TeX, LaTeX and Friends

Sean Allred

03:21

I'm not Apple's biggest fan right now. We're likely to get many questions per this change – read up on this El Capitan upgrade issue. Might be a worth a bookmark for the next few weeks.

Far and above the easiest fix is to tell users to upgrade to MacTeX 2015, but this doesn't mean editors are automatically updated to use the new /texbin search path.

Adam Liter

03:58

@SeanAllred Did you see this?

36

Q: Mac OS changes that affect lots of answers

Those us who are Mac users may be aware that the next version of OS X will change major aspects of the the way users can interact with the underlying Unix infrastructure of the OS. In particular, access to /usr/ will essentially be eliminated, even with admin privileges. As a result, starting fro...

discussion

There were a few suggestions of what to do in the comments, but nothing definitive ever seems to have been decided.

Sean Allred

04:20

@AdamLiter Ooooh, I did not :) Thanks for pointing it out – I'll give that page a read-through and try to offer some thoughts.

Joseph Wright

04:39

@DavidCarlisle It's the priority business: if I add that back in then everything works

Joseph Wright

05:15

@DavidCarlisle I've fixed one issue: latexrelease wasn't getting things back on track, now it is

@DavidCarlisle We might have to simply put up with the priority business: that can get fixed today if we agree and sent (CTAN will love me)

David Carlisle

06:39

@JosephWright sent you mail but then saw you'd done it already, yes agree we should do whatever to get it working today although I will try to look later I think it should be possible to emulate priority without it being in ltluatex, can always do that next time.

Joseph Wright

@DavidCarlisle OK

David Carlisle

@JosephWright yes was worrying about that last night but decided after realising I was dosing at the keyboard that I wouldn't trust any changes I made, so gave up last night:-(

David Carlisle

07:39

@JosephWright so just got in, I see ctablestack just got announced...

Joseph Wright

@DavidCarlisle :-)

@DavidCarlisle Working on the ctex tests for expl3 [which also pick up the LaTeX2e issue :-)]

David Carlisle

@JosephWright I think I can see how to do priority better in the luatexbase emulation, or have you already added it to ltuatex? Also the only values ctex seems to use are 1 and 1000 it might make sense to have a version of add_to_callback which reliably adds to the other end rather than just hoping 1000 is enough, so we have something to tell people whet to do instead of using priority

Joseph Wright

@DavidCarlisle Hope my comments on the pull request are OK

@DavidCarlisle I've only fixed latexrelease

@DavidCarlisle Did my tests on what would fix the issue in my local tree, not in the sources

David Carlisle

@JosephWright was it just priority being ignored or other things as well

Joseph Wright

@DavidCarlisle Just added the priority code from original luatexbase to add_to_callback in ltluatexand adjusted new luatexbase to pass the value through

@DavidCarlisle, @PauloCereda On the issue I asked about last night: I've done a complete install of TL'15 on Ubuntu 12.10LTS and it still says the Chinese font has no Chinese chars! Must be an issue with the OS or something.

Right, next step is to set up a more recent Ubuntu I guess

David Carlisle

07:46

@JosephWright so, ... do you want to do that or do it in luatexbase (it already catches the priority argument but I forgot I only used it there in the "first" case, but if priority > 1 it could iterate through the description list to unregister all callbacks then put back the ones with position less than the stated priority then the new one then the rest of the old ones

Joseph Wright

@DavidCarlisle That sounds like a better solution if it works: I really don't like the priority idea in general, so if we can keep it out of the kernel ...

David Carlisle

@JosephWright OK I'll have a go, see if it works....

Joseph Wright

@DavidCarlisle I'll still need to fix the kernel w.r.t. latexrelease but that's then a lower priority

@DavidCarlisle You're happy with that plan I guess?

David Carlisle

@JosephWright if it works:-)

David Carlisle

09:18

@JosephWright it works!

Joseph Wright

@DavidCarlisle Yay!

