Re: What Software to Type Math In?

In article <45hekvF6sb6oU1@xxxxxxxxxxxxxx>,
Marc Olschok <invalid@xxxxxxxxxxx> wrote:
Herman Rubin <hrubin@xxxxxxxxxxxxxxxxxxxx> wrote:
In article <45c6bbF5m2p8U2@xxxxxxxxxxxxxx>,
Marc Olschok <invalid@xxxxxxxxxxx> wrote:

Herman Rubin <hrubin@xxxxxxxxxxxxxxxxxxxx> wrote:
Marc Olschok <invalid@xxxxxxxxxxx> wrote:
Herman Rubin <hrubin@xxxxxxxxxxxxxxxxxxxx> wrote:
[...]
What is "plain text"?

Something that I can process with simple editors and other text-tools
across a variety of different platforms, independent of the particular
fonts or encodings available. Something I can e-mail or post in newsgroups
without having to assume any special capabilities on the part of other
readers.

If you post TeX, the reader will need a TeX compiler and
reader. If you post LaTeX, a little more. If you post
what we call "mathtex", which leaves off some of the
TeX punctuation, the reader needs to know enough TeX to
be able to tell what you are doing. To do fully what you
want is not possible.

The reader will need TeX or LaTeX to _typeset_ the document and
some viewer/printer in order to look at the result.
But in order to read or manipulate the documents source, no TeX is needed.
This is the point of using plain text for _the_source_ instead of
some format, which can only be manipulated with some special program.
And (in the context of my above remark) of course I know what I do.

If it is complex, the reader will need that to know what
the document intends. Otherwise, it is like someone who
knows no Russian, but knows the alphabet and how to pronounce
it, reading Russian.

If I often cannot find an error in a TeX file without
compiling and reading it with a viewer or printer, I
am unwilling to say that I am reading the document.

In short: I want to be able to use ed in a dumb terminal, if needed.
I admit that such a notion of "plain text" is a moving target: right now
it is still ASCII for me; in the near future, when all those tools will
have migrated, it will perhaps be Unicode.

To do things with ASCII, you may be able to get away with
the clumsy diagramming of characters; this is not enough.

Well, this is why markup languages (as html and TeX) are so popular.

And we have our emails cluttered up with thousands of markup
characters and little content. I am interested in getting the
information efficiently, without all that clutter. Markup
languages are messy.

Actually further down in this posting you provide a perfect example
why it is sometimes a good idea to restrict to ASCII.

[...]
You want an Unicode-capable editor.

I want an editor capable of a sufficient amount of Unicode
in fixed-width type. I have not seen such.

Did you try Yudit? (e.g. <http://www.yudit.org/>)

I have no experienc with it, but it may be a little step towards
your need.



The old Apples has a way to put the typewriter decoding for
a particular font in a corner of the screen. One did not
have to mouse the character in, but knew how to type it.

And one needed an Apple to begin with.

There were other PCs with the capabilities. The computer people
tell me that what I want is "trivial"; the people producing the
software insist on giving me features which I do not want, and
not keeping it simple.

There was an old computer with one very bad design feature
which had fixed-size characters, and allowed the used to
implement 16x10 characters in various fonts, and to change
characters to suit. This was with a memory of 640 kbytes.

I fear the two groups interpret your question in different ways.
When you mention "type mathematics" the latter group will think of
"typesetting mathematics" just as I did before, when I failed to take
the original phrase literally. But, as you pointed out you do not want
or need this.


[...]
Another advantage of at typewriter rather than a typesetter
is that the author has easy control of line breaks. Also,
fixed width fonts are necessary for easy communication.
This goes completely against the typesetting mentality.

Depending on the final format it might also go completely against
the idea of readability. Nobody suggested that you typeset your e-mail.

I still use Berkeley mail for sending email or responding.
The email with this newsreader is like that as well. Email
sent by many of the fancier mailers lacks line breaks, and
can be difficult to handle.

There is no disagreement here. I also use ordinary text terminals
with fixed-width fonts and hard line breaks, but only wanted to remark
that different output media may have different demands.

Do they have different demands? Or are they just trying
to set up a user-"friendly" system which at least a very
large part of their users find to be an overly intricate
mass of arbitrary devices?

