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.
1 to 83 of 83
As the “corresponding editor” I am glad to be able to announce that a new article has just been accepted for publications in the Publications of the nLab:
Domenico Fiorenza and Elena Martinengo,
A short note on -groupoids and the period map for projective manifolds,
Publications of the nLab vol. 2 no. 1 (2012)
A pdf of the final version of the article is availabe behind the above link. We now need volunteers, as for our first publication, to instikify this submission… therefore just for the moment also the tex-source is linked to there.
The obvious first step is for me to run it through my magic machine that produced Tom’s article. Let me have a go at that, first, to determine what would be most useful for others to concentrate on.
Good, thanks!
Okay, I’ll put the temporary stages up at FiorenzaMartinengo2012 (doriath). You can see the results of the first run there. To get that to compile I had to remove all the xy stuff - that will need replacing with either ordinary matrix-type diagrams or proper SVG diagrams. I’ve also had to remove the TeX commands \bf, \rm and so forth (these are TeX commands, not LaTeX ones, and are a bit dubious in math-mode anyway!).
Obviously, I’ll need one or both of the authors to keep an eye on how things look.
Thanks, Andrew!
Obviously, I’ll need one or both of the authors to keep an eye on how things look.
Yes. I had informed them already and pointed them to this thread here. I guess as soon as they find the time they will come by here.
Glad to see this proceeding.
Hooray!
Domenico Fiorenza asks me by private email to forward here the information, with his apologies, that he’ll be mostly offline and unable to show up here until coming Tuesday.
Here I am, at last! :)
Andrew, you did a great job (as usual)! I see only a few little things need to be fixed before processing the article to get the relevant links to nLab pages. I can manage these few corrections in a couple of days. As I fix anything I’ll post here what I’ve done, so that you can see whether that is a bug whcih is worth fixing in general for future articles or something which is so specific to this submission that it is not worth automatically implementing it in general.
Will now be off line till this evening,
best,
d.
I thought that there was quite a lot to be fixed! In particular, I commented out all the xymatrix stuff so that will need to be redone. I also had to undo your use of \bf, \it, and \rm (which aren’t good LaTeX anyway) so those should be reimplemented as \textbf and so forth (and done differently in maths mode).
We will need a “For Authors”-page on the Publications-site with instructions and/or hints for how to prepare submitted LaTeX sources.
I made a start here: For Authors (publications). But it’s just a first idea. Please feel free to expand on that. You know of course more about that script than I do.
I thought that there was quite a lot to be fixed! In particular, I commented out all the xymatrix stuff so that will need to be redone. I also had to undo your use of \bf, \it, and \rm (which aren’t good LaTeX anyway) so those should be reimplemented as \textbf and so forth (and done differently in maths mode).
Right, the xymatrix stuff has to be redone (it will be an occasion for me to learn a bit of svg :) ), and there is the \bf, \it, and \rm issue. but, as you wrote that was bad latex so it is nonsense to adapt the latex-to-itex converter to take care of an author’s bad latex habits. I will fix those by hand. something which would instead be nice to have automated is the bibliography (maybe there is some bad latex dialect of mine here, too)
it also seems the automated converted has expunged me from the authors.. :)
Right, the xymatrix stuff has to be redone (it will be an occasion for me to learn a bit of svg :) ),
While that is probably the preferred option, one could also use some other way to produce the diagrams. The quickest is probably to use codecogs, as described at HowTo here.
I am mainly saying this because I am getting worried that the task of conversion becomes a bit overwhelming. I seem to remember that when the first Publications article appeared, there were several volunteers helping with the conversion. This time it seems to be different.
(This is something that needs to be thought about generally: suppose the Publications are being announced more broadly, once the second article is fully published – as was the original plan. Then: how to ensure that there are enough volunteer man-hours to make it “run” – or at least to make it walk? )
Then: how to ensure that there are enough volunteer man-hours to make it “run” – or at least to make it walk?
