 | | | | | |
All,
I've just posted to lugnet.faq all 202 FAQ items I've collected (1). Just
look for messages with the subject line starting "autoFAQpost"; they should
be pretty hard to miss.
Warning: If you're subscribing to lugnet.faq via e-mail, you might want to
go get a beverage or three while they download. Don't worry; this is (very
likely) a one-time offer.
These messages are posted in the full syntactical glory, with headers and
all. Please feel free to reply to any of them with changes, or to add new
items!
Cheers,
- jsproat
1. Uh-oh, I've just reverse-spammed LUGNET...! :-,
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | |
In lugnet.faq, Jeremy H. Sproat writes:
> All,
>
> I've just posted to lugnet.faq all 202 FAQ items I've collected (1).
Wow! I'm impressed. Thank you. Now the FAQ newsgroup has FAQ's (-:
While I was waiting, I answered most of the MindStorms questions and posted
them here. Maybe some of the information will be new or useful. If not, well I
had fun doing it anyway. I even learned one or two things by figuring out
answers to questions I hadn't thought of before (-8
- Robert Munafo
| | | | | | | | | | | | | | | | |
In lugnet.faq, Robert Munafo writes:
> While I was waiting, I answered most of the MindStorms questions and posted
> them here. Maybe some of the information will be new or useful.
Do you intend to include this FAQ into the grand LUGNET FAQ? (1) If so,
they'd need to be put into the proper format. You could do it if you want, or
I can get to it when I get more time.
> I even learned one or two things by figuring out
> answers to questions I hadn't thought of before (-8
That's my motivation for doing a FAQ. :-,
Cheers,
- jsproat
1. I think I took a bunch of items from the older version.
| | | | | | | | | | | | | | | | | |
On Tue, 13 Jul 1999 03:20:55 GMT, "Sproaticus" <jsproat@io.com> wrote:
> In lugnet.faq, Robert Munafo writes:
> > While I was waiting, I answered most of the MindStorms questions and posted
> > them here. Maybe some of the information will be new or useful.
>
> Do you intend to include this FAQ into the grand LUGNET FAQ? (1) If so,
> they'd need to be put into the proper format. You could do it if you want, or
> I can get to it when I get more time.
Is there a FAQ for the FAQ format and posting properly to lugnet.faq?
My eyes sort of glazed over during the various discussions in this group...
Steve
| | | | | | | | | | | | | | | | | | | | | |
| |
|
Steve Bliss wrote:
> On Tue, 13 Jul 1999 03:20:55 GMT, "Sproaticus" <jsproat@io.com> wrote:
> > In lugnet.faq, Robert Munafo writes:
> > > While I was waiting, I answered most of the MindStorms questions and posted
> > > them here. Maybe some of the information will be new or useful.
> > Do you intend to include this FAQ into the grand LUGNET FAQ? (1) If so,
> > they'd need to be put into the proper format. You could do it if you want, or
> > I can get to it when I get more time.
> Is there a FAQ for the FAQ format and posting properly to lugnet.faq?
The latest (autoFAQpost) can be found at article:
http://www.lugnet.com/faq/?n=431
However, there were several threads discussing the continuing evolution the
the format; a good one to start with would be:
http://www.lugnet.com/faq/?n=180&t=i&v=a
> My eyes sort of glazed over during the various discussions in this group...
Have you tried more caffeine? No, wait -- I get *too* glazed after too much
caffeine. :-,
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | | |
In lugnet.faq, Jeremy H. Sproat writes:
> Do you intend to include this FAQ into the grand LUGNET FAQ? (1) If so,
> they'd need to be put into the proper format. You could do it if you want, or
> I can get to it when I get more time.
I'll look around and try to find out what the "proper" format is. Maybe you
could reply with a pointer to the LUGNET message I need to read.
However, I want to point out that text-only ASCII is the world standard format
for FAQ files and I'm reluctant to spend a lot of time and effort turning it
into some elaborate nonstandard format, particularly if it will make the result
impossible to convert back to plain ASCII. FAQ files are about information, not
style and they need to be easy to distribute across all operating systems by
many different protocols, including plain text email.
| | | | | | | | | | | | | | | | | | | | | |
Okay, I found it. It looks like I can handle it. I can cut and paste to make
the job easier. I have 202 examples to work from! (-;
Good luck to anyone of average intelligence who tries to figure it out, though.
Everyone on LUGNET must be at least MENSA level or higher. I had to read
through the "format" answer http://www.lugnet.com/news/raw.cgi?lugnet.faq:431
three or four times before I could understand what the heck Jacob and Jeremy
were talking about!
In lugnet.faq, Robert Munafo writes:
> I'll look around and try to find out what the "proper" format is. Maybe you
> could reply with a pointer to the LUGNET message I need to read. [...]
| | | | | | | | | | | | | | | | | | | | | | | | | | | |
Robert Munafo wrote:
> Good luck to anyone of average intelligence who tries to figure it out, though.
> Everyone on LUGNET must be at least MENSA level or higher. I had to read
> through the "format" answer http://www.lugnet.com/news/raw.cgi?lugnet.faq:431
> three or four times before I could understand what the heck Jacob and Jeremy
> were talking about!
You want to re-write the answer? :-,
...or, to quote Bones:
I'm an engineer, dammit! Not a technical writer! :-D
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Jeremy H. Sproat writes:
> Robert Munafo wrote:
> > [...] I had to read through the "format" answer
> > http://www.lugnet.com/news/raw.cgi?lugnet.faq:431
> > three or four times before I could understand [...]
>
> You want to re-write the answer? :-,
Here's my attempt at a more understandable answer:
Subject: How do I format an entry in the Lugnet FAQ?
Content-Language: en
Topic-Level: 1
Revision: Jacob Sparre Andersen, 1999-05-06
Revision: Jeremy Sproat, 1999-05-10
Revision: Robert Munafo, 1999-07-13
Reference: http://www.lugnet.com/news/display.cgi?lugnet.faq:83
Reference: http://www.lugnet.com/faq/?n=180&t=i&v=a
Location: /faq/
<P>If you already have a FAQ answer you want to add to the LUGNET FAQ database,
you first need to add headers like the ones you see above this paragraph, and
then add HTML formatting tags like you see scattered throughout this answer.
LUGNET uses HTML formatting and extra header fields so that it can supply FAQ
information to its web browser users as well as more traditional ASCII FAQ
files to other users.</P>
<P>Add a header first. You can cut and paste a header from another FAQ entry
like this one, but make sure to change everything that needs to be changed!</P>
<P>The following entries are allowed in the header, but not all are
required:</P>
<PRE>
Subject: [the question]
Content-Language: [ISO 639 language code. Use "en" for English
and "en-US" for the USA dialect of English.]
Topic-Level: [integer, 0 is beginner/easy/simple]
Revision: [author, ISO date]
Location: [comma-separated list of Lugnet relative
URI's]
Translated-From: [ISO 639 language code]
Translator: [translator, ISO date]
Originator: [originator, ISO date]
Comment: [will be ignored]
Include: [relative (Unix) file name for another
FAQ entry - only header entries that
not already occur will be included]
Reference: [URL to the source of the answer]
</PRE>
<P>After you have added the header, go through the text. A basic familiarity
with the commonly used HTML tags is useful; if you need help look at other FAQ
answers on LUGNET for examples.</P>
<P>Mark the beginning and end of each paragraph with P and /P.</P>
<P>Mark boldface with B and /B, and italics with I and /I.</P>
<P>Any text that you want to appear in a monospaced font, such as ASCII
formatted tables or ASCII graphics illustrations, should be marked with PRE and
/PRE.</P>
<P>If you want a paragraph to be indented to stand out, you can use BLOCKQUOTE
and /BLOCKQUOTE .</P>
<P>Mark URLs with A HREF and /A. Make sure the URL appears twice, once inside
quotes for the HREF and once as plain text, so the user can view and copy the
URL text if he/she wants to.</P>
<P>Other valid HTML tags include DL and /DL, DT and /DT, DD and /DD, TT and
/TT.</P>
<P>After you have finished formatting your FAQ entry in HTML, save it in a text
file. You should be able to view the file with a web browser and the formatting
should look correct (except that the headers will all be run together into one
paragraph). You might need to put an "HTML" tag at the very beginning of the
file to get the browser to format the text in HTML.</P>
<P>When the formatting looks good, submit your FAQ item to the LUGNET group
lugnet.faq. The message will be automatically processed by a LUGNET computer
program. The LUGNET program creates a FAQ answer database entry by starting
with the following text:</P>
<PRE>
<HTML Lang="%Content-Language%">
<Title>%Subject% - The Lugnet FAQ</Title>
<H1>%Subject%</H1>
%Answer%
</HTML>
</PRE>
<P>It then substituts the string <Code>%Answer%</Code> with all of your
HTML formatted content following the headers, and substitutes
<Code>%Content-Language%</Code> and <Code>%Subject%</Code> with the text you
put in the corresponding headers. For the FAQ answer submission to work, the
result should be a strict subset of valid HTML. </P>
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Robert:
[...]
It appears that you were correct, when you wrote that the
explanation of the FAQ file format is incomprehensible.
I will try to write something easier based on what you wrote
(and what I think you misunderstood).
Play well,
Jacob
------------------------------------------------
-- E-mail: sparre@cats.nbi.dk --
-- Web...: <URL:http://www.ldraw.org/FAQ/> --
------------------------------------------------
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Jacob Sparre Andersen writes:
> Robert:
>
> [...]
>
> It appears that you were correct, when you wrote that the
> explanation of the FAQ file format is incomprehensible.
>
> I will try to write something easier based on what you wrote
> (and what I think you misunderstood).
Thanks, that would be great. The three parts I think I misunderstood are:
1. The question says "How do I format an entry for the Lugnet FAQ?" so
naturally I assumed the answer would be in the form of a step-by-step how-to.
Instead it's a terse technical statement of the format.
2. The part of the answer which begins with "The answer should be written so
the following document..." is particularly confusing, because it doesn't
explain what's going on. What's really going on is that an automated process of
some kind will be extracting parts of the answer a Lugnet user has submitted
and inserting them into the "following document". You need to explain this in a
simpler way (like I did in my rewrite)
3. The most confusing part of all results from the many "<" and ">" in
the "following document". This results simply from the fact that I am reading
your raw submission as a LUGNET message and there is no formatter present to do
the formatting. As you can see, my rewrite took this implicitly into account,
and in fact my rewrite will no longer be understandable once the automatic
formatter exists because the embedded HTML tags it self-refers to will no
longer be visible to the reader.
So I would actually suggest taking the whole answer and putting it inside <PRE>
and </PRE> markers. That way, you don't have to deal with any of the issues of
whether the answer is being viewed in formatted or unformatted state.
Unfortunately (I think) < and > are still requred inside <PRE> blocks if
they are part of a valid HTML tag.
- Robert Munafo
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Robert Munafo <munafo@gcctech.com> wrote:
[snip]
> once the automatic formatter exists
[snip, snip; goodbye context]
So, there will be but does not currently exist a tool which extracts
messages from this newsgroup, checks for actual properly-formatted ones,
checks revision, and compiles a useful human-readable FAQ (in plain text
and/or html for web display)?
Where is the authoritative FAQ repository? Here?
--
Matthew Miller ---> mattdm@mattdm.org
Quotes 'R' Us ---> http://quotes-r-us.org/
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
This might not be my place to suggest, as I have not yet contributed anything,
but is there some particular reason why we are custom designing out own thing,
instead of using a standardized format which already has conversion tools, like
Docbook? IMHO, that would be much easier for everyone involved- conversion
tools already exist for everything short of stone tablets, and volumes have
been written about using it. I for one certainly don't have time to learna
format specifically for this, and would love to contribute. Anyway, that is my
two cents-
Luis
P.S. If you want to check out a useful example of Docbook (well, linuxdoc, but
same general idea), check out my Mindstorms LegOS HOWTO at:
http://arthurdent.dorm.duke.edu/HOWTO/HOWTO.html
In the same directory, you'll find .sgml source, and conversions to .dvi and
.ps. Converting it to very well formed text is also exactly one command away. I
strongly recommend this as a solution.