But here is the promised example you provided in favour of
plain ASCII:

Sure. The same point could be made against special fonts in mail.
Compare "Ji visited Wrzburg" to "Ji\v{r}\'\i\ visited W\"urzburg".
Observe, that I can type the left version directly into LaTeX.

If that right side is supposed to be LaTeX, I cannot read
it. what does the string of characters \v{r}\'\i\ mean? I
do know the TeX way of indication umlauts, but I do not
like it. I would not mind the European typewriter version
of having such things as umlauts and accents as non-spacing
superimposed characters.

The \v puts a "czech hook" accent over its argument, \' puts an / like
accent over its argument (which here indicates a long vowel), \i is a
dotless i and "\ " was just a leftover from the TeX source in order to
ensure a space (see e.g. the TeXbook). As I said, LaTeX can work with
different input encodings, so you would not need type these TeX makros.

But the interesting question is:
could you read the equivalent text on the left side?

I do not see anything corresponding to the hacek-r or
dotless i with an accent on the left side, or the u-umlaut
in Wurzburg, although I guessed that that was meant.

However, my interest is not in producing linguistic text,
but mathematical notation added to English, for notes or
class assignments, or even for drafts of articles. So I
need at least the Latin and Greek alphabets, "standard"
mathematical symbols, and I am annoyed that {}\ are not
available for them, a convenient way to handle subscripts
and superscripts, and half lines would do well for that
even without smaller fonts. Bold face and such should be
handled by shifts, which may require typing escapes, but
would still appear as single characters (of different
lengths in binary) in the text file, the whole thing in
fixed width characters, so if I want to line things up I
have no problems. If I have to type "\alpha" to get a
small Greek alpha, I only want to see the small Greek
letter on the screen.

Of course the headers of my postings include the necessary information
about charset and encoding (yours do _not_, which is tolerable as long
as you post within the ASCII subset of iso-8859).

But a terminal that can handle only ASCII will not display the indicated
characters correctly. A terminal that can handle iso-8859-1 will display
the german umlauts and the long i but will incorrectly display a
scandinavian \o (o with a slash) instead of the hooked r.
You would need an iso-8859-2 capable terminal in order to view it correctly,
which is why I rarely post outside ASCII in english speaking newsgroups.

Moreover, your newsreader has lost these characters completely at least
when you incorporated the quote in your response.

Now why should the characters have been completely lost?
I consider that a major flaw; at least give another
rendering. At least indicate that there is a character
there. I often see characters which I can tell are
definitely not what they look like on the screen, as that
is nonsense.

What _has_survived_ is the TeX source, which is plain ASCII.

Marc


--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
hrubin@xxxxxxxxxxxxxxx Phone: (765)494-6054 FAX: (765)494-0558
.

Relevant Pages

  • Re: What Software to Type Math In?
    ... characters when 10 will do is not a good idea. ... replace fonts; also, the user could even modify characters in ... division of labour in the process of typesetting your document. ... What is "plain text"? ...
    (sci.math)
  • Re: NFSS2 and math mode
    ... in bold will be only x. ... available fonts could solve the problem) but the b) require further ... read the TeXbook or TeX by Topic. ... TeX knows "alphabetic" characters which are selected ...
    (comp.text.tex)
  • Re: Why does German favor long compound words?
    ... Pulleyblank's EMC/LMC/EM list with Old Chinese readings added, ... Unfortunately, they don't seem to have produced a Unicode version of their fonts, and I don't think the Mac OS X input routines work with Shift JIS any more. ... You can generate a listing of all the gif versions of the characters they have on the site fairly simply if you analyse the directory numbers and gif numbers carefully. ...
    (sci.lang)
  • Re: Letter Punch Even Spacing - How?
    ... First of all you need you need to ensure a succession of characters is ... fonts: proportional and fixed pitch fonts. ... again referenced to clamped steel rule. ...
    (uk.rec.models.engineering)
  • Re: unable to insert symbol with latest version of Office
    ... Use Character Palette to do the inserting - and you get ... > literally thousands more characters to play with. ... > Lucida Grande if the standard Unicode Microsoft fonts ... Is there any way on a Mac to be assured that a Unicode glyph you are ...
    (microsoft.public.mac.office.word)