Posted by robert maas, see http://tinyurl.com/uh3t on 02/18/07 17:37

> From: John Hosking <J...@DELETE.Hosking.name.INVALID>
> I am not suggesting that pages (of the OP or anyone else) need to
> be slick, snazzy, flashy, Flashy, or animated. I require no
> particular abundance of color. Simple is just fine with me, and
> has been from about the fifth week of my experience with the web.

Good, we're in agreement to this point.

(Referring to me, the OP:)
> he was giving what I saw as excuses for why he couldn't possibly
> *know* what IE users saw

More like it takes hours/days/weeks between the time I make a
chance to my Web page until I have a chance to get to some place
where there's IE so that I can see what it looks like i IE. That's
too long a change-test feedback cycle to be of any use whatsoever
in development, so basically when I'm writing/adding/changing my
Web pages it's impossible to adjust my methods per what it might
look like in IE. I just have to check what I write against how it
looks in lynx and what the validator says (each available within a
minute or less any time I want to check my work), and just *hope*
it will come out right in Mozilla and IE and NetScape too.

So-far in fact I haven't had even one chance to look at any of this
in any of those browsers, because during the rare times I had
access to the other browsers I had much more urgent matters to do
during the limited time allowed (1.5 hours at Connect, or 1 hour if
a terminal happens to be available at public library, each a
half-hour bike ride from here, or half hour at a new place that
opened up much closer just a few days ago but has only one working
computer that has to be shared so I have to go there and sign up
for future slot then go back there later). But I finished most of
those more-urgent matters Friday, so hopefully on Monday I'll
finally be able to take a look for the first time.

> because he only has the one computer

I have only one computer with working modem that would allow upload
or download of text, but both upload and download are very clumsy:
To upload, I start a 'cat' command on Unix, input from terminal and
output to file, copy some text from a text editor on my Mac, and
paste it into the VT100 emulator, watch it spew out with a
half-second pause after each newline, and finally when it's done I
press ctrl-D to end the transfer. Copy&Paste is limited to 30k
bytes at a time due to limit in MacOS. To download, I start a
'more' command on Unix, sit there pressing space manually over and
over until it's reached the end, then scroll back the VT100
emulator to find where the 'more' started, sweep the text with the
mouse to bottom of screen, then rest mouse on bottom of screen as
VT100 emulator auto-scrolls one line at a time, until I reach the
end (assuming it's less than 30k total), then copy what I've swept,
and paste into local text editor. If I want to up/down load more
than 30k at a time, I have to break it into <30k chunks. For
example, on upload I start the 'cat', sweep part of the local file
and use the "Statistics" command to verify my sweep is less than
30k, the copy that and paste, repeat until last chunk done, finally
ctrl-D. To download more than 30k, if the whole file will fit in
the VT100 emulator's scrollback buffer then I use 'more' to get the
whole file, but then I have to guess how much of what I did would
be less than 30k chunk, repeat that guessing, until I reach the
end. If it's too much for VT100 emulator's scrollback buffer, then
I have to run 'more' only partway through the file, do my sweep,
run 'more' some more, do another sweep, etc., which is so much
trouble that it's not really feasible to get correct.

Kermit is available, but it doesn't map newlines between Mac and
Unix, so I have to go into emacs and manually convert from Unix to
Mac before download or from Mac to Unix after upload, which isn't
worth the trouble except for really large files where breaking into
30k pieces is more pain than manually converting newlines.

> with what I presume to be 5 1/4" single-sided floppy drives

No, never had one of those pieces of crap. You are deliberately being
derogatory/insulting at this point.

> and a flaky 300 baud modem.

No. I used one of those for about ten years before I got access to
a 1200/150 split modem, which I used for a few years before I got
access to a 2400/2400 modem, which I used for a few years before I
bought this FAX/data modem which goes up to 56k but a different
standard from what PPP uses nowadays so the one time I tried PPP I
had to use 28k, but the VT100 emulator is limited to 19200 so
that's what I use all the time when on shell dialup.

> I inferred that since he couldn't *possibly* know what graphical
> browsers showed he thought he didn't need to *care*.