I think there will be. Consider that by now Publications have been extremely fast in converting the articles from the sources to the Publications format (a few days compared to the months a journal usually takes only to change the latex class at the beginning of the latex file - when the journal does not directly ask the author to submit an article compiled with thisjournal.cls). And in the particular case of the second article I guess much of the enthusiasm has been slowed down by my being away while the publication was announced. I will now be editing FiorenzaMartinengo2012 (doriath) and reporting about that.
I’ve now edited till section 4. Only two things worth mentioning to Andrew’s attention so far:
there’s a problem with ditto marks: in latex I use (see source), but the first causes a problem. I’ve now replaced that with “ljlkjkjl”
title of section 4 was missing, can’t figure why
The “script” is not really a script, it’s a LaTeX document class. It therefore works fairly well with ordinary LaTeX but it has to know what the right output is for a given input, so things like {\bf something} don’t work because I don’t know how to correctly map that to **something**. Similarly, the amount one can do with xymatrix is so complicated, I haven’t tried to work out how to convert it to SVG.
With my initial import, I commented out everything that made it fail so that I could get a barebones to start with. That’s why some things are missing - such as the title of section 4 as it contained mathematics and that’s tricky to handle in “moving arguments” (search for “fragile latex” … on second thoughts, don’t!).
Strangely, ditto marks are one of my biggest challenges. TeX handles them as ligatures. But in general, I don’t want ligatures in this script since ffi should stay as ffi and not ffi so I disable ligatures. What I really need to do is design (adapt) a font and take out all ligatures except the quote marks (and dashes).
I’m now editing the two-arrows diagrams which were not automatically implemented. The solution I came to is the following which is almost ok,
but the two arrows are a bit drifted up. any idea on how to make make this centered? otherwise I’m ok with the solution above
also footnotes are missing: any suggestion on how to include them?
Perhaps as little blocks of text that appear when moused over? Andrew has done a bit of this for his own lecture notes, and this is used in the menu bars throughout the nLab.
I’ve now added a couple of svg diagrams. two questions:
is there a way to have something like fraktur fonts in sgv?
is there a way to center the diagrams on the nlab page (now they are aligned on the left)?
I remember removing one footnote because it was causing the tex compilation to hang (maths again, I really must fix that). There is a standard Markdown syntax for footnotes that I keep forgetting.
Yes, SVGs can be centred. And you should be able to use iTeX in SVGs. The easiest way is to create them with the svg editor that is built in to Instiki.
also footnotes are missing: any suggestion on how to include them?
Instiki supports footnotes: you put a
[^footnotelable]
in the text and then
[^footnotelable]: Footnote text goes her.
See currently at Sandbox for an example.
Sadly, the inbuilt SVG editor is not working for me today.
I’ve centred the SVGs. The trick is to surround them with
+-- {: style="text-align:center"}
...
=--
neither for me, so I used the SVG editor provide by googlecode. but I wasn’t able to call latex from there. anyway, now all the needed svg diarams are in place and one only needs to replace a few letters with the corresponding gothic.
The only big issue still to be fixed now is the bibliography, which is missing. But that should be easy: can you point me to an nLab page with a bibliography at the end?
With this done the note should be ok, but let me go through it once more before processing it to provide links to nlab pages
Most Lab entries have lists of bibliographies, certainly all the big ones like category theory etc.
But maybe best is if you look at the other Publications article Leinster2011 (publications) to see how precisely it was done there.
As, sorry, maybe you can’t see the source of Leinster2011 (publications) due to the password protection? I’ll send you the code by email.
The cheap way to replace the letters in the SVG is simply to use the right characters. You can cut-and-paste the gothic characters into the SVG. Looks like you’ll need , , and . But they don’t look quite right as they aren’t set in MathML. To do that, you need an SVG editor that can cope with foreignObject. I had a go at doing it manually at SVGs for FiorenzaMartinengo2012 (doriath) but as you can see, the (the one I tried) is misplaced.
I’ll see if Jacques knows anything about why the SVG editor isn’t working.
You can cut-and-paste the gothic characters into the SVG
Great! thanks!
