Want to take part in these discussions? Sign in if you have an account, or apply for one below
Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.
As has long been requested, I have now added the ability to use LaTeX style theorem environments rather than Instiki’s syntax. See (the source of) comma category and hypothèse inspiratrice for two pages where I have changed things to use the new syntax. Also the current version of the Sandbox.
I do not plan to automate a change from the old syntax to the new, or at least not for the moment. But if people prefer the new syntax, it might be good to gradually phase the old one out. Here is a list of the available environments (the key is on the left, i.e. what you type, e.g. \begin{defn}, and on the right are what will be displayed, and the name of the HTML div class that will be used to render it (you can probably ignore this one; they are the old Maruku classes, I have not changed these)).
"defn": ("Definition", "num_defn"),
"definition": ("Definition", "num_defn"),
"thm": ("Theorem", "num_theorem"),
"theorem": ("Theorem", "num_theorem"),
"prop": ("Proposition", "num_prop"),
"prpn": ("Proposition", "num_prop"),
"proposition": ("Proposition", "num_prop"),
"rmk": ("Remark", "num_remark"),
"remark": ("Remark", "num_remark"),
"cor": ("Corollary", "num_cor"),
"corollary": ("Corollary", "num_cor"),
"lem": ("Lemma", "num_lemma"),
"lemma": ("Lemma", "num_lemma"),
"notn": ("Notation", "num_defn"),
"notation": ("Notation", "num_defn"),
"terminology": ("Terminology", "num_defn"),
"scholium": ("Scholium", "num_cor"),
"proof": ("Proof", "proof")
With the new renderer, it is quite straightforward to add this kind of functionality. Let me know if you’d like any further theorem enviroments, or any further ’blocks’ in general. I plan to add support for table of contents and sections at the moment.
Forgot to say that one uses \label{} within the theorem environment block to label it, like in LaTeX. There is an example currently in the Sandbox.
Wonderful! I hope Dmitri Pavlov sees this.
Incidentally, I was just about to say that you should point to Sandbox rev 1647 to provide a stable link, but now I find that following this direct link produces a page where the environments do not render.
That’s awesome! When this is judged stable (now?) we should modify the FAQ/HowTo/etc. pages.
That’s really nice, thank you! Would it be possible also to fix the syntax for bibliographic references so that one can simply say \cite{…} and \bibitem{…}?
Thanks for the feedback!
Incidentally, I was just about to say that you should point to Sandbox rev 1647 to provide a stable link, but now I find that following this direct link produces a page where the environments do not render.
Yes, for the moment I have not switched on the new renderer for revisions, only for the current content. Before this was not much of a problem, because the old renderer and the new one produced more or less the same content, but now it becomes more urgent. I will do it when I get the chance (also the new renderer is only activated for the main nLab at the moment, not for the other webs).
When this is judged stable (now?) we should modify the FAQ/HowTo/etc. pages.
Definitely. I am hopeful that it is reasonably stable already, but a few gremlins will no doubt be discovered once people start using the new syntax in earnest. Let’s wait a short while, I suggest; if things seem OK, we can update the things you mention.
Would it be possible also to fix the syntax for bibliographic references so that one can simply say \cite{…} and \bibitem{…}?
Absolutely. One of the main things that I would like to do (it is on the Technical TODO list (nlabmeta)) is to make some improvements to the referencing, which go slightly further than just the syntax. But at least \cite{} should easily be possible to add already, I will do that when I get the chance.
By the way, I believe that Mike and possibly Dmitri as well mentioned that use of macros would be helpful to them with regard to writing tex in the nLab. It occurred to me yesterday that it would not be too difficult to add support for this, at least in some simple cases. Can you point me to some examples of the kind of macros you have in mind?