There's an old saying, which I'll re-phrase to avoid the stupid
religious aspect of the original: A good policy is to change what
can be changed, accept with what can't be changed, and know the
difference between the two. I know the difference between
interactive checking with lynx and validator, which I *can* do, and
interactive checking with IE/Mozilla/NetScape, which I can *not* do
at this time or in the foreseeable future, so I do the former and
accept lack of the latter. You seem to have it in your mind to
squeeze blood out of a turnip, and can't accept that it's
impossible. You seem to be insistant that I do something that is
beyond my present means, namely have a more expensive computer
system than I can afford, and use it to develop Web pages with
interactive use of IE/Mozilla/NetScape during development, and
won't be happy until you see that blood dripping out of that
turnip.

Do you *care* that we're all going to die, our whole species go
extinct, all life on Earth wiped out, in about one billion years
when increase energy from the Sun boils away the oceans? What are
*you* doing right now to prevent that disaster??

If you can't do anything about something, what's the point of
arguing whether you care or don't care about it?? It's a
methaphysical question if you can't do anything about it. Whether
you care about billion-year-hence extinction, or whether I care how
my Web pages look in IE, is moot since neither of us can do
anything about either. Now if somebody with access to IE would take
a look at my Web pages every few minutes while I'm actively working
on them, and send me an instant alert whenever I create something
that passes validation yet looks like crap in IE, that would allow
me to promptly fix anything I break before I go on to create ten
more copies of the same ugly format. But somebody volunteering to
help me like that ain't gonna happen so I'm not gonna worry about
it.

> None of that would matter if he wasn't getting all uptight about
> what *his* Lynx browser ... on his VT100 did spatially with the
> <pre> element.

Well I'm just trying to get a flowing paragraph with a code snippet
in the middle of it. Apparently <pre> wasn't designed to do that,
so I'm stuck for a way to do it per HTML standard. <pre> almost
does it, sigh. It's so close, it makes me think it *should* do it
right. So if you want to include code snippets in the middle of
running text, how do you do it? For example, if you wanted an
effect somewhat like this:

Now once you somehow get lisp "interpretor" started, it'll probably
print out some banner identifying what version it is, then it'll enter
the read-eval-print loop. So you just type an integer literal on one
line, for example 42, and then press ENTER or RETURN, whatever the key
is on your computer for finishing lines of input, and lisp will print
that number 42 on a new line and then wait for further input. So
suppose you enter two numbers, with a space between them, such as
42 69
on a single line of input? Lisp will read the 42, and print it out
on a new line. Then lisp will read the 69 and print it out on another
new line. Then lisp will see you haven't typed any more input, so lisp
will wait for new input. If you have access to lisp, you should try
that right now.

The basic idea is that there's continuous flow of text, a single
"paragraph", but there's one line set out specially in the middle
of it, not wrapped automatically. How to achieve that effect? I'm
talking about semantic effect, line set out in middle of paragraph,
not the precise way it's formatted. Just that however it's
formatted, it shouldn't look like there are two separate
paragraphs. Here's another example, a long transcript (only part
shown here) with running commentary alternating with the
transcript. From time to time when the topic changes there's a new
paragraph started, but so long as the topic is the same it should
be just one big paragraph with snippets of transcript set out at
various places in the middle:

[First paragraph:]
Now I translate the single line of code in the body of the function.
(Note I don't need a return statement in lisp.)
(* n n)
9
Good, I got it right.

[Second paragraph:]
Now to build my function definition around that single line of code:
(defun sq (n)
(* n n))

[Third paragraph:]
Now to unit-test the function to see if it works.
(sq 4)
16
Good, it works.

[Fourth paragraph:]
Next I want to translate the function showsq into lisp. First I
translate each line separately. Because I used the same variable
name n, I already have a test value set up there. I check it just
to be sure:
n
3

[Fifth paragraph, very long, not all of it shown here:]
The first line of code to translate is: res = sq(n); which translates
to
(setq res (sq n))
9
Working, next to translate: printf("The value of ");
(princ "The value of ")
The value of
Working, next to translate: printf("%d", n);
(princ n)
3
Working, next to translate: printf(" squared is ");
(princ " squared is ")
squared is
...Working, next to translate: printf("%d", res);
(princ res)
9
Working, next to translate: printf(".\n"); which requires two lines
of code in lisp the way I've been doing up to now ...

