@WillRobertson redefine what oh you mean locally redefine DeclareTextSymbol to ignore the encoding argument and use something else? tempting but I think that would be harder to pull into some future unified framework, I think I'd rather use a macro in place of {TU} so you only need to redefine that, I think, perhaps...

@DavidCarlisle Do you prefer \UnicodeEncodingName or \LastDeclaredEncoding?

@WillRobertson depends on time of day:-)

@DavidCarlisle I guess the main decision is whether \DeclareFontEncoding goes in the tuenc.def file or not

Would you be against \DeclareFontEncoding{\UnicodeEncodingName}... ?

@WillRobertson i think \UnicodeEncodingName as it solves the "where to put the \DeclareFontEncoding" question and it also more obviously hints that it has to be unicode, if \LastdeclaredEncoding is OT1 and you input tuenc.def it's going to be unhappy

@DavidCarlisle I agree! :)

So I think that's mostly that?

@WillRobertson I could just commit tuenc with \providecommand\UnicodeEncodingName{TU} and replace every {TU} by \UnicodeEncodingName and then you can see if you can fit that in with fontspec

@DavidCarlisle I can guarantee that'd work well for me. Keeps things simple for you too, I hope.

@WillRobertson expect a commit email in a few minutes..

@DavidCarlisle Thanks — over the weekend I'll try to sort out the interface to fontspec (regardless of l3 plans...), mark it HIGHLY EXPERIMENTAL, and be ready to send to CTAN. Still not sure how the synchronisation will work — I guess I will need to add some redundant code and a warning message for people who update fontspec but not 2e.

$ texlua build.lua check -e xetex

@WillRobertson ^^^ just in case...

@WillRobertson you've got mail:-)

@AlanMunn That's funny. That was written by Archimedes Plutonium - an infamous mathematical crank. He said similar things about John Conway, Andrew Wiles and many other mathematicians who refused to answer his cranky queries (e.g. about his claimed proof of the Twin Prime Conjecture). Anyway, you are welcome to read my answers on math.SE and judge yourself. Constructive feedback is always welcome.

@yo' -- well, we're back in boston, where the temp is just about freezing. the flight that was supposed to leave just before ours -- to frankfurt -- was canceled because of weather, but ours -- to munich -- flew as scheduled. nice to be back in the home time zone. supposed to get to about 10 c tomorrow. weather is crazy.

@BillDubuque ? Some context might help me here. I haven't been in chat for days, so I'm not quite sure what this is replying to.

The little arrow at the start of the comment takes you to the comment being replied to. When you dig up ancient comments in Google searches you should pay attention to the author. Like that, the web is full of nonsense authored by cranks, crackpots, etc (and some of them know the tricks to make their nonsense get high Google page rank, like stating it as the title of a sci.math thread - as there)

@AlanMunn That's the first time I saw anyone quote that! Anyway, no offense taken. Actually, I found it quite hilarious that someone quoted an infamous crank's opinion on that matter.

@PauloCereda Progress in the delivery: "Object left the International Unit" yesterday at 17:38 :)

@BillDubuque wow that's a name from the past, I used to enjoy reading those posts in comp.math newsgroup (I suppose it was?) (for some warped definition of enjoy:-)

Aditya is performing the ConTeXt flood :)

@percusse a flood would be if you were to re-tag all the tikz-pgf questions as picture-mode changing a few context questions is more of a trickle :-)

@DavidCarlisle Something very close has been done by lockstep. S/He practically retagged the whole site

Thanks to those efforts now our tag system is pretty decent

@percusse yes that was happening at about the time I (re)joined the site I think.

@percusse actually I am not sure why people worry about this effect of edits/tags I almost never use the "active" tab which is I think the only place it affects? I just use the "newest" tab as a rule.

@DavidCarlisle There are many things I still wonder why people ever bothered about this place :)

@percusse true

@DavidCarlisle Well, you have to actively go to Questions first, on the front page you don't have "newest".

tag wiki edit badge is my favorite :)

@TorbjørnT. I am not sure Ive ever been to the front page of any sx site:-) I either have a browser tab open already or I land on some question via google search

Anoybody is using Accelerate on OSX here?

@DavidCarlisle They are meant to help not so much the 'active' people more the 'occasional searcher', plus of course on the main site tags are rather more important (they govern what you see at all on the main page)

@DavidCarlisle Do you have any opinion about this issue or better if you have time can you skim through this thread ?

github.com/scipy/scipy/pull/6051#issuecomment-265407699

@percusse hmm interesting, I have some opinions but not here, drop me a mail if you like (although I'm not sure what I can say in mail at the moment either:-)

@DavidCarlisle Sure, if I can test on an OSX, will surely do that

That's just crazy though the whole is stopped because of either Microsoft or Apple all the time

@DavidCarlisle tex.stackexchange.com/a/348110: is this about right?