In lugnet.faq, Matthew Miller writes:
> Robert Munafo <munafo@gcctech.com> wrote:
> [snip]
> > once the automatic formatter exists
> [snip, snip; goodbye context]
>
> So, there will be but does not currently exist a tool which extracts
> messages from this newsgroup, checks for actual properly-formatted ones,
> checks revision, and compiles a useful human-readable FAQ (in plain text
> and/or html for web display)?
>
> Where is the authoritative FAQ repository? Here?
>
> --
> Matthew Miller ---> mattdm@mattdm.org
> Quotes 'R' Us ---> http://quotes-r-us.org/
>
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Luis Villa wrote:
> This might not be my place to suggest, as I have not yet contributed anything,
But isn't this a contribution? :-,
> but is there some particular reason why we are custom designing out own thing,
> instead of using a standardized format which already has conversion tools, like
> Docbook?
Well, we are and we aren't using standardized formats. BTW, I've never
heard about Docbook; I'd like to hear more.
The FAQ format was chosen as a compromise between using a format many people
are already familiar with, and being able to smoothly integrate the FAQ into
the LUGNET heirarchy. It (the FAQ format) is based on already-existing and
widely-used standards, and is essentially broken into two parts:
...The Header section, which follows basic Internet header protocol,
enabling quick and painless parsing of the file's meta-information.
...And the Body section, which consists of a (soon-to-be) well-defined
subset of HTML. Restricting the type of HTML used in the body will enable a
good-looking and well-formed output in true HTML and text, as well as a
plethora of other output options as they present themselves.
Because of these choices of standards, the FAQ format presents several
benefits for the authors.
> P.S. If you want to check out a useful example of Docbook (well, linuxdoc, but
> same general idea), check out my Mindstorms LegOS HOWTO at:
> http://arthurdent.dorm.duke.edu/HOWTO/HOWTO.html
This link seems to be broken. The server gives me the "file not found"
error.
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, "Luis Villa" <liv@duke.edu> writes:
> This might not be my place to suggest, as I have not yet contributed anything,
That's OK -- there doesn't necessarily have to be a strong correlation
between the two, right? You've got an opinion and if it's logical (and not
a drive-by joke), then it's welcome here no matter what you have or haven't
contributed.
> but is there some particular reason why we are custom designing out own thing,
> instead of using a standardized format which already has conversion tools, like
> Docbook?
For certain types of things, it's faster, easier, and wiser in the long run
to make something new from scratch tailored to your needs than to go with
something that exists out there already and try to adapt it (even with the
source). It depends on the project. FAQs, for example, tend toward simple
questions and answers in simple paragraphs, with an occasional something-
extra...they don't need to be (and shouldn't be, IMHO) overly complex or
expressive, else they become difficult to maintain or contribute to.
Also keep in mind, there are very few web things in the LUGNET chute that
aren't bleeding-edge community design. For example, although the newsserver
is 10-year-old technology, there's nothing else like the web interface which
sits on top of it because nothing else out there fit the bill.
I'm extremely skeptical that there's any web tool out there that'll do
exactly (or even come anywhere near close to doing) what we want for the
FAQ. Sure, there might be tools which help with DTD'ing and HTML'ing, but
those aren't really a significant part of the equation...not enough to be
worth grabbing a huge software package for (but certainly worth
investigating Perl libraries or 'lynx --dump').
I don't doubt that it would be possible to use Docbook and make something
really really wonderful. But I can't imagine it not taking months and
months and then getting painted into a corner of having to install more and
more external tools to work with it and interface to other things, and then
still not having it come out as originally envisioned. It just seems like a
horrible can of worms to me.
> IMHO, that would be much easier for everyone involved- conversion
> tools already exist for everything short of stone tablets, and volumes have
> been written about using it.
"Volumes have been written about using it." -- isn't that a minus rather
than a plus? Sounds like a red-flag to me.
> I for one certainly don't have time to learna
> format specifically for this, and would love to contribute. Anyway, that is my
> two cents-
> Luis
Don't worry -- anything you see now is still in a protoplasm stage; it's a
mess because it's still evolving. When it's all ready for general
contribution, it'll really be a snap to contribute and maintain things --
all through simple news discussions to talk about suggestions/edits, and
through simple web pages for making the actual changes.
> P.S. If you want to check out a useful example of Docbook (well, linuxdoc, but
> same general idea), check out my Mindstorms LegOS HOWTO at:
>
> http://arthurdent.dorm.duke.edu/HOWTO/HOWTO.html
>
> In the same directory, you'll find .sgml source, and conversions to .dvi and
> .ps. Converting it to very well formed text is also exactly one command away. I
> strongly recommend this as a solution.
That URL didn't work for me, but I found this...(was it the best place to
look?)...
http://www.oasis-open.org/docbook/
http://www.oasis-open.org/docbook/documentation/reference/index.html
Well, Docbook sure looks nice and robust and well-thought-out, but given its
size and complexity, how could it possibly be comprehensible to a beginner?
Why, just downloading the reference zips online results in five hundred HTML
files of documentation totalling three and a half megabytes. :-(
That's great and all if you're writing a series of technical manuals for 747
jetliners, but it's probably about three orders of magnitude more complex
than what's needed here for simple FAQ things.
I looked through about 50 of the pages and got lost pretty quickly. I guess
that's why there must be other books on it, huh? :-)
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Matthew Miller wrote:
> Robert Munafo <munafo@gcctech.com> wrote:
> > once the automatic formatter exists
> So, there will be but does not currently exist a tool which extracts
> messages from this newsgroup, checks for actual properly-formatted ones,
> checks revision, and compiles a useful human-readable FAQ (in plain text
> and/or html for web display)?
I have written a tool for my own uses which formats FAQ files on the hard
disk. It does not, however, index, validate, searchify, or otherwise
process the files for LUGNET consumption -- that'll be implemented by Todd
or person(s) commissioned by Todd.
As for extraction of FAQ items from the newsgroup, there are too many
variables to work with -- such as which version of this FAQ item do we
process, how to moderate FAQ submissions, etc.
I suspect (largely because this is the most recent plan) that the FAQ items
will reside on LUGNET's hard disk as individual files, much in the same
fashion as NNTP articles.
> Where is the authoritative FAQ repository? Here?
This newsgroup (lugnet.faq) for now. I'm working on getting all of Robert's
new items into my local collection, and I guess that would be as good as
official. Putting the ZIP file up for grabs, however, is proving to be
something of an adventure.
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.faq, Sproaticus <jsproat@io.com> writes:
> [...] I'm working on getting all of Robert's
> new items into my local collection, and I guess that would be as good as
> official. Putting the ZIP file up for grabs, however, is proving to be
> something of an adventure.
If you're having trouble uploading it somewhere to distribute it, you could
e-mail it to me and I could make it available for anonymous-ftp.
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Todd Lehman wrote:
> In lugnet.faq, Sproaticus <jsproat@io.com> writes:
> > [...] I'm working on getting all of Robert's
> > new items into my local collection, and I guess that would be as good as
> > official. Putting the ZIP file up for grabs, however, is proving to be
> > something of an adventure.
> If you're having trouble uploading it somewhere to distribute it, you could
> e-mail it to me and I could make it available for anonymous-ftp.
That sounds good. I tend to update it from time to time; e.g. most recently
adding all of Robert Munafo's entries and giving them filenames, fixing some
headers, etc. For more than one submission, would you prefer e-mail or does
LUGNET's FTP have a write-only incoming directory? I'm fine either way.
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.faq, Jeremy H. Sproat writes:
> > If you're having trouble uploading it somewhere to distribute it, you
> > could e-mail it to me and I could make it available for anonymous-ftp.
>
> That sounds good. I tend to update it from time to time; e.g. most
> recently adding all of Robert Munafo's entries and giving them filenames,
> fixing some headers, etc. For more than one submission, would you prefer
> e-mail or does LUGNET's FTP have a write-only incoming directory? I'm
> fine either way.
There's a write-only directory called 'incoming' -- you could drop it there,
then send me a note and I'll put in into the read-only directory called
'outgoing'.
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Sproaticus <jsproat@io.com> wrote:
> > Where is the authoritative FAQ repository? Here?
>
> This newsgroup (lugnet.faq) for now. I'm working on getting all of Robert's
What happens if I post random extremely unhelpful (counter-helpful, even)
but properly-formatted messages to the newsgroup? Will I break things? :)
--
Matthew Miller ---> mattdm@mattdm.org
Quotes 'R' Us ---> http://quotes-r-us.org/
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Matthew Miller wrote:
> Sproaticus <jsproat@io.com> wrote:
> > > Where is the authoritative FAQ repository? Here?
> > This newsgroup (lugnet.faq) for now. I'm working on getting all of Robert's
> What happens if I post random extremely unhelpful (counter-helpful, even)
> but properly-formatted messages to the newsgroup? Will I break things? :)
Nope! That's the beauty of not being automated yet! :-,
If I see a new article in this ng (or anywhere else in LUGNET, for tha
tmatter) that's probably appropriate for the FAQ, I snag a copy of it and
store it in my "LUGNET FAQ -- TODO" folder. This includes all sorts of
items, a lot of which have no formatting at all.
However you want to put it in, is fine -- have at it!
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.faq, mattdm@mattdm.org (Matthew Miller) writes:
> So, there will be but does not currently exist a tool which extracts
> messages from this newsgroup, checks for actual properly-formatted ones,
> checks revision, and compiles a useful human-readable FAQ (in plain text
> and/or html for web display)?
Right, basically. One thing we were talking about a while back was using
'lynx --dump' to convert the source data (which includes minimal HTML
markup) into ASCII for spitting out into the ng.
> Where is the authoritative FAQ repository? Here?
For right now, yes, but ultimately this group will probably be more like the
.cad.dat.* groups, where the groups are a "dumping group" or "echo area" of
data posted automatically by the server when it's been submitted to a
web-page. The advantage of echoing it into this ng is that it's a great way
to process and record discussions that pop up.
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Todd Lehman <lehman@javanet.com> wrote:
> Right, basically. One thing we were talking about a while back was using
> 'lynx --dump' to convert the source data (which includes minimal HTML
> markup) into ASCII for spitting out into the ng.
Good idea. The current 200+ messages aren't any fun to read.
--
Matthew Miller ---> mattdm@mattdm.org
Quotes 'R' Us ---> http://quotes-r-us.org/
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.faq, mattdm@mattdm.org (Matthew Miller) writes:
> Todd Lehman <lehman@javanet.com> wrote:
> > Right, basically. One thing we were talking about a while back was using
> > 'lynx --dump' to convert the source data (which includes minimal HTML
> > markup) into ASCII for spitting out into the ng.
>
> Good idea. The current 200+ messages aren't any fun to read.
Wouldn't it be cool if there were something other than HTML that was
actually readable as plaintext but which also looked great when converted to
HTML?
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Todd Lehman writes:
> Wouldn't it be cool if there were something other than HTML that was
> actually readable as plaintext but which also looked great when converted to
> HTML?
It's interesting you said that. For my own purposes (and this is unrelated to
LUGNET) I've developed a few simple ASCII tags that I use in my web page source
files. My web pages (on http://www.mrob.com/ ) are all typed in as plain ASCII
text files. I use a fairly simple Perl script to convert them to HTML using the
following rules:
A bunch of characters with *asterisks* on either side is put into *italics*.
Similarly, #sharp signs# become #boldface# and =equals signs= generate =H1
style headers= (these are usually by themselves on a line).
An A NAME anchor appears by itself on a line followed by two colons, like this:
anchor1::
An A HREF anchor appears like this: [lugnet|www.lugnet.com]. The first part
inside the brackets is the text and the second part is the URL. The "http://"
is optional.
An A HREF that points to an A NAME elsewhere on the same page is like this
[section 1|#section1]. The script is smart enough to treat a reference to a
local label as a special case and it doesn't try to stick "http://" in front of
it.
Any lines that appear with blank spaces in front of them are treated as part of
a BLOCKQUOTE block. Any blank lines (consisting of only whitespace) are treated
as paragraph breaks and generate the appropriate P tags.