There are two or three different ways that macros could be implemented, which would correspond to using them in two or three different ways. On one hand, there could be “nLab-wide macros” as a way of implementing consistent choices of notation across the entire site. On another hand, there could be a way to define “local macros” on a given page, just to simplify the lives of people writing/editing that page by reducing the amount of typing they have to do (and making it easier to change notation on the entire page at once), like \newcommand{\C}{\mathcal{C}} if a category named is being used a lot on that page. Somewhere in the middle, if local macros pass through includes, we could have includable pages of “packages” of local macros, for implementing consistent choices of notation in a collection of related pages and making it easier to change notation on all of them at once.
Fixed a bug coming from the interaction between Maruku and the new renderer for the new tex style theorem environments, which cropped up when one concluded a definition (say) with a list and without any blank line before \end{defn}.
Re #6: It would be nice if the macros \tableofcontents, \title, \chapter, \section, \subsection, and \subsubsection also worked as expected.
Hehe, if you take a look at towards a diagrammatic proof of the Poincaré conjecture for knots, you will see that I am already using (some of) these, because I am intending to make them work this evening :-)!
I have now implemented support for \tableofcontents and for \section, \subsection, and \subsubsection. No support for \chapter yet, and no support yet for adding a \label to a section/subsection/subsubsection; at the moment, the label is automatically set to be ’section_’ concatenated with the section name with an underscore instead of space as separator. See (the source of) towards a diagrammatic proof of the Poincaré conjecture for knots for an example.
Regarding \title yet, how would you envision this working, Dmitri? I wouldn’t recommend that we use it to change the title of an nLab page; or at least, if we’d like to do this, it will currently require quite a lot of work. If that is what you had in mind, I would suggest that a compromise, for the moment at least, would be to support it when ’importing’ tex into a new nLab page, and to add it when exporting to tex.
I have moved the page I linked to #11 to a personal web, where the new renderer is not activated yet. I will activate the new renderer for personal webs later. Some examples of the new syntax are available at Sandbox just now.
By all means, you should of course feel free to use a personal web.
Maybe that would be occasion to clean up our list of personal webs. Some of them have never been used after creation, various others have only trivial edits before having been abandoned years ago. Since the list of personal webs is publically visible (even if the link remains not widely known) I would prefer if we deleted these ghost webs.
Thanks very much!
Here is the output from the database regarding the existing webs and when they were created and last updated.
+---------------------+---------------------+------------------------------+----------------------------+-----------+
| created_at | updated_at | name | address | published |
+---------------------+---------------------+------------------------------+----------------------------+-----------+
| 2008-11-28 02:17:38 | 2017-09-03 14:27:03 | nLab | nlab | 0 |
| 2008-12-02 13:53:35 | 2017-09-13 06:02:35 | Schreiber | schreiber | 0 |
| 2008-12-30 14:51:47 | 2013-05-10 14:13:58 | Eric Forgy | ericforgy | 1 |
| 2009-01-18 15:43:33 | 2016-07-30 23:21:54 | Michael Shulman | michaelshulman | 0 |