@DavidCarlisle Still can't work out why TL on Ubuntu can't find chars in Fandol

David Carlisle

@JosephWright I need to remove debugging print statements and move it to dtx file, but I just changed this function in luatexbase.sty

function luatexbase.add_to_callback(name,fun,description,priority)
  texio.write_nl('\string\n HERE: adding ' ..
                  description ..
                  ' to ' ..
                  name ..
                  ' with priority ' ..
                  (priority or '@@@'))
    texio.write_nl('Original list')
    for k,v in pairs(luatexbase.callback_descriptions(name)) do
      texio.write_nl('    ' .. k .. ': ' .. v)
    end
  local priority= priority
  if priority==nil then
   priority=\string#luatexbase.callback_descriptions(name)+1

Joseph Wright

@DavidCarlisle Good plan

Paulo Cereda

@JosephWright Oh my, that is something complicated to track it down.

Joseph Wright

@PauloCereda I know

@PauloCereda At present I'm minded to go with a bail-out, but I really need to work out why there is an issue

@PauloCereda I'll try Fedora next

Paulo Cereda

09:35

@JosephWright Got it.

@JosephWright Yay!

Joseph Wright

@PauloCereda If it's just Ubuntu that will be very weird, but you never know

Paulo Cereda

@JosephWright I can't think of any explanataion. At first I thought it would be something related to AppArmor (I guess that would be the name of the software firewall thingy) but that would be just silly.

Joseph Wright

@PauloCereda Font loader issue, but as I'm not really up on the detail of the interaction between luaotfload and OS stuff it's not so easy to track down

@PauloCereda Font itself is fine, of course

Paulo Cereda

@JosephWright hmmm perhaps font lookup?

Joseph Wright

@PauloCereda I suspect so but it's hard to be sure. First test is as I say a different Linux, stripped down as for Ubuntu, and see what happens

David Carlisle

09:41

@JosephWright git push... can you give that a try...

Joseph Wright

@DavidCarlisle Works for me

David Carlisle

@JosephWright It didn't help that first I put everything back in exactly the wrong order:-)

Joseph Wright

@DavidCarlisle :-)

David Carlisle

@JosephWright I think as emulation this should be fine but we should probably think of adding a prepend_to_callback or some such that's like add_to_callback but from the other end, I can't see anyone really ever wanting to add to the middle

Joseph Wright

@DavidCarlisle I've sent to CTAN: if deals with the immediate problem

David Carlisle

09:48

@JosephWright yes, I meant for 2016/01/01 or whatever

Joseph Wright

@DavidCarlisle See what gets asked for

David Carlisle

@JosephWright yes but I was thinking if we want to ease people off the luatexbase interface we may need to proactively suggest what they should change to, and as all the ctex examples have priority 1 or 1000 having a canned alternative for each of those would simplify matters

Paulo Cereda

El Capitan downloaded, time to make a bootable USB stick. :)

Joseph Wright

@DavidCarlisle Yes

@PauloCereda Good plan

Paulo Cereda

@JosephWright Have you downloaded it too?

Joseph Wright

10:00

@PauloCereda No, as I've said before, I can't go beyond Mavericks as our work VPN software then breaks

Paulo Cereda

@JosephWright Ouch.

Joseph Wright

@PauloCereda support.f5.com/kb/en-us/solutions/public/15000/700/…

Paulo Cereda

@JosephWright They are now two versions behind. :)

Joseph Wright

@PauloCereda We've been told: As the VPN software is due for replacement then, hopefully in the coming months we should have further information to provide.

Paulo Cereda

@JosephWright ooh

Joseph Wright

10:12

@PauloCereda I suspect that the supplier are pushing us to upgrade the entire VPN hardware

Paulo Cereda

@JosephWright Possibly. :)

@JosephWright: I use VPNC here. :)

David Carlisle

LaTeX2e <2015/10/01> patch level 1

Joseph Wright

@PauloCereda Not up to me, up to my employer

David Carlisle

So I wonder how many people ever used LaTeX2e <2015/10/01> :-)

Paulo Cereda

@JosephWright USP doesn't want "third party" softwares as well, but what they know. :)

Joseph Wright

10:16

@PauloCereda I mean you need the right VPN software to connect to the VPN hardware, and that's determined by the manufacturer of the hardware

Paulo Cereda

@JosephWright boo :)

David Carlisle

I see Mico is planning on writing a very long book!

