCII Art Resources file.  It has a lot of 
>ASCII art in it.  And some of that art is wrapped with text, like:

[snip]

>   o  Line drawing - Such as the 3-D snowman         ..:::::::::..
>      below.  Line art is made using              ..:::::::::::::::..
>      characters for their shapes.              ..:::::: ====== :::::..

Okay, okay, that's going to be pointless to convert to HTML. Not
due to any failings in HTML itself, but due to the conceptual
problems; you've already merged two types of information into one
(text and ascii-art into one plain text document). The equivalent
would be to take a magazine layout page, save it as a GIF, and
then complain when converting it to HTML would be a bother.

The merged format may be pretty, but it isn't really useful, if
someone wants to extract text or artwork from the FAQ.

>   Once the art-text is html'ized, it's real hard to work with, unless 
>you strip the html out of it.  The html code throws offs the alignment of 
>the art-text when it's not being viewed in a web browser.

>   So when you are working on a portion of art-text that contains html 
>code, the code has to be stripped, the art-text modified, and the code 
>put back in.

Alternatively, you could reformat it like this:


o Another popular form of ASCII art is the signature. Here is a
signature which uses greyscale lettering:

.eeeeee..eeeee..eeeeee..eeeeeee..eeeeee..eeeee.eeeeeee...eeeeee..eee..eee..eee
@@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@::@@@::@@@
%%%-----%%%----%%%--%%%-%%%--%%%-%%%----%%%----%%%--%%%-%%%--%%%-%%%--%%%--%%%
&&&&&&++&&&++++&&&&&&&&+&&&&&&&++&&&&&++&&&++++&&&&&&&++&&&++&&&+&&&++&&&++&&&
*||||||*|||****||||||||*||||||***|||||**|||****||||||***|||**|||*|||**|||**|||
====!!!=!!!====!!!==!!!=!!!=!!!==!!!====!!!====!!!=!!!==!!!==!!!=!!!==!!!==!!!
:::::::#::::::#:::##:::#:::##:::#::::::#::::::#:::##:::#::::::::#:::::::::::::
......@@@.....@...@@...@...@@...@......@@.....@...@@...@@......@@@...........@


This lends itself more easily to HTML conversion:

<p>
<a name="signatures"><img src="icons/bullet.gif" alt="o"></a>
Another popular form of ASCII art is the signature. Here is a
signature which uses <a href="#greyscale">greyscale</a> 
<a href="#lettering">lettering</a>: 
</p>
<pre>
.eeeeee..eeeee..eeeeee..eeeeeee..eeeeee..eeeee.eeeeeee...eeeeee..eee..eee..eee
@@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@::@@@::@@@
%%%-----%%%----%%%--%%%-%%%--%%%-%%%----%%%----%%%--%%%-%%%--%%%-%%%--%%%--%%%
&&&&&&++&&&++++&&&&&&&&+&&&&&&&++&&&&&++&&&++++&&&&&&&++&&&++&&&+&&&++&&&++&&&
*||||||*|||****||||||||*||||||***|||||**|||****||||||***|||**|||*|||**|||**|||
====!!!=!!!====!!!==!!!=!!!=!!!==!!!====!!!====!!!=!!!==!!!==!!!=!!!==!!!==!!!
:::::::#::::::#:::##:::#:::##:::#::::::#::::::#:::##:::#::::::::#:::::::::::::
......@@@.....@...@@...@...@@...@......@@.....@...@@...@@......@@@...........@
</pre>

*shrug*

Again, the conversion to plain text is MECHANICAL, there is no
work involved, and no-one without a HTML viewer need look at the
HTML document: they can read the posted, fingered, or FTP'd
plaintext version you generate with a browser. And if you can't
handle using &lt; in place of <, a simple search/replace on
everything new that you add to the document, you have bigger
problems.

I'd prefer to browse through HTML documents with hyperlinks. When
it mentions Figlet, a click can take me to the interactive Figlet
page. When it mentions various archives, a click can take me
right there. Remember, you're trying to sell your services to the
user, in a sense. If it makes life easier for them, you win.

I guess it boils down to: do you want just "eye candy" documents
which have no extrinsic value, or a useful resource document
which might be a little less compact and less cute, but is
useful?

Of course, you can combine the two, and have it all
merged, <pre>-formatted, with anchors inside, but that'd be
really painful to edit.

