@percusse This guy is great. Exactly my opinion.

@Johannes_B unfortunately we don't have a good one, eg the new one just now ought to be a duplicate of tex.stackexchange.com/questions/98988/strange-figure-floats/… but Martin closed that as dup (but the duplicate doesn't have the right answer)

@DavidCarlisle We also have something along the line of figure near where its called

@Johannes_B yes somewhere although that's normally placeins: stopping it going too far forwards, not flafter: stopping it going backwards.

@DavidCarlisle I think we should have one (or a blog post) shortly introducing the floating behaviour and then link to Q/As on main.

if anyone can spare a vote, one here will not affect my rep and would take a Q off the unanswered list: tex.stackexchange.com/a/228392/21344

@PaulGessler It's quite a radical change though to suggest changing engine just to scale cmss (since almost everyone uses scalable type1 versions of cm these days)

@DavidCarlisle obviously the OP does not, otherwise the problem would not have showed up...

@PaulGessler Not at all, you just need to tell latex that cm is scalable, fix-cm package (or the older type1cm) for example

@DavidCarlisle ok, edited. I've never used fix-cm, I guess because I only deal with "normal" English-language stuff most of the time.

@PaulGessler thanks, voted:-)

@DavidCarlisle Now reading the docs, I see the changes are "substantial", even for English-language work. I understand that these fixes (fixltx2e and fix-cm) are not baked into the kernel because of the ever-present pixel-level backward compatibility requirement, but why do I almost never see these used in normal documents?

@PaulGessler that is such an excellent question:-)

@JosephWright ^^^^^^

@DavidCarlisle :-)

@DavidCarlisle oh boy, what did I start here... lol :-)

@PaulGessler texdev.net/2014/12/28/fixing-latex2e

Hello all

@DavidCarlisle ah yes, I remember reading this but totally forgot that the "backout mechanism" was in the works.

Thanks for the link.

@PaulGessler I spent most of the day running the latex test suite testing the new release (I'm on "holiday" this week:-)

@DavidCarlisle I've noticed from your messages. :-) New release as in the release will go into TL 2015? Or does it not propagate out that fast?

lol

Does it have an equally excellent answer?

To be decided. If it's stable by the time of the TL2015 pretest then hopefully it'll go in, if there are still things we decide need more thought better to hold it off so it doesn't go in the DVD, and push it out via tlmgr update later in year. We'll see....

@DavidCarlisle You don't know why something happens. I do. ;-)

@DavidCarlisle ok, that makes sense. So it sounds like a goal, but not necessarily a mission-critical goal.

@egreg Oh I could have run \tracingall and find out what was happening, but what I meant is I don't know what the packages do (just removing them would also make the error go away:-)

@egreg My answer is shorter and got a greener tick

@GonzaloMedina 305:-) You are not making it easy to catch up!

@JosephWright oh you simply removed the toc? Hmm In that case I could put some text back that I pruned to make it fit in 2 pages....

@DavidCarlisle Not exactly: only the last two had TOCs at all, and the script we've taken from Heiko doesn't work with multiple TOCs

I don't want to go back to Emacs is usually said by vim users, but never so far by texmaker users. Ref

@ChristianHupfer As it seems, @PaulGessler joined the secret society tex.stackexchange.com/a/228392/37907

@JosephWright yes that's what I mean, but it means we can put some of the text back perhaps

@JosephWright oh so the toc in ltnews21 in the current texdoc ltnews just worked by accident as it was the only one?

@DavidCarlisle Yes, basically!

@Johannes_B I'm still lost when you and @ChristianHupfer talk about this banana guy. My answer was a reference to the reddit-phenomenon "banana for scale". :-)

@PaulGessler google.de/…

@PaulGessler That user made quite a few gray hairs on my head. I nearly lost hope. It was a hard time back then.

@Johannes_B :-)

Well, all right

It won't log me into my normal account, so I guess I use this

Hi. Could anyone run the following and tell me whether it compiles?

\documentclass{article}

\usepackage{chemfig}



\begin{document}



stuff

\end{document}

Good..

user55789 is me.. somehow it wouldn't let me log onto this.

Hi..

@1010011010 worked for me (and I remembered that chemfig was updated today, so I updated and tested with the new version as well: no issues).

I get this on new documents:

aTeX Error: Two \documentclass or \documentstyle commands.