Joseph Wright

@DavidCarlisle ?

@PauloCereda Installing Fedora VM now

Christian Hupfer

10:41

Good maen

David Carlisle

11:04

@JosephWright luatex list (he wants \setcounter{chapter}{2^63})

So if I want w3c.github.io/xml-entities/2007doc/Overview.html to automatically rebuild on commit I need to use something called Travis CI @JosephWright know anything about that? :-)

Paulo Cereda

11:33

@DavidCarlisle ooh

Joseph Wright

@DavidCarlisle :-)

Paulo Cereda

@DavidCarlisle Maybe we might need another workflow.

Joseph Wright

@DavidCarlisle Rebuilds on push not commit, but there we go

David Carlisle

@JosephWright push/commit what's the difference:-) That file has been in CVS for the last 20 years, takes a while to get the new terminology.

Paulo Cereda

@DavidCarlisle hm so you are old.

:)

David Carlisle

11:44

@PauloCereda not as old as ...

Joseph Wright

@DavidCarlisle :-)

Paulo Cereda

@DavidCarlisle ooh :P

Johannes_B

12:28

Does anybody understand the question with minimizing the white space?

David Carlisle

12:44

@Johannes_B no

Ho1

Hi. Once I found a great answer on how to fix lost fonts of a PS file converted to PDF on TeX.SE. I can not find it in my web searches. How can I find that answer?

Paulo Cereda

Ho1

Nice. :-)

Johannes_B

@DavidCarlisle Got more unclear, imho.

Paulo Cereda

13:13

0

Q: Mission control font is aliased in OSX El Capitan

I just upgraded my Mac to 10.11 and noticed the aliased font in Mission Control. I tried to include a screenshot of it, hope the issue is clear. :) The font was fine in Yosemite. Is anybody facing the very same issue? I could not find a proper setting about it in the System Preferences.

osx mission-control el-capitan

Already an issue. :)

Paulo Cereda

13:33

brrrrrr

yo'

@PauloCereda LOL. I would rollback.

Paulo Cereda

@yo' :)

yo'

What is the space for relations? Is it \: or something else?

David Carlisle

13:51

@yo' thats \medmuskip relations get thickmuskip mostly (\;) (depending on what's next to them)

yo'

@DavidCarlisle ord is next to them. But I found out that {} works better than any attempts for a fixed space :)

@PauloCereda You're not alone with problems to climb El Capitan:

0

Q: TeXmaker and El Capitan, Spinning beachball of death

I upgraded to El Capitan and since then i haven't been able to compile. I found som helt that MacTex have been moved from /usr/texbin to /Library/TeX/texbin/ and changed the paths. I can now compile some of my Tex documents while others makes my computer go into some infinity-loop and overheating...

texmaker mactex osx

Johannes_B

Can somebody have a look at this? latex-community.org/forum/viewtopic.php?p=91942#p91942 <- @egreg Something for your collection.

yo'

@Johannes_B Is "have seen worse" any relief? :-)

Johannes_B

@yo' The OP says the document compiles, but there is \batchmode in the class file. I bet he just didn't notice that stuff went wrong.

Paulo Cereda

14:07

@yo' Ouch! At least, there's a new spinning beachball. :)

yo'

14:19

@Johannes_B wait, there's a published document class with \batchmode? Tell me I'm sleeping and this is a nightmare!

Torbjørn T.

@yo' Indeed. mirrors.ctan.org/macros/latex/contrib/turabian/turabian.cls

yo'

@TorbjørnT. I'll have to get a drink to get over this.

