References: <504FAAC6.8000605@morningstar2.co.uk>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="876660292-382686053-1347403564=:97942"
Message-ID:  <1347403564.97942.YahooMailNeo@web181204.mail.ne1.yahoo.com>
Date:         Tue, 11 Sep 2012 15:46:04 -0700
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@listserv.uni-heidelberg.de>
Sender: Mailing list for the LaTeX3 project <LATEX-L@listserv.uni-heidelberg.de>
From: Paul Thompson <patjah@SBCGLOBAL.NET>
Subject: Re: Managing/tracking module prefixes
To: LATEX-L@listserv.uni-heidelberg.de
In-Reply-To:  <504FAAC6.8000605@morningstar2.co.uk>
Precedence: list
Status: R

--876660292-382686053-1347403564=:97942
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Great idea. Why not a list of prefixes of a standard length - say 4 chars -=
 that all variables would need to start with (or would be encouraged to). Y=
ou would select one from the list, indicate that you had chosen it, and use=
 it for all variable within your style/class (or am I confused). That way, =
you would begin with a list which was automatically conflict free.=0A=0A=A0=
=0APaul Thompson =0A=0AProfessor and Senior Scientist=0ADirector, Methodolo=
gy and Data Analysis Center=0ASanford Research/USD=0A2301 E 60th St N=0ASio=
ux Falls, SD    57104=0A=0A=0AO: 605-312-6462=A0=A0 M: 618-974-0473=A0 H: 6=
05-332-1587=A0 F: 605-328-0401=0A=0A=0A909 N. Charleston Circle=0ASioux Fal=
ls, SD 57110=0A605-332-1587=0A=0A=0A________________________________=0A Fro=
m: Joseph Wright <joseph.wright@MORNINGSTAR2.CO.UK>=0ATo: LATEX-L@LISTSERV.=
UNI-HEIDELBERG.DE =0ASent: Tuesday, September 11, 2012 4:19 PM=0ASubject: M=
anaging/tracking module prefixes=0A =0AHello all,=0A=0ATake up of expl3 as =
a programming language raises new issues both in the=0Acode itself and in t=
he wider structures.=0A=0AOne area which deserves attention is namespacing:=
 management of module=0Aprefixes (and related information). In LaTeX2e, thi=
s is handled in a=0Abroadly successful but somewhat ad hoc manner, relying =
on searching over=0Areleased material to find what prefixes are in use. As =
expl3 formalises=0Athe idea of using a namespace prefix for each module, it=
 seems=0Aappropriate to consider a more formal approach to managing these.=
=0A=0ALooking outside of the TeX world, it is notable that similar concerns=
=0Acome up for example in the Perl community:=0Ahttp://www.cpan.org/modules=
/04pause.html#namespace. Clearly, our=0Arequirements are different as expl3=
 code is a subset of (La)TeX code,=0Aand so management at the CTAN level is=
 inappropriate.=0A=0AI think we can see namespace management in two parts:=
=0A- 'Outside LaTeX': having a system available for consultation=0A=A0  bef=
ore any code is written.=0A- 'Inside LaTeX': making module information avai=
lable from within a=0A=A0  TeX run.=0A=0AAt this stage, I want to focus on =
the 'outside' part of the question.=0AThere are many approaches that one ca=
n imagine, varying from the very=0Asimple to the very complex. An approach =
I'd like to raise here is at the=0Alow-tech end of the spectrum. I envisage=
 a simple list of module=0Aprefixes with associated information: the module=
 name, developer(s),=0Acontact details, web site, bug tracker, etc. This wo=
uld be available in=0Apublic (CTAN/LaTeX3 SVN), but updates would rely on a=
 simple process:=0Acontacting the team and making a request. This could the=
n be made more=0A'high tech' in the future if necessary.=0A=0AAn 'open' lis=
t of prefixes offers advantages, for example begin able to=0A'reserve' pref=
ixes in advance of code release (for the team) and making=0Aclear that some=
 prefixes are 'free for all' (perhaps "tmp" and "foo",=0Afor example). At t=
he same time, there will be some issues, for example=0Ahow to handle confli=
cts and how to make sure information remains up to date.=0A=0AClearly, any =
such approach requires agreement within the expl3 developer=0Acommunity. Th=
us what I am seeking here is in the first instance feedback=0Aon this idea.=
 Does this seem sensible, workable and useful?=0A--=0AJoseph Wright
--876660292-382686053-1347403564=:97942
Content-Type: text/html; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

