12:34 AM
@Sebastiano -- the original edition of this book was published in 1965 by academic press. it was almost certainly set with metal type. the present edition is published by dover. when dover publioshes a new edition of a book originally published by one of the traditional "major technical publishers", they almost always use the existing copy and patch in updates. so it will be very difficult to match the fonts used.

6 hours later…
6:57 AM
@barbarabeeton Hi, I have understood. I remove my question. Thanks
@DavidCarlisle Hi Sir David I will delete my question.

@Sebastiano I see you've knighted @DavidCarlisle :)

@JosephWright Yes of course :-)-Some time ago when I was writing Dear David he told me and I remember very well writing Sir
@TeXnician Good morning and welcome into chat

@Sebastiano Good morning.

8:02 AM
@Sebastiano up to you, I didn't say you should delete, I just said I wouldn't answer it.
@JosephWright don't you think that would be a good idea?

@DavidCarlisle :)

8:18 AM
Microsoft officials have been talking to GitHub about possibly acquiring the company, according to a June 1 report in Business Insider. BI claims that the two have discussed the possibility of an acquisition on an on-and-off-again basis over the years "but in the last few weeks talks have grown more serious." BI is citing unnamed "people close to the companies" as its sources. "This isn't as surprising as it would have been ten or more years ago," writes The Welcome Rain. "Microsoft is investing a lot in git, including GVFS, a Git Virtual File System to help Git work with very large codebas
Uh-oh.

@PauloCereda GitLab to the open-source rescue ;)

@TeXnician github has never been open source, I can't see it necessarily makes any difference if commercially it's owned by one big company or another even bigger one.
@PauloCereda stackexchange next
@PauloCereda all the interweb gossip talks of acquisitions and buyouts except the register which suggests a git merge :-) theregister.co.uk/2018/06/02/microsoft_github

@DavidCarlisle I meant more for the open-source projects. I know that GH is proprietary. But let's say the bigger the company the more problems might occur (speaking of this specific company).

@TeXnician If problems occur repos can be hosted somewhere else, but I would not do that just because of an ownership change that's just following some historic bias against one company

@DavidCarlisle Not yet ;)

8:32 AM
@TeXnician Indeed :)
@DavidCarlisle ooh :)
MICROSOFT GIT
@marmot: mr. marmot, when you have some time, can I ask you about Asymptote?

@PauloCereda makes handling latex merge conflicts seem not so bad "large merges across branches (10,000’s of changes with 1,000’s of conflicts). " :-)

@PauloCereda Which vim-plugin for TeX do you use?

@DavidCarlisle oh :)
@DavidCarlisle amateurs these MS blokes, don't you think. :)
@JohnDorian none. :)

8:49 AM
@PauloCereda interesting :-)

2 hours later…
10:21 AM
4
@CarLaTeX @egreg ^^^ One of the local bears celebrating today's „Festa della Repubblica“ by climbing our fridge. He couldn't get to the Dolomiti in time.

May your new government be wiser than some fear!

10:42 AM
@UlrikeFischer Did you see the parachutist with the flag?
@UlrikeFischer At least we have a government, I'm not very confident about its duration :)

10:55 AM
@CarLaTeX No. Where?

11:15 AM
@UlrikeFischer At the ceremony in Rome

12:05 PM
@UlrikeFischer ooh Die Bär

12:29 PM
@CarLaTeX Bella figura! - as was to be expected ;-) Our celebrations will happen tonight: Ossobucco alla milanese.

12:59 PM
@UlrikeFischer Tipical Milanese dish :):):)

1:36 PM
@DavidCarlisle, @JohnDorian -- would like your opinion of the second comment on this question. it's in response to a comment i made in 2013. i think it's misguided, and could both cause overload problems and not have the intended result. but i'm not able to phrase that cogently or accurately, and am not inclined to experiment. can you help?

@barbarabeeton well the comment is true as written, but it's like saying that carrying a crutch might mitigate the bad effects of shooting yourself in the foot.

@DavidCarlisle -- but wouldn't the whole thing have to be repeated every time that object was invoked? that would require yet another command definition to make it convenient. (or am i missing something simple?)

@barbarabeeton yes but it can make sense, if you have a chapter in an external file already written that uses lots of weird commands you had rather not globally redefine eg because they mess up the bibliography then if editing to change the names seems too hard using \renewcommand locally so it only applies to that one chapter may make sense.

@DavidCarlisle -- okay, thanks. that's reasonable, but it's really a very special case. the comment seems to be saying it's a good thing generally, an opinion with which i heartily disagree. (i'm really rather sick of having to make repairs to long documents because an author has tried to be too "smart".)