I’m now doing that. Ive also included the bibliography, but I must have done something wrong since the links to the bibliography seem not to work :(
The bibliography thing is strange. Looking at the source of the generated page (ie the XHTML), I see that the keys have been rendered as, for example citegetzler}. That final } is clearly not meant to be there. What I would try is to see if adding a space between the identifier and the closing brace helped:
* {: #citepridham2 }
J. P. Pridham. _The homotopy theory of strong homotopy algebras and bialgebras_; [arXiv:0908.0116](http://arxiv.org/abs/0908.0116)
Looking at the Sandbox (since you’re editing the original) this seems to work.
(This is clearly a bug, I’ll report it to Jacques.)
Thanks, it worked perfectly! Now I see a couple of equation I still have to add to section 7 (will do that later), and then all big issues should be over :)
We also still need to hyperlink all technical terms.
I may forget how that worked. I guess Andrew has another piece of script or similar and then we will need to clean up after it a little? Shouldn’t be a big deal, but let’s not forget about this.
Both the referee and the authors have agreed that the refereeing process (submissions, reports, reactions) be made public, together with the published article.
I have now collected all the documentation of the refereeing process at
(and linked to it from 2012 (publications)).
The formatting of that material on that page is still very rough. I need to call it quits now for the moment, so I just made sure that everything is visible at all. The page needs further editing for nice readability. Also, the sources of the original submission (superceded by the final version) are currently only being linked to, they still sit on Domenico’s server. Probably we want to upload them to the nLab server, eventually.
We also still need to hyperlink all technical terms.
Sure. I’m considering that “phase 2”, where “phase 1” is having the whole text converted in the nLab format. I should have just now accomplished that, but I’d like to go through the whole text once more tomorrow before processing it to add the hyperlinks.
I have now collected all the documentation of the refereeing process at …
The exact submission date is July 30th 2011
Okay. I have added that date and dates for all the other items in the refereeing process:
check out FiorenzaMartinengo2012 - refereeing (publications)
Should we add some kind of explanation of the length of the time interval between August 2011 and June 2012?
Might it make sense to add a remark like
Referee notifies steering committee that report will be overly delayed due to act of nature beyond control. Steering committee decides to continue the process with this referee nevertheless.
?
(Is that sensible English, even? “act of nature beyond control”? In German I would say “höhere Gewalt”.
Should we add some kind of explanation of the length of the time interval between August 2011 and June 2012?
since this has been an interaction between the referee and the Steering committee, it could be a good idea to add this, too, to the other data. “due to act of nature beyond control” does not sound well to me. what about “due to circumstances beyond control”? but best thing here is to wait for advice from native English speakers.
Concerning the article, I’ve now gone through the text at FiorenzaMartinengo2012 (doriath) and it seems ok: let’s go for hyperlinking!
Looks very good. How did you end up producing the diagrams?
As a native English speaker, I endorse ‘due to circumstances beyond control’, except that there ought to be a possessive adjective (‘our’, ‘the referee’s’, etc) before ‘control’.
But Urs’s phrase ‘höhere Gewalt’ put me in mind of ‘act of God’ (a legal phrase for extreme weather, earthquakes, etc).
Looks very good. How did you end up producing the diagrams?
svg :)
Thanks, Toby.
The term “act of God” seems appropriate (though it wasn’t an earthquake). Now I have put the following into the page:
Referee eventually notifies steering committee that report will be overly delayed due to act of God beyond the referee’s control. Steering committee decides to continue the process with this referee nevertheless.
But I’d be happy if you, or somebody, makes another concrete suggestion.
Needs to be ’due to an act…’
I guess an act of God is by itself beyond one’s control. I’d probably prefer the less dramatic “due to circumstances beyond the referee’s control”
Okay, I made it “due to circumstances”.
Hi Urs,
another thing: to make the refereeing process completely documented: could you add that at a certain point the authors spotted an inaccuracy and sent a revised version before the first referee’s report? the inaccuracy was minimal, so this is not strictly necessary, but I think that providing a full detailed account of all the process could be a good thing to have. In case, I’ve sent you via private email the source and pdf files for the original submission and of the revised submission with that inaccuracy corrected.
Also if you can recover the date of the referee’s communication of the necessity of delaying the refereeing process, adding that to the timeline could be fine (my idea is taht the more detailed account one can provide for nPublications refereeing processes, the better)
Sure, I’ll do that.