| 2009-01-28 22:24:43 | 2010-07-20 00:47:48 | Todd Trimble | toddtrimble | 1 |
| 2009-02-10 07:31:30 | 2010-01-17 19:57:36 | David Roberts | davidroberts | 0 |
| 2009-03-22 21:13:06 | 2010-01-19 17:57:03 | Toby Bartels | tobybartels | 0 |
| 2009-03-23 11:54:59 | 2009-06-06 00:41:20 | James Dolan | jamesdolan | 1 |
| 2009-04-02 18:22:03 | 2009-12-30 13:03:26 | Tim Porter | timporter | 0 |
| 2009-04-17 15:00:39 | 2010-02-02 22:15:14 | Zoran Skoda | zoranskoda | 0 |
| 2009-04-27 15:17:53 | 2017-03-13 05:56:19 | David Corfield | davidcorfield | 0 |
| 2009-05-13 16:37:36 | 2009-05-14 00:38:11 | Alex Hoffnung | alexhoffnung | 1 |
| 2009-06-07 16:22:39 | 2013-06-21 00:29:48 | John Baez | johnbaez | 0 |
| 2009-06-17 10:10:48 | 2013-05-10 14:04:56 | Bruce Bartlett | brucebartlett | 1 |
| 2009-07-09 17:10:21 | 2010-06-18 02:14:48 | nLab meta | nlabmeta | 0 |
| 2009-07-23 12:11:49 | 2009-08-05 21:50:12 | Schreiber private | schreiberprivate | 0 |
| 2009-09-04 13:29:54 | 2011-08-22 12:56:19 | Doriath | doriath | 0 |
| 2011-10-04 16:27:35 | 2011-10-17 13:01:06 | Topology Group Moerdijk | topologygroupmoerdijk | 1 |
| 2009-10-24 07:08:02 | 2010-08-10 04:08:18 | BWebster | bwebster | 0 |
| 2009-11-10 23:54:26 | 2009-11-11 00:02:26 | Chris Rogers | chrisrogers | 0 |
| 2009-12-21 14:07:49 | 2013-03-20 16:48:46 | Domenico Fiorenza | domenicofiorenza | 1 |
| 2009-10-30 16:01:50 | 2013-05-10 14:01:10 | DStevenson | dstevenson | 1 |
| 2010-01-18 21:35:06 | 2013-05-10 14:10:35 | Joyal's CatLab | joyalscatlab | 1 |
| 2010-02-26 12:40:09 | 2010-02-26 12:43:05 | Eregion | eregion | 0 |
| 2010-06-15 20:09:26 | 2010-06-15 20:13:43 | fpqc's Effective Descent | fpqc | 0 |
| 2010-07-21 09:15:40 | 2010-07-23 12:07:25 | lSpace | lspace | 0 |
| 2010-07-23 12:08:53 | 2010-07-23 12:12:34 | Igor Bakovic | ibakovic | 1 |
| 2010-07-28 12:50:20 | 2013-05-10 14:15:57 | Kevin Lin | kevinlin | 1 |
| 2010-07-28 12:53:35 | 2010-07-28 12:55:23 | Kevin Lin Private | kevinlinprivate | 0 |
| 2011-01-02 02:54:14 | 2011-11-02 23:36:12 | Publications | publications | 1 |
| 2011-01-20 13:25:02 | 2011-01-21 02:48:09 | Finn Lawler | finnlawler | 0 |
| 2011-01-29 13:10:52 | 2011-01-29 13:25:13 | nLab (reviewed) | nlabreviewed | 1 |
| 2011-06-24 07:30:00 | 2011-06-24 07:34:33 | Zoran Private | zoranprivate | 0 |
| 2011-04-07 05:44:18 | 2013-05-10 14:17:31 | Roytenberg | roytenberg | 1 |
| 2012-02-08 23:14:20 | 2012-02-08 23:16:53 | Todd Trimble 2 | toddtrimble2 | 0 |
| 2012-05-29 17:26:13 | 2012-05-30 13:00:59 | Spahn | spahn | 0 |
| 2013-08-24 06:59:50 | 2013-08-24 07:02:05 | Problems in homotopy theory | problemsinhomotopytheory | 0 |
| 2014-02-17 04:18:43 | 2015-09-24 11:52:11 | Homotopy Type Theory | homotopytypetheory | 0 |
| 2014-06-09 03:38:50 | 2014-06-09 21:38:25 | Holmstrom | holmstrom | 1 |
| 2014-06-09 03:42:35 | 2014-06-09 21:39:42 | Hasse-Weil | hasseweil | 0 |
| 2014-08-10 02:10:04 | 2014-08-11 01:47:40 | Higher Structure Conferences | higherstructureconferences | 0 |
| 2018-08-27 05:21:06 | 2018-08-27 05:21:06 | Richard Williamson | richardwilliamson | 0 |
| 2015-03-30 13:20:44 | 2015-04-01 15:41:01 | HoTT MURI | hottmuri | 0 |
| 2016-07-30 23:15:21 | 2016-07-30 23:16:17 | Noam Zeilberger | noamzeilberger | 0 |
| 2018-03-28 17:13:32 | 2018-04-26 17:39:28 | UF IAS 2012 Archive | ufias2012 | 1 |
+---------------------+---------------------+------------------------------+----------------------------+-----------+
Just advise me on how to proceed. There’s no going back once they are deleted, of course, so we would probably need to contact the relevant people first. Alternatively, I could add something which allows us to mark a web as inactive, so that it does not show up; in this way we would not lose the content, if the person ever asked for it.