@barbarabeeton yes as I say it's better not to do it at all rather than discuss ways to localise the damage.
@barbarabeeton \renewcommand\def{\operatorname{def}} ...

1:50 PM
@DavidCarlisle -- lovely. just lovely.

@barbarabeeton I have a feeling I've seen a real user do that, but couldn't find a reference just now:-)

@DavidCarlisle ooh

@PauloCereda YAY

@DavidCarlisle YAY

@barbarabeeton Me too. But it's worse when the “clever” author turns out to be myself, a few years ago.

1:56 PM
@PauloCereda ‮ΥΑΥ
3
@PauloCereda my YAY is more international than yours

@DavidCarlisle -- i think i've seen it too. the \cses that hit us most often are \c, \i, \l and \L. the typesetting system ams was using before tex suffered a similar problem, when the company that created the system changed the code for undotted i to instead produce a subscripted zero. instant havoc in all transliterated russian names. so i've carried the scars for over 40 years.
@HaraldHanche-Olsen -- i think someone said something about shooting oneself in the foot?

@barbarabeeton I used to do all sorts of crazy stuff in the old days, especially when I thought nobody else would need to work with the source code. For stuff to be published in journals, I used to me more careful, however.
@barbarabeeton LaTeX makes a great foot gun.

@HaraldHanche-Olsen -- true, though i find word to be even more dangerous, or at least more unforgiving. (i've learned to debug (la)tex; word entirely escapes my comprehension.)

@barbarabeeton Word has mostly been so incomprehensible to me, I rarely try to do anything clever in it. That has saved me much grief, I think.

@HaraldHanche-Olsen -- unfortunately, i sometimes get given a word document to update. what happens is usually total disaster. the beginning and end of format changes aren't where they are most expected, so unless everything is plain, unformatted text, there goes the better part of a day when i have to try to make repairs.

2:10 PM
@DavidCarlisle oh

2:22 PM
@JosephWright any reason not to have ''[Things with TeX in their name are](FAQ-texthings)''. rather than ''[Things with TeX in their name are](/FAQ-texthings)''. I just realize (having synced my repo with texfaq) that links don't work in forks that are at project level so one level down.

@DavidCarlisle I'd noticed the same: I see no particular issue with changing it
@DavidCarlisle I'm taking a bit of a break from trying to sort the theme: I might ask Jonas about it. (On siunitx v3 at the moment: really want to get it done)

@JosephWright yes I noticed you'd gone quiet this week:-) what was the issue, I just junked my old fork (of tex-faq) and re-forked texfaq and enabled minmal theme it seems to more or less work, apart from no links working

@DavidCarlisle The minimal theme works, but I was trying to use minima as it's closest to where I think I want to go ... but it doesn't work :(

@JosephWright oh an l less, I see:-)

@DavidCarlisle Yup

2:31 PM
@JosephWright thanks

@DavidCarlisle Key is it has a simple header.html and footer.html that I imagine altering whilst keeping everything else ... but no matter what I do, it fails

@JosephWright as you may have seen, I did a couple of passes over the files during the week cleaning up various markup issues, and then a few content things I spotted in passing.

@DavidCarlisle Saw that
@DavidCarlisle Perhaps I should mail Jonas ...
@DavidCarlisle You making the change or should I?

@JosephWright the links? I can do it as I'm "there" at the moment.

@DavidCarlisle Cool
@DavidCarlisle The Slate theme is not too bad ..

2:36 PM
@JosephWright does it work? :-)

@DavidCarlisle I'll try in a fork: probably yes as it's directly supported by GitHub

@JosephWright If we plan to fiddle with the header/footer anyway getting something up that is "close enough" may be better then add stuff in a local _includes or whatever to over-ride (can't we?)

@DavidCarlisle Yes, that's what I'm hoping: just 'tweaks' on a base style

@JosephWright we could work from there I think for comparison I have minimal at moment davidcarlisle.github.io/texfaq.github.io/FAQ-newlang

@DavidCarlisle Just a question of choosing a look then editing the minor stuff we need, e.g. question ID
@DavidCarlisle For most of the standard themes, I think they just have default.html so we edit the 'entire' thing
@DavidCarlisle I think we want most of the screen space for content, hence thinking that the minimal theme isn't quite right ...

2:52 PM
@JosephWright is the text body any wider in the theme in your fork? doesn't look much different to me?
@JosephWright markdown is so annoying: why isn't Caveat in bold italic:-) in the last section of that page. I suppose it should be ** ...