Torbjørn T.

@yo' Best make it strong.

Johannes_B

@yo' Sorry, this is reality. Do you want to ping the author, or shall I?

yo'

@TorbjørnT. 55% scots on the rocks will do :D

@Johannes_B I'm not sure I wanna do anything with it :D

Johannes_B

14:28

@yo' \batchmode % Enables TeXworks to continue LaTeX compiling on error.

yo'

@Johannes_B I've seen it, thanks to @TorbjørnT.

David Carlisle

@PauloCereda not too late to switch to windows 10

Paulo Cereda

@DavidCarlisle :)

Johannes_B

@yo' @TorbjørnT. Look at the block in line 46 of the class file. What is that redef supposed to do? All diff i can see are missing EOL-%-signs.

yo'

@Johannes_B Sorry it seems my computer refuses to open it :D

(anyway, the way how lips@dolipsum is written is not optimal, the author obviously missed \loop ... \repeat)

Torbjørn T.

14:44

@Johannes_B Sounds about right. I.e., doesn't do much.

Johannes_B

@TorbjørnT. Well, the end of the block does something unexpected. :-)

Johannes_B

15:01

@yo' @TorbjørnT. Please check section 5 of the user guide.

Paulo Cereda

15:16

@David: Mico's book is no more, apparently. :)

Torbjørn T.

@Johannes_B Charming.

Johannes_B

@TorbjørnT. :-)

Johannes_B

15:50

I don't see why it shouldn't have lipsum loaded. The point I tried to make is how ridiculously haphazard it is. @Johannes_B You are developing an unhealthy dislike to templates. They are not all evil. ;) — percusse 29 mins ago

@percusse Not all, but some :-)

David Carlisle

@PauloCereda yes. seems he'll have to restrict the number of chapters quite considerably.

egreg

@yo' Or, at least, removing the conditional before the next step:

\newcommand\lips@dolipsum{%
  \ifnum\value{lips@count}<\lips@max\relax
    \addtocounter{lips@count}{1}%
    \csname lipsum@\romannumeral\c@lips@count\endcsname
    \expandafter\@firstofone
  \else
    \expandafter\@gobble
  \fi
  {\lips@dolipsum}%
}

Paulo Cereda

@DavidCarlisle Blame Hans. :)

David Carlisle

@PauloCereda He writes such short documents.

yo'

@egreg still, why not

\loop\ifnum\value{lips@count}<\lips@max\relax\stepconuter{lips@count}{1}\csname lipsum@\romannumeral\c@lips@count\endcsname\repeat

?

egreg

16:08

@yo' This is how kantlipsum does it

\cs_new_protected:Nn \kgl_print:
  {
   \int_step_function:nnnN
     {\l_kgl_start_int} {1} {\l_kgl_end_int} \kgl_use:n
  }

@yo' Or \@whilenum

yo'

@egreg both methods are too advanced :D

@DavidCarlisle "This package implements a version of the tabular environment in which the widths of certain columns are calculated so that the table is is a specified width." -- You've got a typo in the first sentence of tabularx.pdf. Should I report it by the official means?

David Carlisle

@yo' amazing, it must be one of the other team members, they have write access of course.

:24473150 $ svn commit -m "is is" tabularx.dtx
Sending        tabularx.dtx
Transmitting file data .
Committed revision 892.

yo'

@DavidCarlisle ok, thanks :-) (must have been some practical joke by @Joseph)

David Carlisle

@yo' probably.

Joseph Wright

@yo' Indeed

yo'

17:02

@percusse Where can I see that it was made by mathematicians?