Hmm, updated_at seems to refer to config changes, so disregard that. I can look up last page edit for those webs which we need info about this, just let me know.
Maybe those marked ’published’ can be deleted, except the UF IAS 2012 Archive, the CatLab, and maybe one or two others?
Re #11: \title should set the content of the HTML element <title> as well as the title displayed in green color on top of the page.
Re #17: Thanks for the explanation. I’ll keep it at the back of my mind for future refactorings, and am not opposed in principle, but for now I think that I’ll not implement this one, unless others ask for it as well. I think page names are handled OK at the moment; it is after all just one line in the TeX file, and more metadata than content. But the main reason, as in #11, is that the code around the handling of page names would need quite a bit of rewriting, and I don’t feel that the importance of \title would be significant enough for that (but, as I say, I’m open to changing my mind).
Thanks for doing this!
It looks like currently \tableofcontents generates a # Contents header above the TOC. I think it would be better not to do this, and instead have \title generate a #-level header. Then the people who like the #-level header to be Contents could write \title{Contents}, while those of us who prefer the #-level header to be a page name could pass something informative to \title.
re #16, why would the ’published’ ones be more deletable than the others?
Re #19: good idea! Funnily enough, I was thinking about the same a short while ago. I’ll implement it when I get the chance. Is it good to use \title for this purpose, as it might prevent using it as Dmitri suggests in the future (one might wish to have a custom table of contents header and a title)? Maybe I could add syntax like: \tableofcontents{header: My header} or something? Or is there some good exsiting LaTeX syntax for this that people would like to use?
Re #20: it looked to me like many of those which have been made published have been made so to prevent them being edited, e.g. by spammers, which suggests that they are not active. But I might be wrong about that. Am happy to delete or otherwise handle whichever people request.
Richard, could you remind me where the new LaTex-style syntax is documented, if anywhere?
Ideally it would eventually have something on this at HowTo.
I don’t like the idea of using anything in the text of the page to affect the wiki-level pagename/URL; it feels to me like there should be a clean separation between the two. I think \title is the natural LaTeX syntax to use for a top-level header # of which, by general nLab convention, there is only one.
Re #22: It is not properly documented yet. In #6, I suggested waiting a bit before following Mike’s suggestion in #4 to add to the HowTo, etc. I’ve tested it a fair bit, now, though, and I think we can go ahead and advertise it now. I am a little tired/busy just now and may not get to it for a few days or next week, but if anyone else feels able to add to HowTo, they are very welcome, and I can add it to the right of the editing page when I get the chance.
Basically one just uses ordinary LaTeX syntax \begin{thm} \end{thm}, and one can put a \label{} anywhere inside the block to make a reference. The available environments are listed in #1.
One can use \tableofcontents rather than the usual table of contents syntax for creating a table of contents.
And one can use \section, \subsection, and \subsubsection instead of the ##, ###, and #### syntax for creating sections. There is no possibility for adding a custom label name to a section just yet, but one is automatically generated (see #11), so they can be referenced as usual. The sections are the least tested of these. Just let me know if more info is needed.
Re #23: Having thought about it some more, I agree. An additional benefit is that, whilst it may not fully achieve what Dmitri would like, it is a step in that direction, because one does not have to modify one’s TeX file to remove \title if one wishes to just copy and paste. Will implement when I get the chance.
one does not have to modify one’s TeX file to remove \title if one wishes to just copy and paste
Well, I think one does still have to make some modifications, because (at least with most LaTeX document classes) the \title command goes in the preamble, so you would have to remove the \begin{document} and \maketitle. But I agree with the overall point.
True!
1 to 26 of 26