Joshua
--
"There's far too much to take in here, | MIME: jsb...@acs.ucalgary.ca
 more to find than can ever be found." | WWW:  http://www.ucalgary.ca/~jsbell/
      - Tim Rice, "The Circle of Life" |

Joshua Bell <jsb...@acs.ucalgary.ca> wrote:
>
>In article <34af74$5n...@gagme.wwa.com>, Bob Allison <bob...@wwa.com> wrote:
>>
>>   The big problem, as I've mentioned before, is that I would probably 
>>use the FAQ portion of the ASCII Art Resources file.  It has a lot of 
>>ASCII art in it.  And some of that art is wrapped with text, like:
>
>[snip]
>
>>   o  Line drawing - Such as the 3-D snowman         ..:::::::::..
>>      below.  Line art is made using              ..:::::::::::::::..
>>      characters for their shapes.              ..:::::: ====== :::::..
>>
>>   Once the art-text is html'ized, it's real hard to work with, unless 
>>you strip the html out of it.  The html code throws offs the alignment of 
>>the art-text when it's not being viewed in a web browser.
>
>>   So when you are working on a portion of art-text that contains html 
>>code, the code has to be stripped, the art-text modified, and the code 
>>put back in.
>
>Alternatively, you could reformat it like this:
>
>
>o Another popular form of ASCII art is the signature. Here is a
>signature which uses greyscale lettering:
>
>.eeeeee..eeeee..eeeeee..eeeeeee..eeeeee..eeeee.eeeeeee...eeeeee..eee..eee..eee
>@@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@::@@@::@@@
>%%%-----%%%----%%%--%%%-%%%--%%%-%%%----%%%----%%%--%%%-%%%--%%%-%%%--%%%--%%%
>&&&&&&++&&&++++&&&&&&&&+&&&&&&&++&&&&&++&&&++++&&&&&&&++&&&++&&&+&&&++&&&++&&&
>*||||||*|||****||||||||*||||||***|||||**|||****||||||***|||**|||*|||**|||**|||
>====!!!=!!!====!!!==!!!=!!!=!!!==!!!====!!!====!!!=!!!==!!!==!!!=!!!==!!!==!!!
>:::::::#::::::#:::##:::#:::##:::#::::::#::::::#:::##:::#::::::::#:::::::::::::
>......@@@.....@...@@...@...@@...@......@@.....@...@@...@@......@@@...........@
>
>
>This lends itself more easily to HTML conversion:
>
><p>
><a name="signatures"><img src="icons/bullet.gif" alt="o"></a>
>Another popular form of ASCII art is the signature. Here is a
>signature which uses <a href="#greyscale">greyscale</a> 
><a href="#lettering">lettering</a>: 
></p>
><pre>
>.eeeeee..eeeee..eeeeee..eeeeeee..eeeeee..eeeee.eeeeeee...eeeeee..eee..eee..eee
>@@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@@@@:@@@@@@:@@@@@@@@:@@@@@@@@:@@@::@@@::@@@
>%%%-----%%%----%%%--%%%-%%%--%%%-%%%----%%%----%%%--%%%-%%%--%%%-%%%--%%%--%%%
>&&&&&&++&&&++++&&&&&&&&+&&&&&&&++&&&&&++&&&++++&&&&&&&++&&&++&&&+&&&++&&&++&&&
>*||||||*|||****||||||||*||||||***|||||**|||****||||||***|||**|||*|||**|||**|||
>====!!!=!!!====!!!==!!!=!!!=!!!==!!!====!!!====!!!=!!!==!!!==!!!=!!!==!!!==!!!
>:::::::#::::::#:::##:::#:::##:::#::::::#::::::#:::##:::#::::::::#:::::::::::::
>......@@@.....@...@@...@...@@...@......@@.....@...@@...@@......@@@...........@
></pre>
>
>*shrug*
>
>Again, the conversion to plain text is MECHANICAL, there is no
>work involved, and no-one without a HTML viewer need look at the
>HTML document: they can read the posted, fingered, or FTP'd
>plaintext version you generate with a browser. And if you can't
>handle using &lt; in place of <, a simple search/replace on
>everything new that you add to the document, you have bigger
>problems.

   Having everything laid out so that there's no wrapped text would work, 
but it's not as pleasing.