Johannes_B

@clemens youtube.com/watch?v=5Wh_M3ohtxM

egreg

17:22

@yo' It's clear it was not done by mathematicians.

barbara beeton

@yo' -- as i understand it (and i've been a u.s. rep on an iso working group, though not the one responsible for iso 80000 and its predecessors), the perpetrators of this standard were mostly engineers and their ilk, not practicing mathematicians. if one peruses the mathematical literature from before the predecessor standard (iso 31, first edition 1978-03-15 -- i'm quoting from a printed copy of same), one will not see some of the designated specifications. (cont'd)

(cont'd) in particular, the roman "d" and "e" were not (and are not) used at least in u.s. mathematical literature. i have only part 11 of iso 31, which does not include the references (which might shed some light on origins), but find the formal title of this part -- "mathematical signs and symbols for use in the physical sciences and technology" (emphasis mine) significant. (cont'd)

(cont'd) i suspect participation in its creation by the u.s. national bureau of standards; certainly not the american mathematical society, where no one i talked to was even aware of its existence.

percusse

17:47

@egreg Don't be so sure.

@yo' This is under ISO technical committee 12. They are also responsible for many other things but this standard I know from some personal contacts that involved mathematicians because the internal organization is pretty strict.

Usually, the standard office secretariat is pretty OK but with this one they wanted to have a generic standard that lightly covers the technical documents. However it went sideways.

You can see the context why they wanted to have it

iso.org/iso/home/store/catalogue_tc/…

And when they wanted to have this standard they contacted the mathematicians so @barbarabeeton is right about the product owner but not about the content owner. It is a very important practice that you don't do things on your own.

barbara beeton

@percusse -- what was the situation/membership of the committee around 1978? iso 30000 simply inherited what was already in iso 31, and (unless what i learned when i was active in the working group that spawned the sgml standard has changed radically) iso/iec committees are very conservative when it comes to changing what was in a predecessor standard or edition.

percusse

@barbarabeeton

No engineer bothers to add standard number sets ;)

and the intervals for units are specified pretty strongly in specialization standards.

barbara beeton

@percusse -- fair enough. but i would still like to know what explicit "mathematical" participation was active for the first edition of iso 31. there never was a representative to ansi or iso from ams before i was invited to participate in 1984. i can try to find out if there was a representative from siam (applied mathematicians and pure mathematicians don't alway see eye to eye), but if you have more specific information, i'm interested.

percusse

18:03

@barbarabeeton I don't know the names but as far as I understand from the context, initially, this was an issue in technical drawings where most of them were drawn by hand and annotated by hand too.

barbara beeton

as stated in the foreword of iso 31, i note that canada "expressed disapproval of the document on technical grounds" (which are not specified).

percusse

So they wanted to have a typographical standard for say greek nu which is a big deal in fluid mechanics and pipe engineering and v for velocity.

That's why the standard is written super-weirdly. Because it is meant for another breed.

But after the plotters etc. this should have been decomissioned in my humble opinion

I know basically no organization actually used it in anywhere

However, I know for a fact that engineers will not define bessel function standards themselves.

barbara beeton

the need for standards in technology isn't in question. whether the same practices are followed for pure mathematics, however, is the issue here. i claim that they're not, and that there's no real conflict, but a lot of argumentation and (in my view, wrongheaded) opinions that practices must be identical.

percusse

@barbarabeeton They can't even decide on imaginary number i or j so I am pretty sure no engineering standard secretariat on their right mind will attempt to do this by themselves.

barbara beeton

@percusse -- engineers aren't really interested in theory, only that there is a practical approach that is known to work reliably in real life situations.

percusse

18:08

But identical practice is a different issue

For technical drawings for example, I witnessed very strange problems say, regarding the 3rd view/1st view standard

That is this stuff

barbara beeton

@percusse for i and j, i will try digging into the nist document, to see what's specified there.

percusse

this is european way of drawing it

american standard says you have to swap the projections

so it's not as physics/mathematics one rule to rule them all.

These actually go into international contracts so they are a big deal

But I have never heard of 80000-2 being imposed on any legal binding

barbara beeton

@percusse -- no contest on this; just like the metric units should have been more explicitly specified on the failed mars lander. but the problem is that "laymen" can think that following international standards is mandatory, and can get very shirty about someone following a different (and traditionally accepted in a specific field) practice.