I don't have a symbol for the other tags like DD and DT but it would be easy to
add them. You'd have to use characters that aren't commonly used or recognize
then only when they appear at the beginning of a line. For example, a
bullet-item could be a line starting with whitespace, a lone "*" and more
whitespace. Other characters to use would be @, %, &, maybe _, |, {, }.
Finally, any line with ":::" is not processed at all. This is to handle those
rare cases where the text has to include characters that would otherwise be
treated as formatting. And, the script converts < and > into < and >
unless they're part of an HTML tag. It also notices the PRE tag and leaves
anything alone if it's inside a PRE block.
If you were willing to use a method like this process the text, then you could
make the submission format much simpler.
- Robert Munafo
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Robert Munafo wrote:
> My web pages (on http://www.mrob.com/ ) [...]
Okay, I've been reading aout numbers for a half hour now. Very interesting
reading, at least for a geek such as I. :-,
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Sproaticus <jsproat@io.com> writes:
> Robert Munafo wrote:
> > My web pages (on http://www.mrob.com/ ) [...]
>
> Okay, I've been reading aout numbers for a half hour now. Very interesting
> reading, at least for a geek such as I. :-,
Me too! Great stuff, Robert! Say, here's a fun tidbit about the golden
ratio if you're interested --
If you plot multiples of phi on the unit interval [0,1), then the nth
multiple always falls into the largest "untouched" sub-interval. That is
to say, if you have the iterated map
f(n) = 0, n = 0
f(n) = frac(f(n-1) + phi), n > 0
where frac(x) is x - int(x), then as n counts upwards, the plot goes like
this:
0 .1 .2 .3 .4 .5 .6 .7 .8 .9 1
|------|------|------|------|------|------|------|------|------|------|
0 x
1 x x
2 x x x
3 x x x x
4 x x x x x
5 x x x x x x
6 x x x x x x x
7 x x x x x x x x
8 x x x x x x x x x
9 x x x x x x x x x x
10 x x x x x x x x x x x
11 x x x x x x x x x x x x
:
etc.
(Or visualize it as points on a circle, since eventually the interval
connecting .94 back to 0 through 1 gets chosen. :-)
IIRC, any irrational can be used to fill the unit interval (via its
multiples) this way, but only the golden ratio conveniently chooses the
largest untouched region at each iteration. I don't remember the proof (my
dad showed it to me many years ago) but it's got some interesting
applications, one of the most interesting of which is the subject of
_phyllotaxis_ in plant left arrangements to help maximize light absorption
for certain leaf-shapes. :)
Another interesting application (my personal favorite) is choosing a
least-visited area of the screen when doing progressive image refinement on
an unknown image (such as a 2D fractal) -- although integer bit-reversal
works just as well at partitioning the space and doesn't have the same
floating point errors to worry about.
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Well, *since* we're in the off-topic, GEEK newsgroup... <-: (-; [-8
Yes, I knew those things about *phi* and I've left out a lot of other stuff
about *phi* for aesthetic reasons.
I've been at a bit of a quandary about what to do with my number and large
number web pages. [1], [2] There are some numbers like
*pi* and *phi* which have so many properties that they would be pages long.
That's a bit awkward because I'm trying to list all the numbers on one page. I
don't want to put each number on a seperate page because I hate other web pages
that are set up that way, it takes so much clicking to skim through something
or to visually search for something.
So the natural solution is to put certain numbers by themselves on a seperate
page. Once it got to that point I started thinking, this is too much work and I
stopped adding to the numbers page. Ideally, I want something that works just
like the site map on LUGNET (thank you Todd):
http://www.lugnet.com/sitemap.cgi?/
See that "level" bar at the top? Click on it and the whole page expands. That's
neat. If I could do that with my numbers page I could hide all the details
about *pi* and *e* until someone clicks on a different level. Trouble is, I
need to find a simple way to do it with static pages. Hmmm...
In lugnet.off-topic.geek, Todd Lehman writes:
> Me too! Great stuff, Robert! Say, here's a fun tidbit about the golden
> ratio if you're interested --
> [...]
(By the way, Todd, in case you're interested -- I once made a robot for a
virtual robot wars tounament called "death blossom", inspired by the movie _The
Last Starfighter_ [3]. The robot's attack worked by rotating and firing a shot
every 2 * PI * PHI radians. Due to the property you described, this had the
effect of optimally mowing down all robots on the field in the minimal amount
of time (statistically speaking) assuming that there is no prior knowledge of
the robots' locations.) "Death Blossom" did nothing else, but nevertheless
managed to do fairly well against many so-called "more sophisticated" robots in
tournaments. It seems many robots turn tail and run whenever they get hit.)
1. http://www.mrob.com/numbers.html
2. http://www.mrob.com/largenum.html
3. http://us.imdb.com/Title?0087597
- Robert Munafo
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.off-topic.geek, Robert Munafo writes:
> (By the way, Todd, in case you're interested -- I once made a robot for a
> virtual robot wars tounament called "death blossom", inspired by the movie
> _The Last Starfighter_ [3]. The robot's attack worked by rotating and firing
> a shot every 2 * PI * PHI radians. [...]
Majorcool geekage!
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Robert Munafo writes:
> In lugnet.faq, Todd Lehman writes:
> > Wouldn't it be cool if there were something other than HTML that was
> > actually readable as plaintext but which also looked great when converted
> > to HTML?
>
> It's interesting you said that. For my own purposes (and this is unrelated
> to LUGNET) I've developed a few simple ASCII tags that I use in my web page
> source files. My web pages (on http://www.mrob.com/ ) are all typed in as
> plain ASCII text files. I use a fairly simple Perl script to convert them to
> HTML using the following rules:
Hee hee! COOL!
Say, I wonder if math-geeks have a strong natural affinity toward custom
tools like that? Why design a special tool and a custom markup when HTML
already exists? Answer: Because it's easy, more pleasant to work with for
lots of semi-complex body text, and hey, making up new special-purpose
notations from scratch (as appropriate) is just a way of life in mathematics,
so it's a perfectly normal thing to do.
> A bunch of characters with *asterisks* on either side is put into *italics*.
> Similarly, #sharp signs# become #boldface# and =equals signs= generate =H1
> style headers= (these are usually by themselves on a line).
What did you use for superscript and subscript? Those all come out so nicely
on your pages!
> [...snip...]
> If you were willing to use a method like this process the text, then you
> could make the submission format much simpler.
"It's interesting you said that." :-) Yes, I think a non-HTML text-markup
approach ("markup" isn't probably the right word, but...) would be a vast
imrovement over HTML for a number of reasons:
1. Safer, quicker, and easier to edit/maintain.
2. Much more directly readable as (doesn't require conversion to) plain text.
3. Better suited to quick-editing via HTML forms than HTML itself is.
4. Much easier for a casual (non-computergeek) user to learn.
Here are some related (and admittedly biased) thoughts on that general topic,
from another thread in another newsgroup earlier this month:
http://www.lugnet.com/admin/general/?n=1977
http://www.lugnet.com/admin/general/?n=1983
http://www.lugnet.com/admin/general/?n=2034 *
http://www.lugnet.com/admin/general/?n=2144
The big question is: Is something like that sufficiently powerful to handle
FAQ stuff...? And I think it is...at least for ~99% of anything anyone would
ever want to write.
The article marked with * above is the one which hints at "exciting related
possibilities with regard to FAQ lists." So yes, I would definitely be
willing to accept FAQ entries formatted in a special non-HTML format...but
it would need to be consistent with the member-page syntax (which is still
in the design and requirements cycle/phase). After realizing that FAQ entries
could actually be handled like that almost just as easily as any other text,
I made it a higher priority to be able to support FAQ entries. As an
experiment last week, I converted about half of the LDraw FAQ (locally) just
to see how far the approach could be taken with a relatively complex FAQ. I
think it shows a lot of promise.
I got a big smile on my face, BTW, when you wrote that your link syntax
used a vertical bar inside the brackets to separate the URL portion from the
visible human-text portion,
[Robert Munafo's Home Page|www.mrob.com]
because I was thinking of the exact same character (only with reversed left &
right portions):
<www.mrob.com|Robert Munafo's Home Page>
(Vertical-bar seemed like the only logical choice.)
Anyway, I was talking (writing) with my sister a few days ago about the link
sub-syntax, and just after I'd sent one of the mails, I noticed something
problematic about the vertical-bar character...
If something is written as:
<http://www.w3.org/Addressing/URL/5_BNF.html|Uniform Resource Locators:BNF>
then, even though | (vline) isn't part of a valid URL, Netscape mail (and
likely MSOE as well) displays everything up through the end of the word
"Uniform" as a hyperlink. It's certainly an unfortunate bug in Netscape's
mail reader, but for practical reasons on those grounds, it's really worth
considering something other than | as the separator (for this LUGNET purpose).
So what other characters are good options? Other illegal characters like \
and ^ are likely to have the same problems, but -- there is always the space
character! :-)
<http://www.w3.org/Addressing/URL/5_BNF.html Uniform Resource Locators:BNF>
Although a space is somewhat less "clean" and less readable than |, it has the
advantage that the text-line can be auto-wrapped more nicely by typical text
editors,
<http://www.w3.org/Addressing/URL/5_BNF.html
Uniform Resource Locators:BNF>
which might be more pleasant in long lists of arbitrarily URLs. (And the
text-to-HTML converter of course simply trims leading space from the human-
readable portion of the link.)
As for the other parts, my sister is in the process of convincing me not to use
[[Foo]] and [[[Foo]]] for headings but instead to use
Foo foo foo
===========
and
Foo foo foo
-----------
since those are more intuitive both to read and to write (especially to read).
The idea of using -'s and ='s comes from the 'txt2html' tool which Tim Rueger
pointed out:
http://www.lugnet.com/admin/general/?n=2097
There's also a huge list of plaintext-to-HTML filters at W3:
http://www.w3.org/Tools/Misc_filters.html
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Todd Lehman writes:
> [...]
> If something is written as:
>
> <http://www.w3.org/Addressing/URL/5_BNF.html|Uniform Resource Locators:BNF>
>
> then, even though | (vline) isn't part of a valid URL, Netscape mail (and
> likely MSOE as well) displays everything up through the end of the word
> "Uniform" as a hyperlink. It's certainly an unfortunate bug in Netscape's
> mail reader, but for practical reasons on those grounds, it's really worth
> considering something other than | as the separator (for this LUGNET purpose).
> [...]
Gulp, I made the same mistake in my URL detection code on the web interface
here. Just tightened up the set of allowable characters a bit and regression
tested...much better now.
BTW, I'm consciously going against what W3 says about the ~ symbol. According
to the 'national' production here...
http://www.w3.org/Addressing/URL/5_BNF.html
...the tilde character (ASCII code 126) isn't valid in URLs. I've seen people
write %7E instead of ~, which is apparently the only correct way to write it,
but almost no one is aware of that (as judged by the huge number of ~'s seen
in URLs *everywhere* on the net). So practically speaking, it would be more
broken (in users' minds) to disallow ~ than to allow it.
So what's up with that, anyway? How the heck did ~ gain such huge popularity
if it's not officially allowed in URLs? Was it allowed once upon a time? Or
is it simply part of today's de facto URL standard because of its extremely
wide misuse? (It's too bad that it's not officially allowed, because it's a
great character for what it's typically used for.)
--Todd
[followups set to lugnet.publish]
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Todd Lehman wrote:
> ...the tilde character (ASCII code 126) isn't valid in URLs. I've seen people
> write %7E instead of ~, which is apparently the only correct way to write it,
> but almost no one is aware of that (as judged by the huge number of ~'s seen
> in URLs *everywhere* on the net). So practically speaking, it would be more
> broken (in users' minds) to disallow ~ than to allow it.
>
> So what's up with that, anyway? How the heck did ~ gain such huge popularity
> if it's not officially allowed in URLs? Was it allowed once upon a time? Or
> is it simply part of today's de facto URL standard because of its extremely
> wide misuse? (It's too bad that it's not officially allowed, because it's a
> great character for what it's typically used for.)