See the LaTeX manual or LaTeX Companion for explanation.
Type  H <return>  for immediate help.
 ...

l.1 \documentclass{
                   article}

Old documents work fine. I can copy and paste the full code of an old document and it won't work. I can also copy and paste all files (aux, etc. included), and it fails to compile with this error.

Oh, and I also just tried with the updated version (thanks for the suggestion) and it fails to compile with 2x documentclass for an error.

@1010011010 does it still give that error if you take out chemfig?

@PaulGessler Nope, only with chemfig.

 *File List*
 article.cls    2014/09/29 v1.4h Standard LaTeX document class
  size10.clo    2014/09/29 v1.4h Standard LaTeX file (size option)
 chemfig.sty    2015/02/15 v1.1 Draw molecule with an easy syntax
    tikz.sty    2013/12/13 v3.0.0 (rcs-revision 1.142)
     pgf.sty    2013/12/18 v3.0.0 (rcs-revision 1.14)
  pgfrcs.sty    2013/12/20 v3.0.0 (rcs-revision 1.28)
everyshi.sty    2001/05/15 v3.00 EveryShipout Package (MS)
  pgfrcs.code.tex
 pgfcore.sty    2010/04/11 v3.0.0 (rcs-revision 1.7)

@1010011010 ^^^ that's my \listfiles output with your MWE. Do you have differences?

(Add \listfiles to your preamble and recompile, the output shows up near the end of the log file.)

 *File List*
 article.cls    2007/10/19 v1.4h Standard LaTeX document class
  size10.clo    2007/10/19 v1.4h Standard LaTeX file (size option)
 chemfig.sty
    tikz.sty    2013/12/13 v3.0.0 (rcs-revision 1.142)
     pgf.sty    2013/12/18 v3.0.0 (rcs-revision 1.14)
  pgfrcs.sty    2013/12/20 v3.0.0 (rcs-revision 1.28)
everyshi.sty    2001/05/15 v3.00 EveryShipout Package (MS)
  pgfrcs.code.tex
 pgfcore.sty    2010/04/11 v3.0.0 (rcs-revision 1.7)
graphicx.sty    2014/04/25 v1.0g Enhanced LaTeX Graphics (DPC,SPQR)

@PaulGessler Chemfig package shows no date.

@1010011010 Is your file called chemfig.tex? ;-)

@Johannes_B Damn.

:-)

@Johannes_B aha! :-)

@1010011010 Happened to me once ;-)

@Johannes_B I'm sure you know a cool story behind this one?

@Johannes_B that shouldn't erase the listfiles info for the package?

@DavidCarlisle And tikz wouldn't be loaded. Wondering myself.

@DavidCarlisle But it does, why?

@Johannes_B correction it would. for chenfig

A seemingly related question that never got an answer:

@Johannes_B because chemfig.sty (I just looked) is just a shell that does \input chemfig.tex (it's a generic plain tex macro set not really latex)

So the first time, the user tex is found, and next time the real chemfig.tex?

@Johannes_B no the users document is input again which is why in the q @PaulGessler just posted the OP got a two docuemsntclass error

so is there a new question about this? or was it just in chat, I just came back and caught the end of the discussion

@DavidCarlisle Yes, this does the two doc error, but it doesn't run in an infinite loop, so when is TL-chemfig.tex found?

@DavidCarlisle we should re-tag it with recursion :-)

@DavidCarlisle just a chat

@DavidCarlisle Was just in chat.

@PaulGessler I answered that one:-)

@Johannes_B never

@DavidCarlisle Ahhhh, now i got it. One of my slow days 3 minutes before the library is closing.

@Johannes_B it doesn't loop as latex does not input packages twice so if you step past the two \documentclass error the "second" \usepackage{chemfig} is ignored as the package is already loaded

Guys, a good night to all of you.

@DavidCarlisle voted, case closed. :)

@Johannes_B good night.

@Johannes_B good night

@DavidCarlisle Got +1 from me too :-)

Might help him out with the !!/battle

@PaulGessler only if the OP comes and gives me a tick (rep capped)

@DavidCarlisle I figured so. OP return is unlikely; he/she's not been around since 8 hr after posting the question

Psmith must be sleeping again, I can never summon him. :'(

:-)

@PaulGessler current battle is with Gonzalo rather than egreg and today's a bit of a setback I'm stuck at 275 with only an hour to go, and Gonzalo's 30 ahead:-)

Thanks all for figuring it out. I would've probably checked stared at the screen for an hour or two without it... :-) I'll restart my PC because I suspect a bug somewhere! Night for now!