@DavidCarlisle Another reminder we need to 'get on with this' (I'm in favour of forgetting all the clever pre-parsing and the like and going with a ConTeXt-like 'all keyval' model, as you can probably guess)

@JosephWright yes although I was a bit late reading that as @percusse had diverted me into reading about python interfaces to fortran libraries, which coincidentally enough just about mirrors the larger state of latex development here:-)

@DavidCarlisle 'Get on with it'?

@JosephWright get on with python or get on with latex?

@DavidCarlisle The thing with xtemplate is all of that re-direction is I suspect a waste of time :(

@DavidCarlisle I meant that they presumably have the same overall statement as we do :)

@JosephWright you mean other than a basic xparse like separation between a user-interface template and the underlying code?

@DavidCarlisle Possibly: certainly the bit in xtemplate where we have a declaration of keys then a separate declaration of their associated values and then a point where they are assigned, and internally a fixed key ordering, and 'pre-parsing' the keys to have the values extracted, is all a bit too much I feel

@DavidCarlisle Ahaha

@DavidCarlisle Taking the question I just linked to, it seems more sensible to allow inheritance or or allow extending templates than requiring entire new ones, but at that stage 'why templates at all'

@DavidCarlisle If we look at beamer, it goes with a simple idea of a template for 'some named part of a document' but with no further set up at the abstraction stage: OK, you could argue it's a special case but then you need that plus 'articles' plus 'chapters' to cover most use cases, so ...

@JosephWright Just a side note: I think that decision is heavily affected by the absence of pgfkeys at the time

@percusse Yes, sure: I'm probably thinking somewhere 'in between'

@percusse Package options has some interesting ideas about inheritance which are inspired by CSS

@JosephWright Ah yes. And I think yax is extremely underrated.

ctan.org/tex-archive/macros/generic/yax

@percusse There are several ideas in the same area: one can do inheritance, etc. in various keyval packages. Partly depends how you want things to work: are keys executed when they are encountered or stored as values. (This came up on the team list recently: @DavidCarlisle pointed to the angle/scale combination in graphicx as a case where the latter doesn't work, at least with the interface as currently described.)

@percusse In any case, the template business is slightly different in concept: that starts from how you conceptualise documents. Then we have the CSS-like questions: how does one describe the relationship between elements and their context-dependence (e.g. section in appendix in minipage ...)

@percusse I'm really hoping to push the team on this area this year: part of the reason for looking over beamer is that whilst it's model is simple we know that it largely works

@JosephWright yes must admit I'm a bit hazy on exactly the current code, but in the original version we kept trying to simplify it but we had a set of use case examples centred around templates for tables of contents and headings mostly, and every time we simplified something one of the use case tests failed, and every time we tried to add another use case, we had to complicate the code to make things work, mostly that lead to a feeling the model wasn't quite right and things stalling (again)

@DavidCarlisle Quite possibly: we were after use cases last time this came up. Once we deal with TU I'll raise on the team list: no point just at the moment.

@DavidCarlisle The beamer side of Leslie's issue is fixed :)

@JosephWright yep just saw the mail:-)

@JosephWright are you going to tell him?

@DavidCarlisle I might tell Frank it's sorted and leave it to him to decide (I don't know Leslie, after all)

@JosephWright I can make the rest of it work but not sure where best to do it, just need to make[a,b][c,d] be viewport=a b c d instead of bb=a b c d but (a) the viewport code isn't defined if only graphics is loaded (it's not much code but pdftex.def might be loaded by eg color.sty before I know what if graphicx is to be loaded) and it's driver specific but I'm not sure if the .def should be changing top level interface definitions (\Gin@iii goes straight from [][] to setting the bb registers)

@JosephWright I'd just leave it this release and think about it, but an undefined command error isn't good.

@DavidCarlisle Yes, true

@DavidCarlisle Let's be honest, there are very few people using the graphics interface (cf. lack of beamer bug reports and fact that keyval methods do actually work quite well!)

@JosephWright and the fact that any attempt to do this has given an undefined command for last 5 years or so (and has never worked since pdftex was introduced) but still now I know it generates an undefined command error.... (well i can \let it to empty so the error goes but then you just get no error and no image

@DavidCarlisle :)

@DavidCarlisle Same with the beamer code (though there are other issues this links to there which do need fixing)

Anyone have a comment on this

It is rather strange: it works in xelatex and lualatex. Produces errors in some versions of AR if compiled with pdflatex. Reverse the other and the error disappears. Insert a paragraph break between them and the error in AR disappears.