>
> --Todd
>
> [followups set to lugnet.publish]
Unix makes the ~ character a users home directory ... so UnixSystem/~lee ... would
be my
home directory ...
Now because of this, and since the Internet (Arpanet) were all college schools when
it 'went public', the system of choice at colleges was Unix, this made the web have
some of the same naming conventions ...
It's an old thing ... but you know what it's like to get hundreds of people to
change ... think of what the colleges went through to get millions of students to
change ... which they didn't.
-Lee.
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.publish, Lee Jorgensen writes:
> Todd Lehman wrote:
> > So what's up with that, anyway? How the heck did ~ gain such huge
> > popularity if it's not officially allowed in URLs? Was it allowed once
> > upon a time? Or is it simply part of today's de facto URL standard
> > because of its extremely wide misuse? (It's too bad that it's not
> > officially allowed, because it's a great character for what it's typically
> > used for.)
>
> Unix makes the ~ character a users home directory ... so UnixSystem/~lee ...
> would be my home directory ...
>
> Now because of this, and since the Internet (Arpanet) were all college
> schools when it 'went public', the system of choice at colleges was Unix,
> this made the web have some of the same naming conventions ...
>
> It's an old thing ... but you know what it's like to get hundreds of people
> to change ... think of what the colleges went through to get millions of
> students to change ... which they didn't.
That's what I thought the etymology of ~ was in URLs too -- but how did it
ever get *allowed* in URLs in the first place? That's what baffles me. The
first time someone tried it, why didn't it fail? The early browsers and httpd
daemons must've silently just passed it on through (as they all do today), and
people must never have bothered to look to see if ~ was actually valid in URLs.
I'll bet it got out of control pretty quickly, especially because it's such
a natural thing to want to use! :-)
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Wouldn't NCSA be the one to blame? ncsa-httpd supported the
feature that caused people to use ~. Did they make that up?
KL
Todd Lehman wrote:
>
> In lugnet.publish, Lee Jorgensen writes:
> > Todd Lehman wrote:
> > > So what's up with that, anyway? How the heck did ~ gain such huge
> > > popularity if it's not officially allowed in URLs? Was it allowed once
> > > upon a time? Or is it simply part of today's de facto URL standard
> > > because of its extremely wide misuse? (It's too bad that it's not
> > > officially allowed, because it's a great character for what it's typically
> > > used for.)
> >
> > Unix makes the ~ character a users home directory ... so UnixSystem/~lee ...
> > would be my home directory ...
> >
> > Now because of this, and since the Internet (Arpanet) were all college
> > schools when it 'went public', the system of choice at colleges was Unix,
> > this made the web have some of the same naming conventions ...
> >
> > It's an old thing ... but you know what it's like to get hundreds of people
> > to change ... think of what the colleges went through to get millions of
> > students to change ... which they didn't.
>
> That's what I thought the etymology of ~ was in URLs too -- but how did it
> ever get *allowed* in URLs in the first place? That's what baffles me. The
> first time someone tried it, why didn't it fail? The early browsers and httpd
> daemons must've silently just passed it on through (as they all do today), and
> people must never have bothered to look to see if ~ was actually valid in URLs.
> I'll bet it got out of control pretty quickly, especially because it's such
> a natural thing to want to use! :-)
>
> --Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Universal Resource Locators (that's what it stands for) started out as
path descriptions. For why twiddle is supported, you might read up on
the Andrew File System.
~lpien is a valid file path (that is, you can cd to it) on a CTP unix
box, or was until I resigned. At various times in my career there it
resolved to various different places as my home directory got moved
around from machine to machine.
Therefore w3.ctp.com/~lpien is a valid URL if you're on the CTP internal
network (the internal nameserver knows how to resolve w3.ctp.com)
That's how I understood it.
If url's disallowed twiddle, I would consider them broken.
--
Larry Pieniazek larryp@novera.com http://my.voyager.net/lar
- - - Web Application Integration! http://www.novera.com
fund Lugnet(tm): http://www.ebates.com/ Member ref: lar, 1/2 $$ to
lugnet.
NOTE: I have left CTP, effective 18 June 99, and my CTP email
will not work after then. Please switch to my Novera ID.
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.publish, Larry Pieniazek writes:
> Universal Resource Locators (that's what it stands for) started out as
> path descriptions. For why twiddle is supported, you might read up on
> the Andrew File System.
>
> ~lpien is a valid file path (that is, you can cd to it) on a CTP unix
> box, or was until I resigned. At various times in my career there it
> resolved to various different places as my home directory got moved
> around from machine to machine.
>
> Therefore w3.ctp.com/~lpien is a valid URL if you're on the CTP internal
> network (the internal nameserver knows how to resolve w3.ctp.com)
>
> That's how I understood it.
Total grokkage there! (Been using ~ for that since '84 :) It's one of the
Truly Great Simplicities of modern day computing life, IMHO.
> If url's disallowed twiddle, I would consider them broken.
I think -that's- the key thing! (People considering it broken.) URLs & URIs
do disallow twiddle (ASCII 126) but nevertheless, I guess people all 'round
the world years must've banded together (or independently) said, "Hey, that's
majorly broken; URLs *should* allow twiddles, so I'm gonna use them anyway.
I'll even hack my local httpd and Mosaic to make it work if I have to." :-)
Mystery still remains why W3 still says twiddle (~) is disallowed in URLs. :)
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Todd Lehman:
> BTW, I'm consciously going against what W3 says about the ~ symbol.
> According to the 'national' production here...
>
> http://www.w3.org/Addressing/URL/5_BNF.html
>
> ...the tilde character (ASCII code 126) isn't valid in URLs. I've
> seen people write %7E instead of ~, which is apparently the only
> correct way to write it, but almost no one is aware of that (as
> judged by the huge number of ~'s seen in URLs *everywhere* on the
> net). So practically speaking, it would be more broken (in users'
> minds) to disallow ~ than to allow it.
The primary reason for disallowing ~ is the special
treatment it gets in several European languages.
If people type <slash> <tilde> <s> <p> (expecting "/~sp")
some systems will just return "sp".
Similarly if people type <slash> <tilde> <n> <o> (expecting
"/~no"...) most systems will return "/ño".
I have actually seen the latter of these two misprints in a
Danish newspaper.
> So what's up with that, anyway? How the heck did ~ gain such huge
> popularity if it's not officially allowed in URLs?
Both the original CERN http daemon and Apache uses ~ for
user home pages.
> Was it allowed once upon a time?
Dunno.
It has probably been allowed until people started thinking
about which characters gives trouble.
"ð" is also disallowed, just because of a few silly American
operating systems.
Play well,
Jacob
------------------------------------------------
-- E-mail: sparre@cats.nbi.dk --
-- Web...: <URL:http://www.ldraw.org/FAQ/> --
------------------------------------------------
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.publish, sparre@sys-323.risoe.dk (Jacob Sparre Andersen) writes:
> [...]
> Similarly if people type <slash> <tilde> <n> <o> (expecting
> "/~no"...) most systems will return "/ño".
>
> I have actually seen the latter of these two misprints in a
> Danish newspaper.
>
> > So what's up with that, anyway? How the heck did ~ gain such huge
> > popularity if it's not officially allowed in URLs?
>
> Both the original CERN http daemon and Apache uses ~ for
> user home pages.
>
> > Was it allowed once upon a time?
>
> Dunno.
>
> It has probably been allowed until people started thinking
> about which characters gives trouble.
> [...]
Wow. OK, that certainly makes sense. So, the hypothesis is that "~" may
have been disallowed so that commonly available software (which used "~" for
special formatting tricks) for certain languages didn't have to be altered
to parse-recognize URLs and avoid converting "~" in those cases? In other
words, it may have been a legacy thing for backward compatibility with
existing text processing tools?
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Todd Lehman:
> Wow. OK, that certainly makes sense. So, the hypothesis is that "~" may
> have been disallowed so that commonly available software (which used "~" for
> special formatting tricks) for certain languages didn't have to be altered
> to parse-recognize URLs and avoid converting "~" in those cases? In other
> words, it may have been a legacy thing for backward compatibility with
> existing text processing tools?
It works like I described (at least the ~n case) even in
Word97/Win95(DK), so I am not sure it is correct to call it
a "backward" compatibility problem[1]. People have to learn
to type <tilde> <space> to get a tilde on most European
PC's.
Play well,
Jacob
1) Despite my low opinions of MS products.
------------------------------------------------
-- E-mail: sparre@cats.nbi.dk --
-- Web...: <URL:http://www.ldraw.org/FAQ/> --
------------------------------------------------
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Jacob Sparre Andersen skrev i meddelandet ...
> Todd Lehman:
>
> > Wow. OK, that certainly makes sense. So, the hypothesis is that "~" may
> > have been disallowed so that commonly available software (which used "~" • for
> > special formatting tricks) for certain languages didn't have to be altered
> > to parse-recognize URLs and avoid converting "~" in those cases? In other
> > words, it may have been a legacy thing for backward compatibility with
> > existing text processing tools?
>
> It works like I described (at least the ~n case) even in
> Word97/Win95(DK), so I am not sure it is correct to call it
> a "backward" compatibility problem[1]. People have to learn
> to type <tilde> <space> to get a tilde on most European
> PC's.
It has nothing to do with different software (MS or others), it's in the
keyboard driver (OK, that's probably MS). On a Swedish keyboard '~' is a 'dead
char', which is automatically composed with the next char if it's possible.
This has been so (at least) since the first PC-keyboard.
I suppose this is defined somewhere in some ISO-standard (don't remember the
number right now :-)
The other 'dead-chars' on my keyboard are ´ (acute accent), ` (grave accent)
and ¨ (trema)
Despite this, all (?) Swedish ISP:s use the tilde in user names (see sig.)
I think the main problem is that US keyboards/users doesn't care for the rest
of the world, while the rest of the world have to care both for themselves and
the US.
(On of my pet peeves: US chauvinism...)
--
Anders Isaksson, Sweden
BlockCAD: http://user.tninet.se/~hbh828t/proglego.htm
Gallery: http://user.tninet.se/~hbh828t/gallery.htm
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.publish, Anders Isaksson writes:
> Despite this, all (?) Swedish ISP:s use the tilde in user names (see sig.)
>
> I think the main problem is that US keyboards/users doesn't care for the
> rest of the world, while the rest of the world have to care both for
> themselves and the US. (On of my pet peeves: US chauvinism...)
Me too -- I know what you mean. And the cluelessness about non-U.S.
conventions is IMHO even worse than the chauvinism...
One thing to remember, though: The A in ASCII does stand for American. :-)
Say, I've got a user-interface question about the ~ keyboard thing. Do you
*have to* type Tilde-Space to get a plain tilde (meaning ASCII character #126),
or can you type Tilde-Tilde (which is probably 1/10 second faster) to get a
plain tilde? Also, when typing Tilde-Space, if you accidentally forget to let
up the Shift key before hitting Space, is Tilde-Shift-Space interpreted the
same as Tilde-Space? Curious minds want to know! :-) I'm sure the answer
depends on the keyboard driver in use, but I wonder what some of the more
common drivers do...?
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Todd Lehman wrote:
>
> In lugnet.publish, Anders Isaksson writes:
> > Despite this, all (?) Swedish ISP:s use the tilde in user names (see sig.)
> >
> > I think the main problem is that US keyboards/users doesn't care for the
> > rest of the world, while the rest of the world have to care both for
> > themselves and the US. (On of my pet peeves: US chauvinism...)
>
> Me too -- I know what you mean. And the cluelessness about non-U.S.
> conventions is IMHO even worse than the chauvinism...
>
> One thing to remember, though: The A in ASCII does stand for American. :-)
>
> Say, I've got a user-interface question about the ~ keyboard thing. Do you
> *have to* type Tilde-Space to get a plain tilde (meaning ASCII character #126),
> or can you type Tilde-Tilde (which is probably 1/10 second faster) to get a
> plain tilde? Also, when typing Tilde-Space, if you accidentally forget to let
> up the Shift key before hitting Space, is Tilde-Shift-Space interpreted the
> same as Tilde-Space? Curious minds want to know! :-) I'm sure the answer
> depends on the keyboard driver in use, but I wonder what some of the more
> common drivers do...?
