LaTeX Warning: Label(s) may have changed. Rerun to get cross-references right.

[Silvio Picano]

I remember someone asked this a few months ago, and now I seem to have
this problem.

[1] $ latex paper
[2] $ bibtex paper
[3] $ latex paper
[4] $ latex paper

After [4], I still get the Warning. If I run "$ latex paper"
one more time, I do not get the Warning. I do not see anything wrong
with the document, but I would like that "warm fuzzy" feeling w/o
running the extra "$ latex paper".

Can anyone explain why or what it could be?

Thanks.
Silvio

[Chris Thompson]

In article <1993Jul1.0...@en.ecn.purdue.edu>, pic...@en.ecn.purdue.edu

There is no guarantee that label convergence will be achieved after two
runs of LaTeX, as the changes in the labels can cause the pagination to
keep on changing. To find out what is changing, you should compare the
contents of paper.aux after each run (take copies).

If you are thinking of the same "someone [who] asked this a few months ago"
as I am: in that case the claim was that label convergence *never* occurred.
This is certainly theoretically possible, and I doubt whether "the chance
of [this] happening by accident" is quite as "infinitesimal" as Lamport
claims in the footnote on page 70 of The Book.

Chris Thompson
Internet: ce...@phx.cam.ac.uk
JANET: ce...@uk.ac.cam.phx

[Gunars V Lucans]

ce...@cus.cam.ac.uk (Chris Thompson) writes:
>There is no guarantee that label convergence will be achieved after two
>runs of LaTeX, as the changes in the labels can cause the pagination to
>keep on changing. To find out what is changing, you should compare the
>contents of paper.aux after each run (take copies).

>If you are thinking of the same "someone [who] asked this a few months ago"
>as I am: in that case the claim was that label convergence *never* occurred.
>This is certainly theoretically possible, and I doubt whether "the chance
>of [this] happening by accident" is quite as "infinitesimal" as Lamport
>claims in the footnote on page 70 of The Book.

I have certainly run into this myself, although for different reasons.
In a large document spanning a dozen included files, I inadvertently
defined the same \label in two places. The result was also a series
of "labels may have changed" messages that of course never converged
(and comparing .aux files doesn't help in this situation). With
various people writing different chapters of the document, the chance
of label conflicts is not "infinitesimal".

--
Gunars V. Lucans -- gun...@spss.com -- SPSS Inc, Chicago, USA

[Donald Arseneau]

In article <gunars.741543998@spssig>, gun...@spss.com (Gunars V Lucans) writes:
>In a large document spanning a dozen included files, I inadvertently
>defined the same \label in two places. The result was also a series
>of "labels may have changed" messages that of course never converged
>(and comparing .aux files doesn't help in this situation). With
>various people writing different chapters of the document, the chance
>of label conflicts is not "infinitesimal".

I agree. And the message

"Label(s) may have changed . Rerun to get cross-references right"

is singularly unhelpful. However, at the beginning of the run you
should get the much more useful message:

"LaTeX Warning: Label `a' multiply defined on input line 8."

Donald Arseneau as...@reg.triumf.ca

[Rainer Schoepf]

Unfortuantely this tends to disappear off the top of the screen. I'll
think about putting in another warning at the end of the LaTeX run for
the next update (no questions please -- I don't know when)

Rainer Sch"opf
LaTeX 2.09 maintenance
--

Rainer Schoepf
Konrad-Zuse-Zentrum
fuer Informationstechnik Berlin
Heilbronner Strasse 10
D-10711 Berlin
Federal Republic of Germany
<Sch...@sc.ZIB-Berlin.de> or <Sch...@sc.ZIB-Berlin.d400.de>