@percusse -- maybe not legally finding, but the "conversations" can become deafeningly heated.

percusse

@barbarabeeton ISO standards are usually pretty comprehensive and usable. This one and actually the whole 80000 family is in my eyes cash-cows of ISO.

They probably know that nobody would care at this point in time. But maybe some person as you described can be lured in for money

@barbarabeeton You can always say no to certain parts of the requirements.

barbara beeton

@percusse -- "cash cows" is a good characterization. which is why i haven't got a copy of the current version. but ams has its own "standards" dating back to the late 1800s for the usage of certain notation, and is not about to change now.

percusse

18:19

@barbarabeeton But on the other hand, I have to say that I consider, system theorists and some ee engineering departments as mathematicians too ;)

Haha, and see who took the bait :D

8

Q: What are the main changes between ISO 31-11 and ISO 80000-2? (math notation standards)

The international standard that defines mathematical signs and symbols, ISO 31-11, was superseded in 2009 by ISO 80000-2. What are the main changes?

notation

Oh my... 160 USD....

Well played ISO

barbara beeton

18:33

i've now checked two "reliable" sources regarding "e" for "exponent". ellen swanson's "mathematics into type" (1972, first edition; based on existing practice of ams and oxford university press) specifies italic "e" or roman "exp". iso 31-11 (first edition, 1978) specifies roman "e" or roman "exp". i don't know where the roman "e" originated. i will guess that if from mathematicians, they were outside the english-speaking tradition; but i can more easily believe engineers or physicists.

yo'

@barbarabeeton May be physicists, by the idea that "all constants ought to be roman"

Christian Hupfer

18:51

@yo' Insulting Physicists, hmmm? Naughty Tom! :D

Paulo Cereda

What have the Time News Romans ever done for us?

yo'

@ChristianHupfer funny thing is that you write the Euler's constant upright, but the gravitational constant italic, because the first one is certainly constant, whereas for the second one, we do not know.

@PauloCereda Someone has messed up your esses, man!

Paulo Cereda

@yo' Oh noes! :)

Christian Hupfer

@yo' Yes, there are theories about a time-dependent gravitational constant or even spatial fluctuations ...

yo'

btw, having a large candle on the table as the lighting and listening to Mozart's Requiem. What an evening!

Christian Hupfer

18:56

@PauloCereda Gave us a working alphabet ;-)

Using randomized enumeration labels is 'easy' but providing the correct reference way to them is a little bit harder ;-)

percusse

@barbarabeeton what makes you think that mathematicians follow those?

@barbarabeeton With all respect, everybody picks their own hero and copies their notation.

yo'

@percusse with you hero mostly being your calculus teacher, needed to say :)

percusse

@yo' Not really. I avoid math notation like the plague. My hero is what current font spits out in the TeX document :)

yo'

@percusse ah math notation and TeX! I have a curiosity question: If a display equation is a part of a sentence, should I use the text comma or the math comma at the end? Maybe @barbara knows what to do (or what DEK does). My text font is Palatino, my math font is Euler, and the comma is very different in the two.

percusse

I remember heated discussions on where should the adjoint sign should be put for a complex valued matrix G^*(iw) or G(iw)^*

yo'

19:05

@percusse ah the thing I denote $G^\dagger$ because $G^*$ is reserved for the Kleene star :-)

percusse

@yo' Consistency dictates it should be a text comma since it is part of the storyline but not a set notation and so on. But I say, as you wish :)

yo'

@percusse this I know, but it knows to look bad :D (I actually used the math comma everywhere, until I got to an equation with complicated cases and a lot of text, and ... I was surprised that something's gone bad with my typesetting ...)