>
> --Todd
Tilde is not a shift+anykey character but rather RightAlt+anykey (ü in
Turkish "q" keyboards) character. RightAlt key (actually named as "Alt
Gr") behaves differently from LeftAlt key in most european keyboards and
only used for performing a third task with a given key.
It is strange but although we don't have any letters in our alphabet
that combined with tilde, we must press space too, to type tilde..:-)
Selçuk
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Todd Lehman skrev i meddelandet ...
> In lugnet.publish, Anders Isaksson writes:
> > Despite this, all (?) Swedish ISP:s use the tilde in user names (see sig.)
> >
> > I think the main problem is that US keyboards/users doesn't care for the
> > rest of the world, while the rest of the world have to care both for
> > themselves and the US. (On of my pet peeves: US chauvinism...)
>
> Me too -- I know what you mean. And the cluelessness about non-U.S.
> conventions is IMHO even worse than the chauvinism...
Agreed, but in my vocabulary 'clueless' rings harsher than 'chauvinist', am I
totally out of line there?
> One thing to remember, though: The A in ASCII does stand for American. :-)
I don't have all the messages in the thread left (and am off-line, as usual),
but I'm sure my message never mentioned ASCII, but ISO (which means
International Standardization Organization, or something like that) :-)
> Say, I've got a user-interface question about the ~ keyboard thing. Do you
> *have to* type Tilde-Space to get a plain tilde (meaning ASCII character • #126),
> or can you type Tilde-Tilde (which is probably 1/10 second faster) to get a
> plain tilde? Also, when typing Tilde-Space, if you accidentally forget to • let
> up the Shift key before hitting Space, is Tilde-Shift-Space interpreted the
> same as Tilde-Space?
Firstly, we have two different Alt keys on our keyboards, the left one is
'Alt', the right one is 'Alt Gr', and it's 'Alt Gr' to get the tilde, not
Shift. Now let's check...
~ (Alt Gr)Tilde-Space 1T, no space
~ (Alt Gr)Tilde-(Alt Gr)Space 1T, no space
~~ (Alt Gr)Tilde-(Alt Gr)Tilde 2T (Naturally!)
In fact Tilde-Anything comes as 'Tilde' 'Anything' *except* for Tilde-Space
and the few letters that can have a Tilde on top (ñ Tilde-n, õ Tilde-o, ã
Tilde-a).
To make (C-)programming even worse for us Europeans, the brackets {}[] also
use (Alt Gr), because our national characters åäöÅÄÖ occupy the keyboard space
where they are 'normally' located (I'm actually *happy* if I can find an
English keyboard when programming in C!). And ^(caret) is also a 'dead key'
like ~(tilde), so you never know if you pressed it until you press the next
key!
No wonder Pascal was invented in Europe!
Sometimes I even think the French have done the right thing (for once) with
their keyboards:
They have the upper row (with numbers and special characters) turned upside
down, so you use Shift to get the digits, and all the special characters are
reached unshifted! The numerical keyboard is always normal, so you have both
digits and everything else without shifting (OTOH, they have swapped a few
letters, just to make sure they differ from everybody else :-)
If you want a really different experience, you should try a Russian keyboard,
with both Cyrillic characters, and our ones (but on completely different
locations!)
(Ah, the memories...)
--
Anders Isaksson, Sweden
BlockCAD: http://user.tninet.se/~hbh828t/proglego.htm
Gallery: http://user.tninet.se/~hbh828t/gallery.htm
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Anders Isaksson wrote:
>
>
> Sometimes I even think the French have done the right thing (for once) with
> their keyboards:
>
> They have the upper row (with numbers and special characters) turned upside
> down, so you use Shift to get the digits, and all the special characters are
> reached unshifted! The numerical keyboard is always normal, so you have both
> digits and everything else without shifting (OTOH, they have swapped a few
> letters, just to make sure they differ from everybody else :-)
>
> If you want a really different experience, you should try a Russian keyboard,
> with both Cyrillic characters, and our ones (but on completely different
> locations!)
We have two different keyboard configurations here, "q" and "f". In the
first one, letters are all located as a "qwerty" layout with a few
additions like ç ö ü and some others that you can't see if I type, and
almost all the special characters are located differently, and the other
one is the true national keyboard layout and strictly preferred by
10-finger typers. Since I'm not a 10-finger typer and I started using
computers with standard US layout keyboards my personal choice is "q"
layout, and I feel myself completely dumb when I face with an "f" layout
keyboard..:-) It looks so stupid when you can't type anything right and
typing so slow if you are supposed to be servicing this computer..:-)
Selçuk
> (Ah, the memories...)
>
> --
> Anders Isaksson, Sweden
> BlockCAD: http://user.tninet.se/~hbh828t/proglego.htm
> Gallery: http://user.tninet.se/~hbh828t/gallery.htm
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.publish, Anders Isaksson writes:
> > > I think the main problem is that US keyboards/users doesn't care for the
> > > rest of the world, while the rest of the world have to care both for
> > > themselves and the US. (On of my pet peeves: US chauvinism...)
> > Me too -- I know what you mean. And the cluelessness about non-U.S.
> > conventions is IMHO even worse than the chauvinism...
> Agreed, but in my vocabulary 'clueless' rings harsher than 'chauvinist', am I
> totally out of line there?
Oh, not at all -- that's exactly what I meant. U.S. cluelessness about
non-U.S. conventions is even worse than U.S. chauvinism. At least chauvinists
know that other conventions exist. :-)
> > One thing to remember, though: The A in ASCII does stand for American. :-)
>
> I don't have all the messages in the thread left (and am off-line, as usual),
> but I'm sure my message never mentioned ASCII, but ISO (which means
> International Standardization Organization, or something like that) :-)
Yes, but the reason ISO-8859-1 (for example) has the ~ at character #126 is
because ISO-8859-1 includes the earlier ASCII set as a subset. Thus (unless
I'm mistaken), any low-numbered ISO-8859-1 character is also an ASCII character
at the same ordinal position. Naturally, the plain tilde character, being part
of ASCII for decades, shouldn't be ruled out as a valid character just because
it is an extra keystroke on some keyboards. In fact, many old American
keyboards were missing certain ASCII characters, and needed sometimes 3 or 4
keys (or sometimes chord sequences) to be pressed in order to summon them.
I remember a keyboard back in the early 80's which required something like
Ctrl-Shift-6 to get a ^ (plain caret)...that's still only 3 keys, but I'm
sure there were some with 4, I just can't recall what they were.
> Firstly, we have two different Alt keys on our keyboards, the left one is
> 'Alt', the right one is 'Alt Gr', and it's 'Alt Gr' to get the tilde, not
> Shift. Now let's check...
>
> ~ (Alt Gr)Tilde-Space 1T, no space
> ~ (Alt Gr)Tilde-(Alt Gr)Space 1T, no space
> ~~ (Alt Gr)Tilde-(Alt Gr)Tilde 2T (Naturally!)
OK, so the shortest sequence to get a plain tilde is pressing 3 keys?
That's not too bad after all. Here, we have to press 2 keys -- Shift plus
the tilde/backtick key.
> In fact Tilde-Anything comes as 'Tilde' 'Anything' *except* for Tilde-Space
> and the few letters that can have a Tilde on top (ñ Tilde-n, õ Tilde-o, ã
> Tilde-a).
>
> To make (C-)programming even worse for us Europeans, the brackets {}[] also
> use (Alt Gr), because our national characters åäöÅÄÖ occupy the keyboard space
> where they are 'normally' located (I'm actually *happy* if I can find an
> English keyboard when programming in C!). And ^(caret) is also a 'dead key'
> like ~(tilde), so you never know if you pressed it until you press the next
> key!
>
> No wonder Pascal was invented in Europe!
Hmm, but Pascal uses ^ all over the place (for pointers).
> Sometimes I even think the French have done the right thing (for once) with
> their keyboards:
>
> They have the upper row (with numbers and special characters) turned upside
> down, so you use Shift to get the digits, and all the special characters are
> reached unshifted! The numerical keyboard is always normal, so you have both
> digits and everything else without shifting (OTOH, they have swapped a few
> letters, just to make sure they differ from everybody else :-)
Oh, WOW -- that is AWESOME! It would take a bit of getting used to, but, man,
I'll bet you could type real text SO much faster that way! Very nice!
> If you want a really different experience, you should try a Russian keyboard,
> with both Cyrillic characters, and our ones (but on completely different
> locations!)
How about a Russian or Cyrillic keyboard driver running on a U.S. keyboard?
(Oh, my head would hurt!)
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Todd Lehman skrev i meddelandet ...
>
> Hmm, but Pascal uses ^ all over the place (for pointers).
Pointers...??? Who needs pointers? We're talking about Pascal, right? :-)
With the Object Pascal implemented in Borland Delphi, you can get a looong way
without ever using pointers (consciously).
BTW, I'm sure there is a digraph defined for it, I just don't remember which
right now. I know there's (* *) for {}, and (. .) for [ ], so there must be
something for ^ too!
> How about a Russian or Cyrillic keyboard driver running on a U.S. keyboard?
> (Oh, my head would hurt!)
The solution is obvious: small stickers that you put on each key (was included
when I bought a Cyrillic keyboard driver/font set for Win 3.11 back in -91).
(I will *not* set follow-up to off-topic.geek, as I don't download that
group!)
Do svidanija!
--
Anders Isaksson, Sweden
BlockCAD: http://user.tninet.se/~hbh828t/proglego.htm
Gallery: http://user.tninet.se/~hbh828t/gallery.htm
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
On Fri, 3 Mar 2000 21:03:54 GMT Todd Lehman <lehman@javanet.com> writes:
> Oh, WOW -- that is AWESOME! It would take a bit of getting used to,
> but, man, I'll bet you could type real text SO much faster that way!
> Very nice!
What about just using a Dvorak keyboard? I've never seem one, but it is
designed to lessen finger movements across the keyboard, where as QWERTY
was designed to slow people down so their typewriters wouldn't jam...
--Bram
Bram Lambrecht / o o \ BramL@juno.com
-------------------oooo-----(_)-----oooo-------------------
WWW: http://www.chuh.org/Students/Bram-Lambrecht/
-----------------------------------------------------------
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Bram Lambrecht <braml@juno.com> writes:
> [...] where as QWERTY was designed to slow people down so their
> typewriters wouldn't jam...
I don't think the original purpose of QWERTY was to _slow_ people
down. Many mechanical typewriters can operate at hair raising speeds.
What causes the typewriter "hammers" to jam is usually two subsequent
letters from the same area of keyboard being pressed rapidly after
each other. So the purpose of the QWERTY keyboard was, if I'm not
mistaken, to make sure that two subsequent keys normally appear in
different regions of the keyboard.
Fredrik
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.publish, Fredrik Glöckner writes:
> Bram Lambrecht <braml@juno.com> writes:
> > [...] where as QWERTY was designed to slow people down so their
> > typewriters wouldn't jam...
>
> I don't think the original purpose of QWERTY was to _slow_ people
> down. Many mechanical typewriters can operate at hair raising speeds.
> What causes the typewriter "hammers" to jam is usually two subsequent
> letters from the same area of keyboard being pressed rapidly after
> each other. So the purpose of the QWERTY keyboard was, if I'm not
> mistaken, to make sure that two subsequent keys normally appear in
> different regions of the keyboard.
Hmm, I thought the original purpose of the QWERTY keyboard was to make sure
that, while giving a demo, the word TYPEWRITER could be plucked out using
all keys from the top row of letters. ;-)
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.publish, Todd Lehman writes:
> In lugnet.publish, Fredrik Glöckner writes:
> > Bram Lambrecht <braml@juno.com> writes:
> > > [...] where as QWERTY was designed to slow people down so their
> > > typewriters wouldn't jam...
> >
> > I don't think the original purpose of QWERTY was to _slow_ people
> > down. Many mechanical typewriters can operate at hair raising speeds.
