Received: from mail.proteosys.com ([62.225.9.49]) by nummer-3.proteosys with Microsoft SMTPSVC(5.0.2195.5329); Sun, 4 May 2003 22:08:36 +0200 Received: by mail.proteosys.com (8.12.9/8.12.2) with ESMTP id h44K8WPQ024626 for ; Sun, 4 May 2003 22:08:34 +0200 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C31278.F1E1FA00" Received: from sun.dante.de (root@sun.dante.de [134.100.9.52]) by rzdspc1.informatik.uni-hamburg.de (8.12.9/8.12.9) with ESMTP id h44K8Klq025683 (version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NO); Sun, 4 May 2003 22:08:20 +0200 (CEST) Received: from rzdspc1.informatik.uni-hamburg.de (root@rzdspc1.informatik.uni-hamburg.de [134.100.9.61]) by sun.dante.de (8.12.9/8.12.9) with ESMTP id h44K84e1014642; Sun, 4 May 2003 22:08:04 +0200 (CEST) X-MimeOLE: Produced By Microsoft Exchange V6.5 Received: from postman.arcor.de (postman1.arcor-online.net [151.189.0.187]) by rzdspc1.informatik.uni-hamburg.de (8.12.9/8.12.9) with ESMTP id h44K7glq025621 (version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NO); Sun, 4 May 2003 22:07:55 +0200 (CEST) Received: from localhost (dialin-145-254-239-199.arcor-ip.net [145.254.239.199]) (authenticated bits=0) by postman.arcor.de (8.12.9/8.12.9) with SMTP id h44K7V74017602; Sun, 4 May 2003 22:07:35 +0200 (MEST) In-Reply-To: <16053.14411.19419.666172@istrati.mittelbach-online.de> Return-Path: X-Mailer: PMMail 2.20.2380 for OS/2 Warp 4.0 X-OriginalArrivalTime: 04 May 2003 20:08:36.0520 (UTC) FILETIME=[F2315280:01C31278] X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang) X-Spam-Score: -9.9 () EMAIL_ATTRIBUTION,IN_REP_TO Content-class: urn:content-classes:message Subject: Re: textcomp once more Date: Sun, 4 May 2003 21:06:19 +0100 Message-ID: <200305042007.h44K7V74017602@postman.arcor.de> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: textcomp once more Thread-Index: AcMSePJa/+2YCwdtQ+y8pZ/1VP458Q== From: "Walter Schmidt" To: "Frank Mittelbach" Cc: , =?iso-8859-1?Q?Lars_Hellstr=F7m?= , Reply-To: "Walter Schmidt" Status: R X-Status: X-Keywords: X-UID: 4611 This is a multi-part message in MIME format. ------_=_NextPart_001_01C31278.F1E1FA00 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable On Sun, 4 May 2003 17:56:59 +0200, Frank Mittelbach wrote: >there is an encguide.tex that was never made publically available, = perhaps it >is one day Are you joking? First you try to dissipate my objections=20 by promising sufficient documentation; then you admit that=20 documentation will be made available only "perhaps". =20 If you don't have the time to document it properly, then=20 don't implement it. Period. =20 >[...] to convince me you should explain to me why >your approach serves the majority and mine the minority It is easy to explain that a certain font family has no Euro symbols, and that \texteuro fails for this reason. Everyone can understand this, and noone will blame TeX for the problem. However, it is very difficult to explain, that a font family _does_ have Euros, but you can't use them, unless you have edited a certain .cfg file, which is _perhaps_ documented. Thus, the latter approach will be confusing for larger number=20 of TeX users. Btw, didn't we already agree on a better approach than the=20 use of a .cfg file? On Mon, 28 Apr 2003 15:12:34 +0200, Lars Hellstr=F6m wrote: >One could probably have some of these declarations in textcomp.sty = (e.g. >those for CM families), use the "full TS1" as default, and suggest that >font-loading packages contain explicit declarations when they don't = provide >this default. One further problem comes to my mind: In a typical TeX system=20 on a mluti-user OS three cumulative .cfg files would be needed: * One that comes with the LaTeX distribustion, so that it is overwritten whenever LaTeX gets updated. * One, where I can add fonts that are locally installed=20 for system-wide use. Thhis file must not be overwritten,=20 when LaTeX is updated; otherwise the local addions would=20 be lost. * A third one, which contains user-specific additions. I can hardly imagine that you actually want to implement that. Thus, the only feasible approach is the one suggested by Lars: Enable all (or almost all) symbols by default, and rely on=20 font-loading packages to provide optionally the additional=20 goodie of improved error messages and warnings.=20 Provided that we include this into package such as mathptmx, mathpazo or lucidabr, a majority of users would profit from it,=20 while any uneeded restrictions are avoided. just my 2c Walter ------_=_NextPart_001_01C31278.F1E1FA00 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Re: textcomp once more

On Sun, 4 May 2003 17:56:59 +0200, Frank Mittelbach = wrote:

>there is an encguide.tex that was never made = publically available, perhaps it
>is one day

Are you joking?  First you try to dissipate my = objections
by promising sufficient documentation; then you admit = that
documentation will be made available only = "perhaps". 
If you don't have the time to document it properly, = then
don't implement it.  Period.

 
>[...] to convince me you should explain to me = why
>your approach serves the majority and mine the = minority

It is easy to explain that a certain font family has = no Euro
symbols, and that \texteuro fails for this = reason.  Everyone
can understand this, and noone will blame TeX for the = problem.
However, it is very difficult to explain, that a font = family
_does_ have Euros, but you can't use them, unless you = have
edited a certain .cfg file, which is _perhaps_ = documented.
Thus, the latter approach will be confusing for = larger number
of TeX users.


Btw, didn't we already agree on a better approach than = the
use of a .cfg file?

On Mon, 28 Apr 2003 15:12:34 +0200, Lars Hellstr=F6m = wrote:

>One could probably have some of these declarations = in textcomp.sty (e.g.
>those for CM families), use the "full = TS1" as default, and suggest that
>font-loading packages contain explicit = declarations when they don't provide
>this default.


One further problem comes to my mind:  In a = typical TeX system
on a mluti-user OS three cumulative .cfg files would = be needed:

* One that comes with the LaTeX distribustion, so that = it
is overwritten whenever LaTeX gets updated.

* One, where I can add fonts that are locally = installed
for system-wide use.  Thhis file must not be = overwritten,
when LaTeX is updated; otherwise the local addions = would
be lost.

* A third one, which contains user-specific = additions.

I can hardly imagine that you actually want to = implement that.

Thus, the only feasible approach is the one suggested = by Lars:
Enable all (or almost all) symbols by default, and = rely on
font-loading packages to provide optionally the = additional
goodie of improved error messages and warnings. =

Provided that we include this into package such as = mathptmx,
mathpazo or lucidabr, a majority of users would = profit from it,
while any uneeded restrictions are avoided.

just my 2c
Walter

------_=_NextPart_001_01C31278.F1E1FA00--