Okay, I have added more information at FiorenzaMartinengo2012 - refereeing
Toby: thanks, yes.
Everybody: so the next is adding all the hyperlinks. I forget how precisely that proceeded with the previous article. I suppose Andrew does something?
We need a HowTo somewhere for How To do all these things. I forget.
And as I recall, yes, Andrew had a script to add hyperlinks, which then had to be gone over by humans.
Yes, eventually. But for the moment we should develop it on a non-write protected web.
problems with the hyperlinking?
It seems we are still at #47.
Given the state of affairs, I guess we are waiting for Andrew to run the script. If Andrew is somehow too busy, we will need to add some hyperlinks by hand…
Been at the 2012 BTM, back now.
2012 BTM ??? Ah! BTM. Anything interesting mentioned? (sorry this is off thread).
So, doing the links is the really, really boring bit! I have a script that goes through the page and looks for words that match nlab page titles. For each one, you have a choice: (a)ccept, (s)kip, (i)gnore name, (r)edo line, a(l)ways accept, or (q)uit. Hopefully these are all self-explanatory. It will always try to find the longest match, but matches have to be exact so a space in a different place will throw it out. There are some obvious things you should ignore: and, the, and similar. There are also some less obvious ones: section, for example (you’ll see why on the first match). If you quit early, it will have saved everything up to that line so you can resume later at that line by passing a line number to the script to start from (but it will overwrite the output file so you need to save the new bit into a new file and then concatenate them afterwards. Obviously, there are improvements that could be done to this script.)
Call as:
wikilinks.pl -i <input file> -o <output file> [-n <page name file>] [-s <start line>]
The <page name file> defaults to nlab_pagenames in the current directory. Make sure that <output file> does not yet exist as it will be overwritten.
It’s a perl script, so should run on any modern system (though if you have an old version of perl it might not. My system uses 5.12.3.)
You can download it: script and almost current list of nlab page names (includes redirects).
Thanks, Andrew!
I have pasted what you just wrote into the instruction page For Authors (publications). (Would deserve further editing, but I can’t do more right now.)
Ideally, eventually we provide enough information and links on that page that the wiki-fication of submitted articles can proceed independently of whether you are available. For instance: can you also provide a link to the TeX-code that you use to transfer from LaTeX to instiki? Ideally this should eventually be available to the authors or to other helpful souls here.
Now Domenico has run Andrew’s script and I have now edited the result by hand a bit further.
Check out the article here
where it is editable, and here
where you can see how the current version would look once published.
I need to do something else now. But as far as I can see from scanning through the article the main remaining thing to be fixed is that in two of the SVG diagrams some labels got currupted, apparently from us chasing this through various editors.
Andrew, by the way: if you have a minute, could you change your script such that it produces
[[nlab:linkname]]
instead of
[[linkname]]
? Or is there maybe an option to choose such that this happens?
Good idea. There’s a few other things to clean up about that script but that one will be easy to do.
Thanks, Andrew!
Let me just mention two other things that I fixed by hand which I guess in principle the script could learn how to do. I am not requesting you to implement this, I am just mentioning it as something to keep in mind:
The first one is because Domenico used the syntax $\infty $-groupoid for ∞-groupoid and that isn’t (currently) a valid redirect for that page. (Actually, Domenico used $\infty$-groupoid without the space - my conversion class added the space.) One option would be to add that as a redirect (how does it work with iTeX in redirects? I’d need to experiment). Another would be to have the script be able to take in a list of aliases, whereby we could come up with a list of common substitutions like that.
The second is a good idea but I’d have to think about how easy it would be to implement. The script is quite crude and doesn’t know anything about Markdown. Making that work would involve teaching it a bit. Sections would presumably not be too hard to detect, though.
Thanks.
If we make
$\infty$-groupoid
a redirect, would it then not also appear this way on the screen, i.e. with the dollar signs?
You’ll know how to handle this. It’s not urgent.
All right, Domenico has fixed the remaining SVG diagrams.
So it looks like we are done with all the mandatory formatting!