> > What causes the typewriter "hammers" to jam is usually two subsequent
> > letters from the same area of keyboard being pressed rapidly after
> > each other. So the purpose of the QWERTY keyboard was, if I'm not
> > mistaken, to make sure that two subsequent keys normally appear in
> > different regions of the keyboard.
Yes, but in the process of making it slow down, you also make it into as
inefficent as possable. (and have the keys as far from one another as is
possible) My mom has a 70 year old _mechanical_ typewriter, and on a good day,
I can still jam it up solid without too much trouble, and I am not the worlds
best typist. Her Electric is much harder to jam up, but it is possible to do
so (key type, not dasy wheel type).
I'd go to a devorak keyboard, but then whenever I was not at home I would be
cursing madly :)
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.publish, Todd Lehman writes:
> In lugnet.publish, Fredrik Glöckner writes:
> > Bram Lambrecht <braml@juno.com> writes:
> > > [...] where as QWERTY was designed to slow people down so their
> > > typewriters wouldn't jam...
> >
> > I don't think the original purpose of QWERTY was to _slow_ people
> > down. Many mechanical typewriters can operate at hair raising speeds.
> > What causes the typewriter "hammers" to jam is usually two subsequent
> > letters from the same area of keyboard being pressed rapidly after
> > each other. So the purpose of the QWERTY keyboard was, if I'm not
> > mistaken, to make sure that two subsequent keys normally appear in
> > different regions of the keyboard.
>
> Hmm, I thought the original purpose of the QWERTY keyboard was to make sure
> that, while giving a demo, the word TYPEWRITER could be plucked out using
> all keys from the top row of letters. ;-)
That's cool! I never noticed that ;-)
-Shiri
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.publish, Fredrik Glöckner writes:
> I don't think the original purpose of QWERTY was to _slow_ people
> down. Many mechanical typewriters can operate at hair raising speeds.
> What causes the typewriter "hammers" to jam is usually two subsequent
> letters from the same area of keyboard being pressed rapidly after
> each other. So the purpose of the QWERTY keyboard was, if I'm not
> mistaken, to make sure that two subsequent keys normally appear in
> different regions of the keyboard.
Just for fun, here are a couple one-liners which spit out a list of words
that can type typed on a single hand using a QWERTY keyboard...
Left hand:
cat /usr/dict/words | grep -i '^[qwertasdfgzxcvb]*$'
Right hand:
cat /usr/dict/words | grep -i '^[yuiophjklnm]*$'
There are some pretty good ones in there. n'joy...
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Todd:
[...]
> Naturally, the plain tilde character, being part of ASCII
> for decades, shouldn't be ruled out as a valid character
> just because it is an extra keystroke on some keyboards.
Assuming we still are talking URL's, it is discouraged/invalid
because there is a significant risk that people get it wrong
if they type it in (which they shouldn't do). You are right
that it shouldn't be necessary to rule it out completely.
> Oh, WOW -- that is AWESOME! It would take a bit of
> getting used to, but, man, I'll bet you could type real
> text SO much faster that way! Very nice!
Not nice at all!
I'm typing text, also in French, much faster with a US
layout (+some private adjustments) than with the French
layout. - And that is no matter whether the actual keyboard
is French or US.
Play well,
Jacob (who wants full Unicode domain names and URL's)
------------------------------------------------------------
-- E-mail: sparre@cats.nbi.dk --
-- Web...: <URL: http://hugin.ldraw.org/LEGO/Skibe/ > --
------------------------------------------------------------
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.publish, Anders Isaksson writes:
> Sometimes I even think the French have done the right thing (for once) with
> their keyboards:
>
> They have the upper row (with numbers and special characters) turned upside
> down, so you use Shift to get the digits, and all the special characters are
> reached unshifted! The numerical keyboard is always normal, so you have both
> digits and everything else without shifting (OTOH, they have swapped a few
> letters, just to make sure they differ from everybody else :-)
That sounds really useful! Although, as Todd mentions, would take a while to
get used to...
> If you want a really different experience, you should try a Russian keyboard,
> with both Cyrillic characters, and our ones (but on completely different
> locations!)
Or try a hebrew keyboard. The english characters are on the same places. But
when you switch the drive to hebrew, you get totally different letters by
typing the same sequence - ie, every key can spew out at least two different
characters, one english, one hebrew. (Every key also has both characters
printed on it. Some keyboards have the hebrew in red for easier typing). When
you are in hebrew configuration and press Shift plus a letter, you get the
same thing as you would get in english (e.g. clicking Shift + 'Alef' results
in a capital T).
But there are a few probs with this. For starters, english has 26 characters
while hebrew has 27 (if you count end letters). So they put one letter on the
":" key. Well, now you can't print a ";" without switching to english and back!
For some bizarre reason, the people who configured the keyboard decided this
needs to be fixed, and fixed it by switching around some letters and
punctuation marks.
So on the comma and period keys (to the right of M, bottom row) there are two
letters, while on the Q and W keys there's a "'" and a slash. Where did the
period go? Instead of the slash key. And the comma? On the "'" key. What about
the ";", which started all the problems? Ah, that goes on the tilde key... ;-)
And to complicate the whole thing, Mac keyboards switch comma and "'" from the
PC keyboards' configuration. But they look the same when printed on the
keyboard.
Ugh!
(Sorry for boring you with these keyboard woes. Me thinks I should go to sleep
now.)
> (Ah, the memories...)
Tell me about it...
-Shiri
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
On Mon, 6 Mar 2000 05:10:07 GMT "Shiri Dori" <shirid@hotmail.com> wrote
concerning 'Re: URL characters':
> But there are a few probs with this. For starters, english has 26 characters
> while hebrew has 27 (if you count end letters). So they put one letter on the
> ":" key. Well, now you can't print a ";" without switching to english and back!
> For some bizarre reason, the people who configured the keyboard decided this
> needs to be fixed, and fixed it by switching around some letters and
> punctuation marks.
>
> So on the comma and period keys (to the right of M, bottom row) there are two
> letters, while on the Q and W keys there's a "'" and a slash. Where did the
> period go? Instead of the slash key. And the comma? On the "'" key. What about
> the ";", which started all the problems? Ah, that goes on the tilde key... ;-)
>
> And to complicate the whole thing, Mac keyboards switch comma and "'" from the
> PC keyboards' configuration. But they look the same when printed on the
> keyboard.
I belive the basic hebrew keyboard configuration comes from hebrew
typewriters - which had the comma and semicolon on the upper left side
(Q and W IIRC), and didn't have the backwards quote (`) at all!
Someone needs to redesign the hebrew keyboard, and make it a dvorak
one too :)
Dan
(yay, finally a real email client, go emacs! :)
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.publish, Jacob Sparre Andersen writes:
> [...]"/ño".
> [...]"ð" ...
I'm guessing these are supposed to be letters + tildes on top. Funny thing,
tho' - on my computer, which has a Hebrew + English system, I see them as
hebrew letters.
I've rarely seen that before - I think the only other time was when someone
sent me email, with one letter that appeared as a hebrew character. OTOH, I
often get emails that are supposed to be in hebrew but arrive as a mass of
unrelated characters ([aeiou] with tildes or accent marks).
Not being a *total* math geek (only a partial one ;-) I'm not really sure what
would cause that. Obviously it has something to do with my system... but what?
Any thoughts?
-Shiri
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Shiri Dori wrote:
>
> In lugnet.publish, Jacob Sparre Andersen writes:
> > [...]"/ño".
> > [...]"ð" ...
>
> I'm guessing these are supposed to be letters + tildes on top. Funny thing,
> tho' - on my computer, which has a Hebrew + English system, I see them as
> hebrew letters.
> I've rarely seen that before - I think the only other time was when someone
> sent me email, with one letter that appeared as a hebrew character. OTOH, I
> often get emails that are supposed to be in hebrew but arrive as a mass of
> unrelated characters ([aeiou] with tildes or accent marks).
>
> Not being a *total* math geek (only a partial one ;-) I'm not really sure what
> would cause that. Obviously it has something to do with my system... but what?
>
> Any thoughts?
>
> -Shiri
No it's not your system..:-) it's just the case of different code pages.
Same extended ASCII codes used for specialized characters of many other
languages at the same time. I mean what you see on the monitor when you
type a special character, say ASCII 135 for example, is completely
depends on which code page is the default one (at your news/mail reader,
at your operating system, or at your keyboards driver, assuming that you
are a MS windows user) when you are looking at the document that
contains this character. If the hebrew support already installed to your
computer, as you mentioned, you can that "mails garbled with special
characters, but supposed to be hebrew" are actually typed in hebrew just
by switching default encoding in your mail reader (i.e. View/Character
set in Netscape -I can't see Hebrew option here..:-( and
View/Encoding/More in MSIE 5.0 -yes I can see Hebrew here..:-)
Selçuk
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Todd Lehman writes:
> Say, I wonder if math-geeks have a strong natural affinity toward custom
> tools like that? Why design a special tool and a custom markup when HTML
> already exists? Answer: Because it's easy, more pleasant to work with for
> lots of semi-complex body text, and hey, making up new special-purpose
> notations from scratch (as appropriate) is just a way of life in mathematics,
> so it's a perfectly normal thing to do.
I actually believe my main reason for doing it was that I had to create
hundreds of pages of HTML, and I didn't trust any of the "HTML authoring tools"
to do what I wanted. Basically, I thought other people's tools would have bugs
or idiosyncracies and I didn't want to deal with the hassle of learning the
tools.
> What did you use for superscript and subscript? Those all come out so nicely
> on your pages!
I didn't mention that because I figured we wouldn't need it for the FAQs. I
denote a superscript like this:
2^{6} to represent 2 followed by a superscript "6"
x^{(x^{(y-1)})} to represent "x" followed by "(x" in superscript,
followed by "(y-1)" in double superscript, followed by ")" in superscript
subscripts are done in a similar way but instead of "^" I use a "v".
For completeness I'll also mention that I have a syntax for tables (another
thing I don't think you'll want to use, but who knows :-). To start and end a
table you use "<table>" and "</table>" (I didn't see the point in making
special symbols for that). You separate table rows with "-" by itself on a
line, and you separate items within a row by "|" with blank space on either
side. Any extra blank spaces next to a "|" get automatically deleted, so if you
want you can make your table line up nicely in ASCII like this (click on "View
Raw Message" or switch to a monospace font, please):
<table>
Part # | Description | Color
-
3001 | 2 x 4 brick | trans red
-
ld3660 | 2 x 2 45^{o} inverted roof brick | blue
-
32133 | *CyberSlam* projectile | teal
</table>
Notice the use of other tags within table items. The sharp sign "#" in the
first item is left alone by my script because it isn't part of a valid boldface
tag (it would need to be adjacent to non-white space and there would have to be
a second # sign somewhere else on the line).
> [...]
> Here are some related (and admittedly biased) thoughts on that general topic,
> from another thread in another newsgroup earlier this month:
>
> http://www.lugnet.com/admin/general/?n=1977
I like where you went in this message, and the use of matching pairs of markers
can be useful to help people see where the bold and italics begin and end. For
example, in my current system, a line like the following is a little hard to
read (although my script has no problem translating it):
*This #sentence* has# overlapping styles.
Also in my system it's easy to inadvertantly leave a * or # out which results
in all the following lines being in boldface or italics. (My script could be a
little smarter to fix this)
The advantage I see of my system is that it leaves [ ] and { } free for other
uses, specifically superscript/subscript and labeled HREFs. In fact, my system
uses almost every character for multiple uses and I suggest that is the way to
go for your system.
> http://www.lugnet.com/admin/general/?n=1983
The only comment I have here is that it would be good to adopt "(r)" and "(c)"
as symbols that get converted into "®" and "©". The "TM" symbol would
of course be "^{TM}" using the existing superscript notation (-:
> http://www.lugnet.com/admin/general/?n=2034 *
When you talk about "upgrading" pages, the comment I have is, if you ever
changed the markup symbols in an incompatible manner, you could convert all the
existing pages to HTML and then convert back to ASCII using the new rules. I
have also written a text-only web browser and of course I used the same
notation to represent web pages as plain ASCII text. The algorithms are just as
simple in either direction.