Do you see how that whole bunch of transcript-with-commentary is
only five paragraphs (that fifth one rather long with repeated
set-out segments of transcript)? Ignoring how it *looks* in the
various browsers, what's the correct *logical* organization of that
text per HTML standard?? If I get the HTML correct, then I expect
all the various browsers will render in some way that is consistent
with that logical organization, *not* in a way that indicates some
*other* organization to the user.

> I admit to a withering despair concerning our brother Robert and
> I thought the best advice would be for him to concentrate as best
> he can on one thing (his text) before shifting his attention to one
> other thing (appropriate, valid markup).

But I'm not writing a novel, I'm writing a matrix, a hierarchial
table of contents linking to cells of informational text. From the
very start it has to be HTML, and it has to be online so that I can
check each new link I put in to make sure it really works, that it
takes the user from the clicked label in the table of contents to
the appropriate cell of informational text. And for the benefit of
anyone who might stumble into it expecting to get an answer to a
question, like how to convert an ASCII character code to the actual
character object, or how to search a string for the first character
*not* among a specific bag of characters, all items in the table of
contents for which there is no content yet are obvious stubs, just
[Topic] (plain text), instead of HREFs that go to blank pages or
even worse, so the user can see at a glance in the table of
contents which topics are covered and which aren't before clicking
on anything. On the other hand, I've been filling in specifics for
only c and lisp so-far, and c++ when it's identical to c, so
anybody looking for differences of c++ from c, or anything in java
or perl or PHP, will be disappointed for the moment, sorry. Maybe I
should put in a "status so-far" section with such general warnings?

> ... he'll probably want to spend time "testing" the look of these
> on his paleo-browser and, potentially, browsers that the rest of
> the planet might use.

No, "paleo" browser is what I found in the campus library at SCU
back around 1967: It was a "programmed text", where you read the
text on the first page and the see alternatives as to what page to
turn to next. It gave me an idea for doing the same sort of thing
for organizing information in a hierarchial way on a computer,
which I implemented circa 1970-71 on the PDP-10 at Stanford A.I.
lab. Apparently Richard Stallman at MIT got the same idea about the
same time, and implemented INFO mode in EMACS. I don't know whether
he was also inspired by seeing an old "programmed text" book, or
whether he thought of the idea originally.

Almost-"paleo" technology would be that mechanical card sorter I
saw used between 1964 and 1967, where you can sort on only a single
column at a time, so you sort on least significant column first,
then work your way toward most significant column last.

Half-"paleo" would be that first access to the Web that I ever had.
I'd find a URL somewhere in a newsgroup, and e-mail that URL to a
lynx-by-email service, and get back in a few hours or a day or so
the lynx-rendered page, with [number] for each link, and mapping
from [number] and URL at bottom. Then I'd find which URL I wanted
to follow next, and do another e-mail cycle. Often I'd get no
response after several days, so I'd have to try again. For a year
or two that's the only Web access I had. Being able to run lynx
live from my own shell account is sooooooo much better!!

> I agree wholeheartedly that such a page could be far more useful
> and information-rich than the common crud. I hope his work is
> fruitful.

<GomerPileUSMC>Surprise, surprise, surprise</GomerPileUSMC>
So you like the general idea of a matrix coordinating different
data types, or the general idea of a "cookbook" that showed the
same task in several common languages side-by-side, or both ideas,
or just the idea of non-swirly organized information, or ...?

(Hmm, I think we've coined a new slang term, "swirly", which sounds
rather like "gnarly" and has similar meaning but specifically deals
with flashy visual effects swirling all over the Web page
distracting from the actual content, rather like the flashing
swirling lights all over the "midway" at a carnival/circus/fair.)

Before posting this I checked Google to see if the word was already
in use, yes it is, with different meanings, such as bogwash, and
something having to do with a prostitute or artist but I can't
figure out the precise meaning, plus the literal meaning applied to
Adobe Illustrator. No problem. It's OK to have the same slang word
with more than one meaning. At least *our* meaning is new, not
already in use. So next time somebody shows us their flashy
"wonderful" Web site, we can simply dismiss it as "too swirly for
my taste", as we lean over sideways, looking like we're going to
get dizzy and either fall over or vomit or both.

[Back to original message]