(needed to say, I may be one of the few people who use the Kleene star on things like matrices...)

percusse

@yo' I think this whole should be/shouldn't be stuff is artificial. Do whatever pleases the looks and make sure it is not confusing

And somebody will definitely get confused so screw it :)

yo'

@percusse LOL, it's better to submit an empty thesis probably then :-)

percusse

@yo' Yes that would be ideal if there is no result :)

yo'

19:09

@percusse there is a result, in each chapter of the (empty) thesis at least one ;-)

percusse

@yo' I recently had a discussion with a music nerd about music theory which doesn't exist for me.

yo'

@percusse music theory is only as good as how much it can help you do practical music.

percusse

The whole point is that western music is structured on certain things which are considered to be nice. But when you take a mathematician in, shit goes sideways and suddenly you have rules.

You can't play that with this, this is where it gets resolved, that creates tension

Take any indian piece and analyze that, I guess that would be the non-euclidian music theory

yo'

@percusse yep. Like the rule of no 4th and 5ths -- so stupid! On the other hand, the hint that a 3rd makes your music "softer" is nice.

percusse

where circle of fifths don't repeat :P

spirals FTW

yo'

19:12

@percusse :)

barbara beeton

@percusse -- mathematicians may or may not have followed those, but compositors surely did, backed up by careful technical editors. at least until mathematicians started typing their own manuscripts. and i have another published source, though not readily at hand, a translation of a manual published in czechoslovakia in the 1950s, that says the same thing.

yo'

@percusse LOL

@barbarabeeton I wonder what Czech manual it could be and whom by...

percusse

@barbarabeeton I think you are still looking from a mathematics publishing editor point of view.

barbara beeton

@yo' -- in text, all commas should be the same shape. so the text comma should be the "master". (which is why it is a bad thing to mix fonts. and one reason why punctuation in theorems should be roman.)

percusse

I can't even read what Partial Differential equations people publish on a math journal, though I know they are doing sparse least-squares and they say it in the text and they are mathematicians.

Because they adopt a different set of rules arrow vectors, underline matrices etc.

SIAM Lin Alg App doesn't have that

yo'

19:16

@barbarabeeton Well, thanks! I'm not sure about the "roman punctuation", but I'll think about it. It's a tough thing to be typographically consistend and try to create something nice and at least slightly unique at the same time.

barbara beeton

@yo' -- the author was karel wick; i don't have the czech title readily available.

percusse

So I can't believe that every mathematician has a reference manual for typesetting and they never romanize e

yo'

@percusse I do romanize e, for one. Oh wait, I don't use it :D

percusse

@yo' I think we should go back to hand-writing

yo'

but I use \mathrm{i} for the imaginary unit.

@percusse I've had handwritten figures in my master project! And I was considering using the xkcd style for some of my drawings in the thesis.

barbara beeton

19:19

@percusse -- yes, admittedly true. but the discussion has been about math and mathematicians. and i fully recognize that pure and applied mathematicians (never mind engineers or physicists) may have a different perspective. (which may be one reason that i ended up with a degree in linguistics rather than in applied math.)

yo'

@barbarabeeton ah you should have gone here: lingustics are studied at the mathematics faculty!

percusse

@barbarabeeton That's why I'm having difficulty understanding why you have difficulty of believing it might be indeed mathematicians that were involved in the standards.

There is an implied we don't do that kind of nonsense in this discussion :)

And yes they do. A lot.

barbara beeton

@percusse -- most of them (who use tex) probably just do what's easiest, and accept what tex provides, which is from a pure mathematician's perspective. check out the references that knuth used, listed in his gibbs lecture article, "mathematical typography" (ams.org/journals/bull/1979-01-02/S0273-0979-1979-14598-1)

percusse

@barbarabeeton Aha, let me check them. Very interesting thanks!