> http://www.lugnet.com/admin/general/?n=2144
I really like the use of | to denote text in monospace fonts and the <pre>
</pre> type thing. It could actually be added to my system without conflict.
> The big question is: Is something like that sufficiently powerful to handle
> FAQ stuff...? And I think it is...at least for ~99% of anything anyone would
> ever want to write. [...]
I agree -- I'm even reluctant to suggest adding the table operators because I
think it might result in overuse in FAQs. However, since you want to use the
same symtax for member pages, I guess you should include them.
> [...] [for] link syntax [...]
> I was thinking of the exact same character (only with reversed left &
> right portions):
>
> <www.mrob.com|Robert Munafo's Home Page>
I put the "label" portion before the "URL" portion because you read the label
first and *then* go to the URL. But I agree it works almost as well the other
way.
> [...] If something is written as:
>
> <http://www.w3.org/Addressing/URL/5_BNF.html|Uniform Resource Locators:BNF>
>
> then, even though | (vline) isn't part of a valid URL, Netscape mail (and
> likely MSOE as well) displays everything up through the end of the word
> "Uniform" as a hyperlink. It's certainly an unfortunate bug in Netscape's
> mail reader, but for practical reasons on those grounds, it's really worth
> considering something other than | as the separator (for this LUGNET purpose).
You can probably fix this by allowing the user to put blank space on either
side of the | character. You need to allow blank space anyway. If that doesn't
work, try using a different character-pair in place of < >.
> [...] Although a space is somewhat less "clean" and less readable than |,
> it has the advantage that the text-line can be auto-wrapped more nicely
> by typical text editors, [...]
Thats the main reason why I allow blank space.
> [...] my sister is in the process of convincing me not to use
> [[Foo]] and [[[Foo]]] for headings but instead to use
>
> Foo foo foo
> ===========
> [...]
I like those better than my own system. It takes a little more work for the
script, though, because it has to count the "=" characters to figure out what
part of the previous line goes inside <H1>.
- Robert Munafo
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.faq, Todd Lehman writes:
> Say, I wonder if math-geeks have a strong natural affinity toward custom
> tools like that? Why design a special tool and a custom markup when HTML
> already exists? [...]
I should also point out that I have an almost pathological affinity for using
ASCII when about 99.5 % of the sane world would use graphics. For example, see:
http://www.mrob.com/muency/r2c12.html
I actually wrote general-purpose code for converting bitmap images of line art
into ASCII characters so that my illustrations would make sense in browsers
that don't support graphics. The algorithm is partly described here:
http://www.mrob.com/muency/asciigraphics.html
If I had written the LUGNET software the "thread tree" at the end of each
message would have been done in ASCII like this (ya gotta click that "View Raw
Message", dude):
<code>
1-+-2
+-3
+-4-5-**-15-+-16-+-17
| | `-18
| `-6-7
`-8-+-9-10
+-11
`-19
</code>
Good thing I didn't write it, huh [-8
- Robert Munafo
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Robert Munafo wrote:
> If I had written the LUGNET software the "thread tree" at the end of each
> message would have been done in ASCII like this (ya gotta click that "View Raw
> Message", dude):
> <code>
> 1-+-2
> +-3
> +-4-5-**-15-+-16-+-17
> | | `-18
> | `-6-7
> `-8-+-9-10
> +-11
> `-19
> </code>
Ho...ly...smokes. This is cool.
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Todd Lehman <lehman@javanet.com> wrote:
> Wouldn't it be cool if there were something other than HTML that was
> actually readable as plaintext but which also looked great when converted to
> HTML?
I _know_ there's a good smartass answer to this comment somewhere, but I
can't think of it. Very tragic.
--
Matthew Miller ---> mattdm@mattdm.org
Quotes 'R' Us ---> http://quotes-r-us.org/
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.faq, Matthew Miller writes:
> Todd Lehman <lehman@javanet.com> wrote:
> > Wouldn't it be cool if there were something other than HTML that was
> > actually readable as plaintext but which also looked great when converted
> > to HTML?
>
> I _know_ there's a good smartass answer to this comment somewhere, but I
> can't think of it. Very tragic.
Ahh, but it wasn't at all a smartass question. :-) Rather, it was 110%
serious, tossed out as food for thought. There actually are things that
fit the bill above.
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Todd Lehman <lehman@javanet.com> wrote:
> Left hand:
> Right hand:
Interesting:
cat /usr/dict/words | grep -i '^[qwertasdfgzxcvb]*$'|wc -l
1447
cat /usr/dict/words | grep -i '^[yuiophjklnm]*$'|wc -l
187
--
Matthew Miller ---> mattdm@mattdm.org
Quotes 'R' Us ---> http://quotes-r-us.org/
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
In lugnet.faq, "Robert Munafo" <munafo@gcctech.com> writes:
> [...]
> <P>When the formatting looks good, submit your FAQ item to the LUGNET group
> lugnet.faq. The message will be automatically processed by a LUGNET computer
> program. The LUGNET program creates a FAQ answer database entry by starting
> with the following text:</P>
> [...]
Robert, I haven't looked closely yet at the rest of your post, but the above
paragraph is only correct if it's in the future. :) That is, there's no
program yet that goes and collects things from the group. It might not
actually even end up working that way; perhaps instead items will be
submitted to a web page script (where a bit of checking can be done) and
then the item would be auto-posted to the .faq group by the server. It's
still up in the air.
--Todd
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Todd Lehman writes:
> Robert, I haven't looked closely yet at the rest of your post, but the above
> paragraph is only correct if it's in the future. :) That is, there's no
> program yet that goes and collects things from the group. It might not
> actually even end up working that way; perhaps instead [...]
Yeah, I realized that. I just wrote it that way to help point out how abysmally
incomprehensible it was before.
A better approach would be to write the answer in a way that takes into account
the fact that there is no automatic processing yet but explains that there
might be in the future. Also, show by example rather than stating a terse
symbolic specification. And, lead the person through step-by-step: first you
add the headers, then go through and add your text style tags, then convert
your URLs into valid HREFs. Finally, put the entire text, that's the whole FAQ
formatting explanation, inside a PRE block so it's intelligible regardless of
whether or not it's being read in "raw" form.
- Robert Munafo
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
At 12:31 PM 7/13/99 , Sproaticus wrote:
> You want to re-write the answer? :-,
>
> ...or, to quote Bones:
> I'm an engineer, dammit! Not a technical writer! :-D
Heh - but he's a doctor, not an elevator. :)
-Tim <><
http://www.zacktron.com
AIM: timcourtne
ICQ: 23951114
If someone tells you you're worth your weight in gold, fatten up.
| | | | | | | | | | | | | | | | | | | | | | | | |
| |
|
Robert Munafo wrote:
> In lugnet.faq, Jeremy H. Sproat writes:
> > Do you intend to include this FAQ into the grand LUGNET FAQ? (1) If so,
> > they'd need to be put into the proper format. You could do it if you want, or
> > I can get to it when I get more time.
> However, I want to point out that text-only ASCII is the world standard format
> for FAQ files and I'm reluctant to spend a lot of time and effort turning it
> into some elaborate nonstandard format, particularly if it will make the result
> impossible to convert back to plain ASCII.
The format won't make it impossible to convert back to plain ASCII. It
will, however, make it possible to incorporate into a larger, searchable,
multimedia FAQ.
The main rationale behind the FAQ format is to have a common source base
from which different publishing formats can be acheived. For the purposes
of LUGNET, these publishing formats include HTML and plain text.
Thanks for your effort!
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | | | | | |
In lugnet.faq, Jeremy H. Sproat writes:
> The main rationale behind the FAQ format is to have a common source base
> from which different publishing formats can be acheived. For the purposes
> of LUGNET, these publishing formats include HTML and plain text.
(A super-restricted subset of HTML, that is.)
--Todd
| | | | | | | | | | | | | | | | | | | | | | |
Todd Lehman wrote:
> In lugnet.faq, Jeremy H. Sproat writes:
> > The main rationale behind the FAQ format is to have a common source base
> > from which different publishing formats can be acheived. For the purposes
> > of LUGNET, these publishing formats include HTML and plain text.
> (A super-restricted subset of HTML, that is.)
Whoops -- thanks Todd; allow me to clarify.
s/publishing/output/ig
The *final* *output* format of the FAQ may be plain text, it may be DHTML,
or it may be etched in 36-bit EBCDIC-encoded XML in Farsi on stone tablets;
the FAQ format supports many different output types.
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | |
Okay Jeremy, and anyone else who is interested --
I have now reformatted my FAQ answers and merged the content with the answers
written by others and recently auto-reposted. I then tested the results for
proper HTML formatting in my browser and then posted the answers as seperate
messages to this newsgroup. Then I compiled an index of all the message URLs
and posted it as a FAQ table of contents which appears as message
http://www.lugnet.com/robotics/rcx/?n=37
Let me know what you think of the formatting. I think I got it right but I was
really just copying what I saw in the other FAQ answers.
- Robert Munafo
In lugnet.faq, Jeremy H. Sproat writes:
> Do you intend to include this FAQ into the grand LUGNET FAQ? (1) If so,
> they'd need to be put into the proper format. You could do it if you want, or
> I can get to it when I get more time.[...]
| | | | | | | | | | | | | | | | |
Robert Munafo wrote:
> Let me know what you think of the formatting. I think I got it right but I was
> really just copying what I saw in the other FAQ answers.
I haven't gone through all of them, but what I've seen so far has been
really good. Thanks again!
One issue (and this may be a non-issue) is the usage of <BLOCKQUOTE> -- has
there been a consensus on the list of usable HTML tags?
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | |
In lugnet.faq, Jeremy H. Sproat writes:
> One issue (and this may be a non-issue) is the usage of <BLOCKQUOTE> -- has
> there been a consensus on the list of usable HTML tags?
Thanks for pointing that out Jeremy. I had to guess on BLOCKQUOTE. I saw it
referred to only in one place, the following suggestion by Todd Lehman:
http://www.lugnet.com/faq/?n=194
and I assumed that it would be okay to use BLOCKQUOTE because there's no other
easy way to do block quotes in HTML. But if you want, just take all the
BLOCKQUOTE and /BLOCKQUOTE tags out and the result should still format well,
although the URLs and maxims will no longer be indented.
I tried to get more information about the use of BLOCKQUOTE, but was frustrated
by the Lugnet search engine's refusal to locate the above message when
BLOCKQUOTE was searched for. I have submitted a question about this to
lugnet.admin.general: http://www.lugnet.com/admin/general/?n=2220
- Robert Munafo
| | | | | | | | | | | | | | | | | | | | |
In lugnet.faq, "Robert Munafo" <munafo@gcctech.com> writes:
> I tried to get more information about the use of BLOCKQUOTE, but was frustrated
> by the Lugnet search engine's refusal to locate the above message when
> BLOCKQUOTE was searched for. I have submitted a question about this to
> lugnet.admin.general: http://www.lugnet.com/admin/general/?n=2220
Sorry about that, Robert... I don't know what I was smoking when I told the
indexer to filter out HTML tags.
--Todd
| | | | | | | | | | | | | | | | | | |
In lugnet.faq, Sproaticus <jsproat@io.com> writes:
> One issue (and this may be a non-issue) is the usage of <BLOCKQUOTE> -- has
> there been a consensus on the list of usable HTML tags?
I think it doesn't hurt to allow <BLOCKQUOTE> -- but I'm having trouble
imagning places where it would be useful. Heh heh, maybe in acutal block
quotes? :-)
"Four score an seven years ago . . .