>I'd prefer to browse through HTML documents with hyperlinks. When
>it mentions Figlet, a click can take me to the interactive Figlet
>page. When it mentions various archives, a click can take me
>right there. Remember, you're trying to sell your services to the
>user, in a sense. If it makes life easier for them, you win.
>
>I guess it boils down to: do you want just "eye candy" documents
>which have no extrinsic value, or a useful resource document
>which might be a little less compact and less cute, but is
>useful?

   I want both.  I want eye candy with links to everything: internal and 
external, FTP, Gopher, and WWW sites.

>Of course, you can combine the two, and have it all
>merged, <pre>-formatted, with anchors inside, but that'd be
>really painful to edit.

   That's what I've been saying.  I guess I want somebody to tell me how 
that can be done.  I need a way to de-html it so I can work on the 
art-text, than re-html it for use.

   Oh well, I may html it as a one shot and see if people like it enough 
to figure out how to do it easily.  This discussion should have been sent 
to the www groups as well.

--
   _--_      ____                                                .
  ||          ||  ||_   _                                       /:\
  ||          ||  || | (/_                                     /;:.\
  ||__.        _                                              //;:. \
   ~  ||  /'\\< \,,._-_ /'\\ /'\\,._-_ /'\\ \\ \\ \\         ///;:.. \
      || ||   /-|| ||  || / ||    ||  || || || || ||   __--"////;:... \"--__
      || ||  (( || ||  ||/  ||    ||  || || || || || --__   "--_____--"   __--
 ,-___-  \\,/ \/\\ \\, \\,/ \\,/  \\, \\,/  \\//\//      """--_______--"""
 

Joshua Bell:
>I guess it boils down to: do you want just "eye candy" documents
>which have no extrinsic value, or a useful resource document
>which might be a little less compact and less cute, but is
>useful?

This is similar to the reason I hate the Motif look&feel.  It's very
heavy on eye candy that chews up screen real estate for no reason.

But I'd argue the layout of the ascii FAQ is not eye candy.  It's
information-dense and makes good use of visual space.  Perhaps the
cost of doing this is too high, but that's for Bob Allison to decide.
ascii art is very much concerned with visual layout, and it's quite
sensible to apply these techniques to the FAQ itself.

>The merged format may be pretty, but it isn't really useful, if
>someone wants to extract text or artwork from the FAQ.

It's only hard to extract text or artwork from the FAQ if you're stuck
with oppressively linear text editors.  Anyone interested in doing
ascii art is probably going to want an editor with 2-D text support
anyway.

>Okay, okay, that's going to be pointless to convert to HTML. Not
>due to any failings in HTML itself, but due to the conceptual
>problems; you've already merged two types of information into one
>(text and ascii-art into one plain text document). The equivalent
>would be to take a magazine layout page, save it as a GIF, and
>then complain when converting it to HTML would be a bother.

but it *is* a failing of HTML.  HTML is okay as a document-structure
language, but it isn't particularly great as a layout language.  It
isn't rich enough to describe non-trivial layouts in a sensible
fashion.  You can't even do simple things like two-column text.  So
why bother using HTML as a source language for the ascii FAQ?

now, going the other way may be plausible.  Or perhaps using a third
form that will generate both plain ascii and html...
--

+---jsb...@acs.ucalgary.ca (Joshua Bell)-writes---
| 
| Again, the conversion to plain text is MECHANICAL, there is no
| work involved, and no-one without a HTML viewer need look at the
| HTML document: they can read the posted, fingered, or FTP'd
| plaintext version you generate with a browser. And if you can't
| handle using &lt; in place of <, a simple search/replace on
| everything new that you add to the document, you have bigger
| problems.
| 
| I'd prefer to browse through HTML documents with hyperlinks. When
| it mentions Figlet, a click can take me to the interactive Figlet
| page. When it mentions various archives, a click can take me
| right there. Remember, you're trying to sell your services to the
| user, in a sense. If it makes life easier for them, you win.
| 
| I guess it boils down to: do you want just "eye candy" documents
| which have no extrinsic value, or a useful resource document

I want "eye candy" documents!!!! They may not have any extrinsic [sic] value,
but they are artistic!!!!

| which might be a little less compact and less cute, but is
| useful?

Well, most FAQs are so boring to read. I would prefer a FAQ which impresses
a newbie and make he/she interested in ascii art instead of one telling
him/her tons of useless "useful" informations.

---------------------------------------achille (eil...@drizzle.stanford.edu)
One who prefer the current form of the FAQ.