@DavidCarlisle Maybe not ... I guess we just need to pick something (the 'no theme' set up is a bit tricky, for example it doens't pick up titles from the YAML)

@JosephWright yes, I don't mind much we can tweak later, the head section in the theme you have seems very large (and very black) but that could presumably be toned down a bit if you want to start from there rather than from minimal?

3:20 PM
@DavidCarlisle All doable, of course: really it's a question of how much effort we want to put in here in terms of appearance

@JosephWright minimal effort to get basic navigation in place so you can switch it to tex.ac.uk, it can always be restyled at a later date

@DavidCarlisle Yup, that's what I meant

@JosephWright as i say I don't mind, I'd probably start from minimal (the sidebar only appears on wide screens where the space is otherwise unused) but the slate one is OK as well, if we start from either we can add some faq-specific data (and you want to get back to siunitx:-)

@DavidCarlisle OK, we'll go with minimal and I'll fiddle about a bit with adding meta-data/adjusting naming. Then we can ask @StefanKottwitz about redirecting texfaq.org ...

@JosephWright OK, agreed:-)

3:36 PM
@DavidCarlisle What do we want for title/description (see github.com/pages-themes/minimal)

@JosephWright "The TeX FAQ" as title and "English language frequently asked questions on TeX" as description (depends what that looks like in situ, might be too long)

@DavidCarlisle Cool: working on a few layout things now

3:51 PM
@DavidCarlisle I've added the title and question ID to the FAQ: other than perhaps linking to the 'About' page from the LHS of the pages, I think we are good-to-go
@DavidCarlisle Time to ask @StefanKottwitz about texfaq.org
All: Feedback on 'new' FAQ set up at texfaq.github.io very welcome (to @DavidCarlisle and me, or on GitHub)

@JosephWright can we change the "visit my profile" link to a link to the github repo page, the profile page for an organisation level thing isn't so useful

Trying out Atom as an editor: it's pretty good, just doesn't have proper DTX highlighting ...
@DavidCarlisle We'll need to copy-and-edit default.html to do that
@DavidCarlisle We need a logo ... @PauloCereda?

@JosephWright probably we'll need to do that in the end anyway

@DavidCarlisle I suspect so

3:57 PM
@DavidCarlisle Indeed: I do wonder if we should get Jonas to suggest some 'LaTeX team like' feel

@JosephWright certainly if he's got any suggestions...

@DavidCarlisle OK: I'll drop him a line

@JosephWright Looks good, but I get a duplicated title on the home page, is that intended?

@TeXnician No, fixing it now

4:18 PM
@JosephWright forget that, it only does the "view my profile" on my fork as it's a user page, on organisation pages it uses <p class="view"><a href="{{ site.github.repository_url }}">View the Project on GitHub <small>{{ site.github.repository_nwo }}</small></a></p>

4:29 PM
@DavidCarlisle Still not exactly what we want ...

@JosephWright no. could copy the file as you said, may be worth doing that anyway, so that we can check the local one over-rides as expected

@DavidCarlisle Just need to fix the 'index' layout so we don't get the title doubled ...
@DavidCarlisle If we want 'About' to link from somewhere other than the top of the question list, we need a custom layout
@DavidCarlisle We could stick to the vanilla default.html and modify just for individual pages

4:44 PM
@DavidCarlisle We could keep the default layout ... would mean we don't have to worry about updating it

@JosephWright yes if we can keep the main layout and force the title index page it would make updates (or switching styles) easier, so just the title, and add a link to about (and contributing?) in the generated list?
@JosephWright I only get the title once on my fork, so there must be a user/organisation switch somewhere, unless you just fixed it?

@DavidCarlisle Should be fixed ...
@DavidCarlisle Something like that, but I do like the idea that every page has an 'About' link (some other themes do that automatically, so it's much of a muchness)

@JosephWright forced a refresh, fixed now thanks
@JosephWright if we copy the template and add an about link to the sidebar it's not going to be so hard to check for diffs later I suppose, although currently the sidebar links to the index and the index has an about link so it's not too bad

5:29 PM
@JosephWright @DavidCarlisle The new FAQ websites looks really nice! A small detail could be annoying: If one mouse hovers over a link, it turns bold which changes its width. This can cause a change in line breaks which could be annoying if the mouse is accidentally placed over a link.

@samcarter We're currently using a standard theme from GitHub: probably we'll look into a 'custom' look shortly

5:46 PM
@JosephWright Ok. Some links are also not clickable, for example on texfaq.github.io/FAQ-hypertex the link to [arXiv.org/hypertex/]. Is this a general problem?