--Todd
| | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Todd Lehman writes:
> I think it doesn't hurt to allow <BLOCKQUOTE> -- but I'm having trouble
> imagning places where it would be useful. Heh heh, maybe in acutal block
> quotes? :-)
Well, if you look at my FAQ submissions (and you'll have to save them as files
and open in a web browser to see the formatting) you'll see that I put
BLOCKQUOTE and P around all my URLs to make them stand out. Here are a couple
examples of that:
http://www.lugnet.com/faq/?n=594
http://www.lugnet.com/faq/?n=592
I also use it in a couple places to make an important piece of advice stand
out. Here is an example of that:
http://www.lugnet.com/faq/?n=554
I think BLOCKQUOTE is useful in limited situations. It's sort of a stylistic
(artistic or judgmental) thing. I happen to think URLs should be set apart from
the text of a paragraph rather than embedded in the text the way most Web
authors nowadays do them. I'm not opposed to doing it that way, but it does
make the grammar a little more clumsy, particularly considering the requirement
that the URL has to appear as plain text to benefit users who for whatever
reason can't just click on the link and jump to the referent page.
- Robert Munafo
| | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Robert Munafo writes:
> In lugnet.faq, Todd Lehman writes:
> > I think it doesn't hurt to allow <BLOCKQUOTE> -- but I'm having trouble
> > imagning places where it would be useful. Heh heh, maybe in acutal block
> > quotes? :-)
>
> Well, if you look at my FAQ submissions (and you'll have to save them as
> files and open in a web browser to see the formatting) you'll see that I
> put BLOCKQUOTE and P around all my URLs to make them stand out. Here are
> a couple examples of that:
>
> http://www.lugnet.com/faq/?n=594
> http://www.lugnet.com/faq/?n=592
I think it's nice how those are set off, but I see a couple problems:
1. there is accidental space around the 'foo' in
<a href="xxx"> foo </a>
which will cause the links to display improperly. Really it should say
<a href="xxx">foo</a>
for proper links.
2. All of the URLs are written out twice. Is that really necessary?
3. In the first example (article #594), why not something like this
instead?--
<p>To get either of these sensors, go to the
<a href="http://www.legoworldshop.com/">LEGO® World Shop</a>, then
select your region and country, then choose LEGO® MindStorms.</p>
<p>For the DACTA online store, go to the
<a href="http://www.pitsco-legodacta-store.com/">Pitsco LEGO DACTA Online
Store</a>, then select "Spare Parts," then select "Sensors."</p>
> I also use it in a couple places to make an important piece of advice stand
> out. Here is an example of that:
>
> http://www.lugnet.com/faq/?n=554
Ahh, that's a perfect example, IMHO. :) Alternatively, though, couldn't
it be written in italics and carry the same mental-weight?
> I think BLOCKQUOTE is useful in limited situations. It's sort of a
> stylistic (artistic or judgmental) thing. I happen to think URLs should
> be set apart from the text of a paragraph rather than embedded in the text
> the way most Web authors nowadays do them. I'm not opposed to doing it that
> way, but it does make the grammar a little more clumsy, particularly
> considering the requirement that the URL has to appear as plain text to
> benefit users who for whatever reason can't just click on the link and
> jump to the referent page.
Yeah, that definitely sounds like sound rationale. I agree, and always write
URLs like this as well:
This is the most amazing M-Set explorer I've ever seen:
http://www.fsf.org/software/xaos/xaos.html
But (I think) it really makes the most sense only when, like you said, when
the URL has to appear as plain text for whatever reason. On that note, lemme
go respond later tonight to your other message on ASCII->HTML conversion,
which I agree in spirit is really the right approach rather than raw HTML.
--Todd
| | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, Todd Lehman writes:
> 1. there is accidental space around the 'foo' in
>
> <a href="xxx"> foo </a>
>
> which will cause the links to display improperly. Really it should say
>
> <a href="xxx">foo</a>
>
> for proper links.
Accidental to you maybe, but it was no accident to me. I left space on either
side of anchor text to make it more readable. I did that because right now
everyone is limited to reading raw FAQ entries.
> 2. All of the URLs are written out twice. Is that really necessary?
Yes. If the formatting and unformatting tools are written, they will strip out
HTML tags when generating the plain-text FAQ. In order for the URL to be in the
plain-text FAQ it has to appear as plain text.
> 3. In the first example (article #594), why not something like this
> instead?--
>
> <p>To get either of these sensors, go to the
> <a href="http://www.legoworldshop.com/">LEGO® World Shop</a>, then
> select your region and country, then choose LEGO® MindStorms.</p>
> [...]
I guess I just answered that (-: The URLs have to appear explicitly, and
explicit URLs as direct or indirect objects in prose are grammatically awkward.
> > http://www.lugnet.com/faq/?n=554
>
> Ahh, that's a perfect example, IMHO. :) Alternatively, though, couldn't
> it be written in italics and carry the same mental-weight?
I wanted it to appear seperate from the discussion, with actual white space on
all four sides. In other words, I prefer the way I did it only on personal
aesthetic grounds.
> [...] But (I think) it really makes the most sense only when, like you
> said, when the URL has to appear as plain text for whatever reason. [...]
Yeah, it's clearly an aesthetic argument and I think the majority (which
includes you) will eventually win. Which is as it should be (-:
I just thought of another idea though, which is to do "normal" links in the
body of the text and have a list of explicit URLs at the end, kind of like a
footnotes list. The list of URLs would be to benefit those users who can't
simply click on the links, which includes users who read it as a text-only FAQ.
- Robert Munafo
| | | | | | | | | | | | | | | | | |
Okay, I've got a little more on the BLOCKQUOTE issue now. I had to load the
entire newsgroup on my local machine and use my special MIRA tools to search it
in order to figure this out.
Here are the only non-Robert-Munafo FAQ answers that use BLOCKQUOTE:
http://www.lugnet.com/faq/?n=228
http://www.lugnet.com/faq/?n=452
Aside from the message by Todd that I referenced before:
http://www.lugnet.com/faq/?n=194
there are no other messages in this newsgroup that include the word
"BLOCKQUOTE" in any form and were posted prior to when I started submitting my
FAQ items.
- Robert Munafo
| | | | | | | | | | | | | | | | | | |
Robert Munafo <munafo@gcctech.com> wrote:
> > 2. All of the URLs are written out twice. Is that really necessary?
> Yes. If the formatting and unformatting tools are written, they will strip
> out HTML tags when generating the plain-text FAQ. In order for the URL to
> be in the plain-text FAQ it has to appear as plain text.
Couldn't these theoretical tools be a little smarter?
--
Matthew Miller ---> mattdm@mattdm.org
Quotes 'R' Us ---> http://quotes-r-us.org/
| | | | | | | | | | | | | | | | | | | | |
Matthew Miller:
> Robert Munafo <munafo@gcctech.com> wrote:
> > > 2. All of the URLs are written out twice. Is that really necessary?
> > Yes. If the formatting and unformatting tools are written, they will strip
> > out HTML tags when generating the plain-text FAQ. In order for the URL to
> > be in the plain-text FAQ it has to appear as plain text.
>
> Couldn't these theoretical tools be a little smarter?
The _actual_ tool (lynx) is!
Play well,
Jacob
------------------------------------------------
-- E-mail: sparre@cats.nbi.dk --
-- Web...: <URL:http://www.ldraw.org/FAQ/> --
------------------------------------------------
| | | | | | | | | | | | | | | | | | | | |
| |
|
In lugnet.faq, "Robert Munafo" <munafo@gcctech.com> writes:
> In lugnet.faq, Todd Lehman writes:
> > 1. there is accidental space around the 'foo' in
> >
> > <a href="xxx"> foo </a>
> >
> > which will cause the links to display improperly. Really it should say
> >
> > <a href="xxx">foo</a>
> >
> > for proper links.
>
> Accidental to you maybe, but it was no accident to me. I left space on either
> side of anchor text to make it more readable. I did that because right now
> everyone is limited to reading raw FAQ entries.
Aha! OK, forgive me -- I assumed it was either an accident or some penchant
for doing that sort of thing. Looking at your web pages (which I love,
BTW!), of course, you don't have anything weird at all like that there, so
that ruled out the penchant thing. :-)
> [...]
> > > http://www.lugnet.com/faq/?n=554
> >
> > Ahh, that's a perfect example, IMHO. :) Alternatively, though, couldn't
> > it be written in italics and carry the same mental-weight?
>
> I wanted it to appear seperate from the discussion, with actual white space on
> all four sides. In other words, I prefer the way I did it only on personal
> aesthetic grounds.
Either way is fine by me; just wanted to point out the italics option in
that case, since the text was so short. :-)
--Todd
| | | | | | | | | | | | | | | | | |
Robert Munafo:
> > 2. All of the URLs are written out twice. Is that really necessary?
>
> Yes. If the formatting and unformatting tools are written,
> they will strip out HTML tags when generating the plain-text
> FAQ. In order for the URL to be in the plain-text FAQ it has
> to appear as plain text.
The HTML-to-text converter already exists (lynx -dump), and it
inserts footnote-like[1] references in the text, so there
isn't a serious need for repeating URL's as plain text.
Play well,
Jacob
------------------------------------------------
-- E-mail: sparre@cats.nbi.dk --
-- Web...: <URL:http://www.ldraw.org/FAQ/> --
------------------------------------------------
| | | | | | | | | | | | | | | | |
In lugnet.faq, Jacob Sparre Andersen writes:
> The HTML-to-text converter already exists (lynx -dump), and it
> inserts footnote-like[1] references in the text, so there
> isn't a serious need for repeating URL's as plain text.
Wow, that's great! That's actually the perfect thing to do. Somehow I
was led to believe someone was writing all the LUGNET tools themselves and that
everything inside angle brackets would be invisible.
So, let's see -- that means I can rewrite the FAQ answers I submitted to make
the URL links have labels, link "DACTA home page". That just leaves the issue
of readability. For now I'd like to leave the URL links set apart in their own
paragraphs because people still have to read these things in raw format, and it
really screws up a well written paragraph to have HTML tags in the middle.
So, this is mainly directed to Jeremy I guess: should I do any editing of the
links and resubmit?
- Robert Munafo
| | | | | | | | | | | | | | | | | |
Robert Munafo wrote:
> So, this is mainly directed to Jeremy I guess: should I do any editing of the
> links and resubmit?
Nah. Don't worry about it.
I'm taking the position that individual FAQ items written or edited by an
individual might as well carry some of that individual's stylistic
techniques -- "stylistic techniques" to me means things like writing style,
punctuation, etc. and ASCII formatting. As long as it falls within certain
social, LUGNETtiquette, and language norms, I'll just leave it alone.
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | | | |
| |
|
Robert Munafo wrote:
> In lugnet.faq, Jacob Sparre Andersen writes:
> > The HTML-to-text converter already exists (lynx -dump), and it
> > inserts footnote-like[1] references in the text, so there
> > isn't a serious need for repeating URL's as plain text.
> Wow, that's great! That's actually the perfect thing to do. Somehow I
> was led to believe someone was writing all the LUGNET tools themselves and that
> everything inside angle brackets would be invisible.
Actually, we were doing just that. We were ready to come out in the open
with our stealth ASCII technology, but then the USAF needed some of it to
bomb someone, and we couldn't talk about it for at least until the media
stopped getting excited.
BTW, this message contains a very nasty stealth ASCII computer virus. It
spreads itself by mailing everyone in your e-mail address book and your
phone book for good measure. Its effect: it'll infect everyone and their
computer and make them knuckle down and learn how to use LDraw. Oh, and
it'll run up to $500,000 US in phone sex calls to China, and it doesn't dial
10-10-220.
So, if you get the overwhelming urge to visit www.ldraw.org or if your
computer case gets purple and splotchy, respond to this message for the
cure.
Cheers,
- jsproat
--
Jeremy H. Sproat <jsproat@io.com>
http://www.io.com/~jsproat
Darth Maul Lives
| | | | | | | | | | | | | | | | |
Hey, while you're at it, will you write one that makes people buy LEGO sets and
send me just the 2 x 2 tiles? They can use an envelope or a box, I don't care.
In lugnet.off-topic.fun, Jeremy H. Sproat writes:
> [...]
> BTW, this message contains a very nasty stealth ASCII computer virus. It
> spreads itself by mailing everyone in your e-mail address book and your
> phone book for good measure. Its effect: it'll infect everyone and their
> computer and make them knuckle down and learn how to use LDraw. Oh, and
> it'll run up to $500,000 US in phone sex calls to China, and it doesn't dial
> 10-10-220.
> [...]
| | | | | | |