Normally I'd just ignore these errors from AR on Linux (because it is very old), but the OP gets them on Win7 with a much newer AR

@daleif the existing answer looks good, if the pdf you are importing is broken then pretty arbitrary whether it works or not.

@DavidCarlisle we are also just trying to understand what is broken, and how can a user test for it.

@daleif I saw later you had moved to chat (didn't follow) if the answer is uncompress the pdf stream and spot by eye a mis-matched q Q graphics state restore then I guess the answer is most users do not have a hope of testing for it.

@JosephWright I can't fix graphics now as ltnews26 is full.

@DavidCarlisle I think Alex just tested that and it did not work correctly. Might have been the wrong pair.

@daleif ah

Texmaker keeps resetting its settings every time it's closed. Should I ask here or in Ask Ubuntu? The ini file isn't being modified by the program.

@JosephWright Can I ask you a (possibly quick) biblatex question?

@AlanMunn You can ask: I'm about to head home so reply may be slightly delayed!

@JosephWright I have people with names like Yi-Hui whose proper initial should be not just Y. I tried just putting {Y--H} in the bib entry, but it still shows up as Y. This is with biblatex-apa.

Something amusing for @JosephWright: Someone was submitting to an ACS journal with your template. The editors decided that the formatting of the TOC was all wrong and made them redo it. They were amused that even the official template was not official enough.

@Canageek Oh, I get this all of the time: I don't really know why editorial offices worry about it (the typesetters redo everything anyway)

@Canageek I've just had a mail today from the ACS concerning a journal they forgot to tell me about :)

@JosephWright shrug

@AlanMunn Hmm, biblatex-apa is PLK

@JosephWright Yeah, I know. Do you know if out of the box biblatex does this too?

@AlanMunn This comes down to the formatting function not the bibliography database. For example

\RequirePackage{filecontents}
\begin{filecontents*}{\jobname.bib}
@Article{Redshaw2014,
  author =    {Ma, Jing and Zhao, Ke-Qing and Walton, Mark J. and Wright, Joseph A. and Frese, Josef W. A. and Elsegood, Mark R. J. and Xing, Qifeng and Sun, Wen-Hua and Redshaw, Carl},
  title =     {Vanadyl complexes bearing bi-dentate phenoxyimine ligands: synthesis, structural studies and ethylene polymerization capability},
  journal =   {Dalton Trans.},
  year =      {2014},
  volume =    {43},
  number =    {22},

Done another draft of the paper we are sending to Dalton Trans.

@Canageek Cool

Now just need to hunt my boss down and give it to him

@Canageek :)

@Canageek I've just moved office so my group can't escape me: door is opposite the lab

@JosephWright We don't want to escape Danny. When we had a larger group there would be lines outside his office

@JosephWright Me and another student once might have put a couch outside his office as a joke, for us to wait on. Undergrads kept making noise on it though so they made us get rid of it

@JosephWright Yes, I know that; I was just trying to temporarily fool it with braces, but it didn't work. It's no big deal; I'll look into it later. Thanks.

Who was surprised when I mixed Meter and Litre? I think @AlanMunn or @PauloCereda? I am taking a poll: strawpoll.me/12079264/r

@Canageek Hmm. I'm so far the only one who uses 'metre' and 'litre'.

@AlanMunn Are you Canadian?

This is supposed to be Canadian only

@AlanMunn In other words, you can actually spell ;)

@JosephWright Canada: The only country where both Americans and Brits look at you like you are crazy when they read your documents.

Another way to spot a Canadian: I think we are the only ones to mix imperial and metric in the same sentence.

@Canageek you forgot Aussies?

@yo' Are they halfway between American and British English?

@yo' Canadian: Colour, Jail, Meter, Litre, Zed

;)

@Canageek well, they certainly are half way somewhere, but who knows where?

Zed as for the letter Z?

@Canageek Mainly UK-like then

@Canageek Oh, we do that too, certainly if you are my age (39): I do 'domestic' stuff in imperial, work obvious in SI units

@JosephWright also, IIRC, Celsius for weather, Fahrenheit for cooking

@yo' Nope, Celsius for weather, gas mark for cooking :)

@JosephWright with a convection/electrical oven?

@yo' No, they tend to have Celsius markings but I tend to think of the equivalent in gas mark

@yo' My flat has no gas so the oven is electric, but at home we do have gas

@JosephWright ah ok.

@JosephWright my flat has got gas, but the cooker/oven is split: gas cooker, electric oven

@yo' Yeah. We named a nuclear reactor Zero-Energy Deuterium-2 mostly to force Americans to say Zed

@Canageek :)

@yo' Yeah, since all our ovens and cookbooks come from the US. I have no idea what 350F is, aside from the temperature I set the oven to.

Oh and we use measuring cups like in the US, not kitchen scales

@Canageek so it's Zed for Britons and Zee for US?