@samcarter We've script-converted from the TeX original: some stuff is still being fixed (@DavidCarlisle has sorted a lot)

@egreg Do you have any suggestions about the best way to get to Siena from N. America? It seems that flying into Florence and then taking a bus is the best, but perhaps you have other choices?

Would a list of such link be helpful or are you able to fix them automatically?

@samcarter that's because markdown is a really stupid markup system
7
@samcarter there, I feel better now.

@DavidCarlisle They didn't name it markdown for nothing. :)

5:49 PM
@DavidCarlisle I know, I just fought with a beamer/markdown Q on the main site

@DavidCarlisle I think long-term we need our own files ... I'm going to chat to Frank about things before raising with Jonas

@samcarter a normal link is [text](http:...) but if you want the link text to be the url you can use the short form [http...] that is you can use the short form unless there is a ? in the second line of the preceding paragraph and there is a y in the month

@DavidCarlisle Short form is <http...> ...
@samcarter now fixed

@JosephWright Great! Thanks a lot!

@samcarter List or even better edit on GitHub and request a merge :)

5:51 PM
@JosephWright you generated [...] so I blame you:-)
2

a y in the month?

@JosephWright i fixed several of those so grepped for them but the ones I looked at worked anyway

@DavidCarlisle I think overall things have gone well, and we now can fix stuff (as just shown), which is a good thing :)

@JosephWright yes I know:-)

@DavidCarlisle Ah

5:53 PM
@JosephWright i think it has some heuristics for making things that look like url links even if they are not marked at all

@AlanMunn Florence and train or bus. If you find Pisa, it's essentially the same and it's easier from the airport to the train station.

@DavidCarlisle Probably
@DavidCarlisle Actually, is texfaq.github.io/FAQ-hypertex really useful today?

@egreg Thanks. So I I'll look for flights either to Florence or Pisa.

@JosephWright just fixed the remaining ones
@JosephWright I wondered about that, I think probably we do need a "historical interest" section. You can't really "update" that other than write a manual for hyperref
@TeXnician more likely than getting tex distributions for an amiga or TOPS20, about which we also have answers

6:07 PM
@DavidCarlisle We can add a 'Historical' category, perhaps omitting it from the standard list, and linking somewhere?
@DavidCarlisle Probably we should link to the change log somewhere ... oh yes, and I should add dates to posts

@JosephWright did you want your real address in the about section? (Or do you think a real person has more legal weight than some nebulous github group?)

@DavidCarlisle It's Impressum: legally required for Germany (and we expect to be visible there ...)
@DavidCarlisle e.g. latex-project.org/contact

@JosephWright never scrolled down that far:-)

@DavidCarlisle Same info is in by blog ...
@DavidCarlisle I could make a separate page I guess and link from About ...

@JosephWright no it's fine

6:48 PM
@JosephWright I answered a beamer question:-)

7:32 PM
@CarLaTeX vvvvvvv

@JosephWright @DavidCarlisle I made a permanent http(s) redirect from texfaq.org and www.tex.ac.uk, please check

@StefanKottwitz I get page not secure error for tex.ac.uk

@DavidCarlisle for https possibly (cert domain names), check http too
and try with deleted browser history/cache or a different browser

@StefanKottwitz tex.ac.uk redirects to texfaq.github.io and works
@StefanKottwitz same in ie/edge/chrome/firefox, with wget I get
$wget tex.ac.uk --2018-06-02 20:44:02-- tex.ac.uk Resolving www.tex.ac.uk (www.tex.ac.uk)... 78.46.26.59 Connecting to www.tex.ac.uk (www.tex.ac.uk)|78.46.26.59|:443... connected. ERROR: The certificate of ‘www.tex.ac.uk’ is not trusted. ERROR: The certificate of ‘www.tex.ac.uk’ hasn't got a known issuer. ERROR: The certificate of ‘www.tex.ac.uk’ was signed using an insecure algorithm. The certificate's owner does not match hostname ‘www.tex.ac.uk’ @StefanKottwitz with http it does: $ wget tex.ac.uk
--2018-06-02 20:46:21--  tex.ac.uk
Resolving www.tex.ac.uk (www.tex.ac.uk)... 78.46.26.59
Connecting to www.tex.ac.uk (www.tex.ac.uk)|78.46.26.59|:80... connected.
HTTP request sent, awaiting response... 301 Moved Permanently
Location: texfaq.github.io [following]
--2018-06-02 20:46:21--  texfaq.github.io
Resolving texfaq.github.io (texfaq.github.io)... 185.199.111.153, 185.199.110.153, 185.199.109.153, ...
Connecting to texfaq.github.io (texfaq.github.io)|185.199.111.153|:443... connected.

hacked github certificate, try now

7:47 PM
@StefanKottwitz wget happy ...
@StefanKottwitz ... and firefox and chrome work too, thanks!

