Received: by nummer-3.proteosys id <01C19443.A54056AC@nummer-3.proteosys>; Thu, 3 Jan 2002 11:44:39 +0100 Return-Path: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C19443.A54056AC" x-vm-v5-data: ([nil nil nil nil nil nil nil nil nil][nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil]) X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message Subject: Extending the lcircle fonts Date: Tue, 31 Mar 1992 22:56:16 +0100 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: From: Sender: "LaTeX-L Mailing list" To: "Multiple recipients of" Reply-To: "LaTeX-L Mailing list" Status: R X-Status: X-Keywords: X-UID: 656 This is a multi-part message in MIME format. ------_=_NextPart_001_01C19443.A54056AC Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I don't want to dampen anyone's enthusiasm for extending TeX, LaTeX, = etc., so that they can give nice illustrations, and I know that not everyone = has PostScript printers, but ... I think one needs to be cautious about advertising (La)TeX as a drawing package. * For example, there are TeX macro packages around that will draw = nice chemical structure diagrams for particular types of structures. = But for general chemical structures, a chemist is better off with a special-purpose drawing package (such as chemdraw or chemintosh) = and importing the result into LaTeX as encapsulated PostScript via a \special. A chemist who actually needs the facilities of a special-purpose drawing package, but happens to stumble across the "chemical structure" TeX macro package first may be (a) = disappointed (b) a nuisance (since they will come along asking for extra = facilities that are difficult to put into TeX macros but easy to get in a structure drawing package). * Similarly, there are TeX macro packages that will draw nice graphs, for certain types of graph. But once one reaches the limits of the TeX macro packages one has to transfer to a graphics package and = paste encapsulated PostScript in via \special again. A user who has got = to the limit of the macro package and "just wants one extra facility" = may (a) be disappointed (b) exert pressure for "just one more feature" = to creep in to the "graphs via TeX" macro package. [Another approach might be to find out what basic operations graphics packages = require, e.g., "move to point", "draw line to point", and see if a TeX interface could be provided that would allow a real graphics = package (such as Simpleplot or NAG Graphics) to write instructions in terms = of those basic operations which TeX could then "plot".] It may be wise to bear in mind the traditional division between what can = be typeset and what gets drawn as artwork and: * try to improve LaTeX to do well what is traditionally typeset * be cautious about attempting to improve LaTeX to do things that are traditionally done as artwork. Treat "pasting in via encapsulated PostScript (perhaps after tidying up via Adobe Illustrator etc.)" = as the norm for artwork. If there is the prospect that everything = that anyone could ever want to do in a particular area could be done via = a TeX macro package (given suitable fonts), it might be worth doing. But advertising (La)TeX as "able to draw ... diagrams" may give = the impression that (La)TeX can "draw ALL ... diagrams" and just cause disappointment and demands for extra facilities when the end-user finds that (La)TeX can only "draw SOME ... diagrams". If a TeX macro package does get written that does what would = traditionally have been done as artwork, I think that the documentation should be = careful to point out any limitations, and to recommend alternative software for = use when the user reaches the limitations. I'm not familiar with what people want out of Feynman diagrams. Would = the proposed addition to the lcircle fonts: 1. stop end-users asking for more facilities ever again (because the proposed extensions would do everything that any Feyman = diagram-drawer could ever want) 2. just encourage end-users to ask for more. "Yes the pion loops and = photon loops are very nice. I just want TeX to draw some ... too. (If I = draw ... in by hand they don't match the pion loops and the photon = loops.)" If its (2), it might not be worth the trouble. David = Rhead ------_=_NextPart_001_01C19443.A54056AC Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Extending the lcircle fonts

I don't want to dampen anyone's enthusiasm for = extending TeX, LaTeX, etc.,
so that they can give nice illustrations, and I know = that not everyone has
PostScript printers, but ...

I think one needs to be cautious about advertising = (La)TeX as a drawing
package.
*    For example, there are TeX macro = packages around that will draw nice
     chemical structure diagrams = for particular types of structures.  But
     for general chemical = structures, a chemist is better off with a
     special-purpose drawing = package (such as chemdraw or chemintosh) and
     importing the result into = LaTeX as encapsulated PostScript via a
     \special.  A chemist = who actually needs the facilities of a
     special-purpose drawing = package, but happens to stumble across the
     "chemical = structure" TeX macro package first may be (a) disappointed
     (b) a nuisance (since they = will come along asking for extra facilities
     that are difficult to put = into TeX macros but easy to get in a
     structure drawing = package).
*    Similarly, there are TeX macro = packages that will draw nice graphs,
     for certain types of = graph.  But once one reaches the limits of the
     TeX macro packages one has = to transfer to a graphics package and paste
     encapsulated PostScript in = via \special again.  A user who has got to
     the limit of the macro = package and "just wants one extra facility" may
     (a) be disappointed (b) = exert pressure for "just one more feature" to
     creep in to the "graphs = via TeX" macro package.  [Another approach
     might be to find out what = basic operations graphics packages require,
     e.g., "move to = point", "draw line to point", and see if a TeX
     interface could be provided = that would allow a real graphics package
     (such as Simpleplot or NAG = Graphics) to write instructions in terms of
     those basic operations which = TeX could then "plot".]

It may be wise to bear in mind the traditional = division between what can be
typeset and what gets drawn as artwork and:
*    try to improve LaTeX to do well = what is traditionally typeset
*    be cautious about attempting to = improve LaTeX to do things that are
     traditionally done as = artwork.  Treat "pasting in via encapsulated
     PostScript (perhaps after = tidying up via Adobe Illustrator etc.)" as
     the norm for artwork.  = If there is the prospect that everything that
     anyone could ever want to do = in a particular area could be done via a
     TeX macro package (given = suitable fonts), it might be worth doing.
     But advertising (La)TeX as = "able to draw ...  diagrams" may give the
     impression that (La)TeX can = "draw ALL ...  diagrams" and just cause
     disappointment and demands = for extra facilities when the end-user
     finds that (La)TeX can only = "draw SOME ...  diagrams".
If a TeX macro package does get written that does = what would traditionally
have been done as artwork, I think that the = documentation should be careful
to point out any limitations, and to recommend = alternative software for use
when the user reaches the limitations.

I'm not familiar with what people want out of Feynman = diagrams.  Would the
proposed addition to the lcircle fonts:
1. stop end-users asking for more facilities ever = again (because the
   proposed extensions would do everything = that any Feyman diagram-drawer
   could ever want)
2. just encourage end-users to ask for more. = "Yes the pion loops and photon
   loops are very nice. I just want TeX to = draw some ... too.  (If I draw
   ... in by hand they don't match the pion = loops and the photon loops.)"
If its (2), it might not be worth the trouble.


          &nbs= p;            = ;            =             &= nbsp;           &n= bsp;   David Rhead

------_=_NextPart_001_01C19443.A54056AC--