@yo' Yes

@Canageek I use both, since me and @barbarabeeton have met on TUG15 :)

@JosephWright ok, will remember to keep saying Zed as I'm used to from Czech

@JosephWright We've adopted a lot of American spellings though. Jail instead of Gaol, and to us a jumper is something you use when your car won't start....

@Canageek 'Jail' is normal in the UK too (looking it up, 'gaol' has not been the main spelling for 100+ years!) (As a chemist, of course, I tend to think JEOL!)

@JosephWright jelly? me likes :)

@JosephWright Ah, OK. A lot of the British books I'd read (Other then the first Harry Potter) would be about 100 years old

@Canageek I've never head gaol, TBH

@yo' jeol.co.jp/en

@Canageek 'Classics' I guess would have 'gaol'

@yo' You wouldn't: It was pronounced identically.

;)

@JosephWright funny enough: is JEOL an abbreviation? If so, it's expansion is not mentioned on the homepage :)

Canadians use -our (British) but -izes (American)

honour, colour, theatre and centre

@Canageek a number of British use -izes FWIW

But center is also used by many people

@Canageek does it depend on the part of speech?

@yo' Technically, both are allowed in UK English but I'd tend to '-ise' unless writing for a US journal

@yo' Nope, pretty much random.

@yo' Depends if you've been reading too many American papers as of late

lukemastin.com/testing/spelling/cgi-bin/…

@JosephWright yeah

Also it is math, not maths. That just sounds weird.

Eh is falling out of use in most of the country though

Also we have no idea WTF a Freshman or Sophomore is.

“connexion” and “inflexion” -- damn, I wouldn't write that other than maybe in calculus speaking about inflexion points

@Canageek these labels are plain stupid

Oh this one confused the heck out of Mara

"In speech and in writing, Canadian English speakers permit (and often use) a transitive form for some past tense verbs where only an intransitive form is permitted in most other dialects. Examples include: "finished something" (rather than "finished with something"), "done something" (rather than "done with something"), "graduated university" (rather than "graduated from university")."

"Some nouns, as in British English, take -ice while matching verbs take -ise – for example, practice and licence are nouns while practise and license are the respective corresponding verbs. (Note that advice and advise are universal across all forms of English in this respect.)"

Also: Curb and Tire, not Kerb and Tyre

and I see my feeling and preference is a good mixture: colourize but analyse, modelling but woolen, ...

holy crap! Someone really writes judgment?!

@yo' ....how else would you spell it?

(I shall probably not be judgemental about people who spell it judgment)

@Canageek Indeed: what's wrong with 'first year' or whatever

@JosephWright the same as having the numbers on a coin rather than just "DIME"

@Canageek Ah, didn't know that ('curb' to me is a verb, 'tire' is almost as bad as 'color')

@yo' How would you spell it?

@Canageek judgement

@JosephWright Worse: Nature now uses Sulfur instead of Sulphur.

@yo' Bleh.

@Canageek Yes, that's IUPAC unfortunately: here, most of stick to 'sulphur' if the document is internal, and only use 'sulfur' for stuff going for publication

@JosephWright Question: Do you know what cutlery, eavestrough, a two-four, and toque are?

@Canageek Cutlery = knives, forks, spoons, etc.

@Canageek Others I don't know

@JosephWright eavestrough: The thing that runs along the edge of the roof & catches rain.

@Canageek Ah, 'guttering'

Two-four: Case of 24 bottles or cans of beer

@JosephWright A gutter is the thing at the edge fo the street

@Canageek Not my area of expertise

@Canageek This is also true

@JosephWright Toque is a warm, knit cap.

Worst Americanism that is like nails on a chalkboard: "Niche" rhymes with "leash" not "itch"

@Canageek I take is as that I can pronounce things any way I like :-)

anyway, I caught some nasty cold, so I'm better off to bed.

@CarLaTeX: YAY

@Canageek Americans don't know French, Canadians do. ;-)

@egreg Ehhhhhhh, we pretend we do. Most of us don't remember much of it.

@Canageek Je rappel très bien mon français (I studied it only in middle school)

@egreg I was so bad at it due to a learning disability that I had to be exempted.

@Canageek At least you didn't have to cope with imperial unit conversions.

@egreg Nope. The worst I had to deal with was km/h to m/s and back. Seriously, who decided base 60 was a good idea?

/me thinks we really need to fix the time system to be base 10

@Canageek Ask Ptolemy and the Babylonians.

@egreg Exactly. Time to fix it to something useful.

@Canageek The French commission for weights and measures tried to make decimal time, they didn't succeed. I saw a “decimal clock” in some French museum.

@egreg Yeah, that is one thing the French have been good at, SI units.

@egreg I think they founded IUPAC as well