@UlrikeFischer Excellent!

@DavidCarlisle my pleasure

@StefanKottwitz @JosephWright is plan now to tell github pages to serve it as texfaq.org ?

@StefanKottwitz Cool: I suspect we need to do something at the GitHub end to make this 'transparent'
@DavidCarlisle I guess so

@DavidCarlisle This would be the next step, yes transparent

7:52 PM
@StefanKottwitz I've added texfaq.org on the GitHub admin page (and probably I should add you as an admin for the 'team')

@JosephWright gh seems to be moaning about that setting though

@DavidCarlisle I've backed out the setting: I suspect the redirect confuses it ...
@StefanKottwitz Thanks :) I think a blog post for me today or tomorrow ...

@JosephWright sure! I'll send to another comment by mail

@StefanKottwitz Great

@JosephWright I'm wondering why that final id and tag footer appears on the upstream version but not my fork

8:16 PM
@JosephWright @DavidCarlisle mail sent

@PauloCereda Yes, of course, I have plenty of time (but no stable internet) and most likely no clue ;-)

@marmot Your knowledge of asymptote is asymptotic?

@AlanMunn Yes, it goes asymptotically to zero (as it cannot become negative;-)

8:41 PM
@marmot Hello! I saw the Rosengarten today, but met no relative of yours.
@marmot And no invisible dwarf either, but this is almost obvious. ;-)

@DavidCarlisle I was demoralized and I canceled my question. :-) anyway thanks for everything always.
Good evening to all users into this chat.

8:56 PM
@Sebastiano Hi Sebastiano!

@egreg Let me tell you: no one has ever seen an invisible dwarf. My relatives may be on strike, this is very typical for the Italians ;-)

@marmot Moreover today was a national holiday.
@marmot I just hope that invisible dwarves are good at dodging motorbikes.

@egreg Some of them are dangerous: they catch and skin tourists. Better be careful!

@marmot The road to the Karerpass (passo Costalunga) is between Rosengarten and Latemar, two beautiful mountains. Just past the pass there's the famous Karersee.
3
The mountain is Latemar

@egreg Really pretty, indeed. I really love that region, but this year I most likely won't make it there. (But two weeks Maroon Lake is also not too bad.)

9:07 PM
@marmot Seems a nice place!
@marmot On the other hand, Karersee is full of tourists.

9:24 PM
@egreg Well, in order to be closer to my relatives and further away from the tourists, you need to hike a bit. (Marmots are not really attracted to the sound of motorcycles.... ;-) And when you hike a bit, you will have no internet, and don't get distracted by crazy chats ;-)

9:40 PM
@samcarter @JosephWright is this better (underline rather than bold on hover) davidcarlisle.github.io/texfaq.github.io/FAQ-startup

@egreg oh a fractal mountain!

@yo' You should be able to compute its Hausdorff dimension.

9:56 PM
@egreg your image has only a finite resolution though

@yo' Interpolate

@egreg well, interpolation here would lead HD 1 (or 2 is speaking about 3d)

@yo' The fact that the image has only some finite resolution does not tell you much about the nature of the mountain. (Yet it is very likely that it is made of atoms, and hence not really fractal.)

@DavidCarlisle Looks good: merge

@marmot well, even several coast lines had their HD determined

10:01 PM
@yo' Not really. First of all, these coasts are made of atoms. Second, if you go to very small distances, our usual notion of geometry will break down (and get replaced by something like string theory). So don't trust these computations.

@marmot Of course, strictly speaking, you cannot measure it. However, you can still tweak the definition and try to look at the "experimental" HD of various coast lines and compare them:
Benoit Mandelbrot has stated that "A fractal is by definition a set for which the Hausdorff-Besicovitch dimension strictly exceeds the topological dimension." Presented here is a list of fractals ordered by increasing Hausdorff dimension, with the purpose of visualizing what it means for a fractal to have a low or a high dimension. == Deterministic fractals == == Random and natural fractals == == See also == Fractal dimension Hausdorff dimension Scale invariance == Notes and references == == Further reading == Benoît Mandelbrot, The Fractal Geometry of Nature, W. H. Freeman & Co; ...

@marmot That's a LaTeX generated fractal mountain:
2