<html><body><div style=3D"color:#000; background-color:#fff; font-family:bo=
okman old style, new york, times, serif;font-size:12pt"><div><span>Great id=
ea. Why not a list of prefixes of a standard length - say 4 chars - that al=
l variables would need to start with (or would be encouraged to). You would=
 select one from the list, indicate that you had chosen it, and use it for =
all variable within your style/class (or am I confused). That way, you woul=
d begin with a list which was automatically conflict free.<br></span></div>=
<div>&nbsp;</div><div>Paul Thompson <br></div><div>Professor and Senior Sci=
entist<br>Director, Methodology and Data Analysis Center<br>Sanford Researc=
h/USD<br>2301 E 60th St N<br>Sioux Falls, SD    57104</div><br><br><div>O: =
605-312-6462&nbsp;&nbsp; M: 618-974-0473&nbsp; H: 605-332-1587&nbsp; F: 605=
-328-0401</div><div><br><br></div><div>909 N. Charleston Circle<br>Sioux Fa=
lls, SD 57110<br>605-332-1587<br>  <div style=3D"font-family: bookman
 old style, new york, times, serif; font-size: 12pt;"> <div style=3D"font-f=
amily: times new roman, new york, times, serif; font-size: 12pt;"> <div dir=
=3D"ltr"> <font face=3D"Arial" size=3D"2"> <hr size=3D"1">  <b><span style=
=3D"font-weight:bold;">From:</span></b> Joseph Wright &lt;joseph.wright@MOR=
NINGSTAR2.CO.UK&gt;<br> <b><span style=3D"font-weight: bold;">To:</span></b=
> LATEX-L@LISTSERV.UNI-HEIDELBERG.DE <br> <b><span style=3D"font-weight: bo=
ld;">Sent:</span></b> Tuesday, September 11, 2012 4:19 PM<br> <b><span styl=
e=3D"font-weight: bold;">Subject:</span></b> Managing/tracking module prefi=
xes<br> </font> </div> <br>Hello all,<br><br>Take up of expl3 as a programm=
ing language raises new issues both in the<br>code itself and in the wider =
structures.<br><br>One area which deserves attention is namespacing: manage=
ment of module<br>prefixes (and related information). In LaTeX2e, this is h=
andled in a<br>broadly successful but somewhat ad hoc manner, relying on se=
arching
 over<br>released material to find what prefixes are in use. As expl3 forma=
lises<br>the idea of using a namespace prefix for each module, it seems<br>=
appropriate to consider a more formal approach to managing these.<br><br>Lo=
oking outside of the TeX world, it is notable that similar concerns<br>come=
 up for example in the Perl community:<br><a href=3D"http://www.cpan.org/mo=
dules/04pause.html#namespace" target=3D"_blank">http://www.cpan.org/modules=
/04pause.html#namespace</a>. Clearly, our<br>requirements are different as =
expl3 code is a subset of (La)TeX code,<br>and so management at the CTAN le=
vel is inappropriate.<br><br>I think we can see namespace management in two=
 parts:<br> - 'Outside LaTeX': having a system available for consultation<b=
r>&nbsp;  before any code is written.<br> - 'Inside LaTeX': making module i=
nformation available from within a<br>&nbsp;  TeX run.<br><br>At this stage=
, I want to focus on the 'outside' part of the question.<br>There are
 many approaches that one can imagine, varying from the very<br>simple to t=
he very complex. An approach I'd like to raise here is at the<br>low-tech e=
nd of the spectrum. I envisage a simple list of module<br>prefixes with ass=
ociated information: the module name, developer(s),<br>contact details, web=
 site, bug tracker, etc. This would be available in<br>public (CTAN/LaTeX3 =
SVN), but updates would rely on a simple process:<br>contacting the team an=
d making a request. This could then be made more<br>'high tech' in the futu=
re if necessary.<br><br>An 'open' list of prefixes offers advantages, for e=
xample begin able to<br>'reserve' prefixes in advance of code release (for =
the team) and making<br>clear that some prefixes are 'free for all' (perhap=
s "tmp" and "foo",<br>for example). At the same time, there will be some is=
sues, for example<br>how to handle conflicts and how to make sure informati=
on remains up to date.<br><br>Clearly, any such approach requires
 agreement within the expl3 developer<br>community. Thus what I am seeking =
here is in the first instance feedback<br>on this idea. Does this seem sens=
ible, workable and useful?<br>--<br>Joseph Wright<br><br><br> </div> </div>=
  </div></div></body></html>
--876660292-382686053-1347403564=:97942--