Here in return :P

Stephen Fry Kinetic Typography - Language

►

2

barbara beeton

@percusse -- my difficulty in believing that mathematicians had anything to do with iso 31 is because when i started working on compiling symbols for the stix project, to be added to unicode, i asked the person who provided the printed copy to me, who was on the committee at the time of the 1992 revision. and he said no, no mathematicians that he was aware of.

percusse

19:28

@barbarabeeton We also equally don't know if engineers are involved either. Still the implication is there :)

Somebody needs to be the legal body if something goes wrong. ISO cannot say, well our engineers are really good at math. So I would believe the other way around which is lightly backed up by some anectodal evidence on my part

yo'

@barbarabeeton he's a genius!

barbara beeton

@percusse -- that's quite delightful! thank you! (what i strive for, as editor of tugboat, is prose description that can't be misunderstood in a damaging way. if i am not sure i understand something, i query the author, explaining how i understood the words on the page, in the clearest way i can. if the author agrees that i "got it right", then i leave it. if not, we work out a mutually intelligible compromise.)

a particular problem i have (and i admit it may be just my problem) is, what time is "12:00 pm". i say it's midnight. a lot of people these days understand it as noon. so in that situation, there's a reasonable alternative: just call it "12:00 noon", which nobody can misunderstand. math notation isn't that easy to tie down, but there's a (usually unwritten) rule that any mathematician can define his/her own notation, as long as it is defined before it's used, and used consistently.

yo'

19:44

@barbarabeeton no, not again the 12:00 pm and 12:01 pm business :-)

barbara beeton

@yo' -- sorry. i've just been subjected to a spate of calendar quibbles. i hate them. there's a good workaround; just use it! there are actually two good workarounds; the 24-hour clock was invented for a reason.

yo'

@barbarabeeton it's just that for me, it sounds wrong to have 11:30am, 12:30pm (change to pm here), 1:30pm (subtracting 12 here)

@barbarabeeton but even that is not perfect. My microwave shows 0:00 at midnight, but my oven shows 24.00 :-) And a fun fact, in the printed train timetable, an arrival at midnight is marked 24 00, but a departure at midnight is marked 0 00 (the colon or dot is omitted for clarity)

Also, as a fun fact, the Czech rules prescribe dot as a minute separator

barbara beeton

@yo' -- there's actually some perverse sense to that. an arriving train is completing it's journey, whereas a departing train is just beginning. i think back to a flight from japan to the u.s., where according to the schedule we would arrive before we left! which is why astronomers stick to "universal time". (sometimes you just can't win.)

yo'

@barbarabeeton ah yeah, the (now called) Zulu time :-) And I do understand that logic. Actually, the midnight train from Prague to Budapest officially leaves at 0:01, which is IMHO good, you have to get a seat/bed reservation, and this way you are 100% sure which day you book.

barbara beeton

@yo' -- that's the practice among u.s. insurance companies, starting a policy at 12:01 a.m. has probably saved a bundle on legal fees.

yo'

19:57

@barbarabeeton but again, I would be confused by 12:01 am, because it's "2 minutes after 11:59 am"...

But again, our law is probably a bit more rural than the American law; here if a long-term contract starts at 5th of October, it's meant "at the beginning of that day", unless stated otherwise. Similarly if anything ends at 8th of October, it ends "at the end of that day".

barbara beeton

@yo' -- 00:01 would be clearer, but if you live in the u.s. long enough, you get used to the discontinuity. i get flak for referring to time by what's considered here "military time", but i really don't want to show up at noon for a midnight concert or show (of which there are a reasonable number in these parts).

yo'

@barbarabeeton :-)

Now excuse me, it's time to practise for the Sunday service, so that the songs settle down over two nights, rather than just one. So have a nice weekend and bye for now!

Kurt

23:24

What to do with a migrated question, the op never logged in and it is rather old. Closing as unclear?

1

Q: Two separate yet connected bibliographies in one document in LaTeX

I'm writing a thesis in LaTeX, and what I need is bibliography split into two separate parts: general citations and own citations (i.e. those positions I'm author/coauthor of). Those parts should have consistent numbering, so if general citations end with number 83, my first own position should s...

bibliographies

Johannes_B

@Kurt Voted as unclear.

Kurt

@Johannes_B Okay, Thanks. I will vote too.

Transcript for

$TeX - LaTeX$ TeX, LaTeX and Friends