There are still a few things one could improve with the links, if one were feeling ambitious. For instance:
“Frobenius manifold” was only linked as “Frobenius manifold” because we didn’t have Frobenius manifold. But clearly we should, and so I now created a stub for it.
“formality” wasn’t recognized as a keyword, while we do have Kontsevich formality. So I added a redirect to make it work.
“-algebra” is currently not recognized. What should we do about cases like this? We could have links “L-∞ algebra”. Does that look bad? Maybe at least the very first occurence should be linked like this.
One could do more along these lines. Depends on how ambitious we feel. Domenico should decide.
I think we still need to make sure that at least in the introduction the important keywords are really linked and linked correctly. Otherwise we are defeating one of the main purpose of the nPublications.
So I have created a stub for period map. We still need now Lab entries (can be just stubs containing just a single reference for the moment) for:
Also where it says “to every nilpotent dgla is naturally associated…” I have made “naturally associated” point to Lie inetegration. We could probably add more links like this. The ideal idea here is that no reader is ever left wondering about what the author is referring to, but always has a link behind which more information is found. We don’t need to push this idea all the way, but let’s briefly look at the article to see if there are some important such links which we should still provide.
I have now created stubs for some of these, but not all. Need to quit now. For some I am not sure where to point to. Domenico, please have a look at the (new) grey links at the beginning of FiorenzaMartinengo2012 (doriath)
Hi Urs,
great work, thanks! For quasi-abelianity we just need the definition: a differential graded Lie algebra is said to be quasi-abelian if it is homotopy equivalent to a cochain complex (i.e. to an “abelian” differential graded Lie algebra). Anyway, this is recalled within teh text of the article, so if we do not have an nLab page with this definition, this is still ok. For Cartan homotopy I’ll now write a stub. “projective manifold” can be linked to “projective variety” which is already in nLab.
Concerning the linking s of -algebra, I’ll later edit on doriah so that -algebra there will actually point to L-infinity-algebra.
Note that in passing from doriath to the nPublication page the gothic letters in the SVG has gone lost: probably an intermediate editor has been used at some point. By copying-and-pasting directly from doriath to the nPublication page one should be able to avoid this problem.
Will post here as I will have edited on doriath (this will happen in 8 hours or so)
Okay, here is projective manifold, check if that serves its purpose.
Even if stuff is explained later in the article, for general purpose could you just paste the corresponding sentence over to its dedicated entry. What should go in Goldman-Millson quasi-abelianity theorem ?
Note that in passing from doriath to the nPublication page the gothic letters in the SVG has gone lost
I wouldn’t think so: I just hadn’t copied over the latest version yet. But check, maybe I am missing something. Here is the latest Doriath version now moved over to the nPublications again:
But I’ll not keep that up-to date until we have a stable version on Doriath.
I am learning from this process that we should add more redirects for “category:people”-entries for fairly well-known people. That way Andrew’s script will better find them and link them. For instance now the following three all redirect to the same entry
(Clearly we can’t do this with every name. But as long as it works, I guess it would be helpful if we follow this pattern.)
I wouldn’t think so: I just hadn’t copied over the latest version yet. But check, maybe I am missing something. Here is the latest Doriath version now moved over to the nPublications again
ah, sorry! I thought it had already been moved.
But I’ll not keep that up-to date until we have a stable version on Doriath.
sure.
Okay, here is projective manifold, check if that serves its purpose
that’s ok
Even if stuff is explained later in the article, for general purpose could you just paste the corresponding sentence over to its dedicated entry. What should go in Goldman-Millson quasi-abelianity theorem ?
I’ll create an nLab entry for this later this evening
I’m now creating Goldman-Millson quasi-abelianity theorem and Cartan homotopy (or better I was, since the nLab seems to have just gone down :( )
Yeah, it’s slooooooow. Not down. I just got one page I asked for 8 minutes later.
But I am not trying any restarts at the moment: because with the steering committee members we are currently trying to figure out if everybody is experiencing this. The opinion has been voiced that on some people’s end the Lab is not actually slow.
Everybody who can access the Lab right now please drop me a message, together with an indication where on the planet your ISP is.
The lab was slow (UK), but how slow I do not know as I left it to load (after several minutes) and went to eat! That was a much better way to pass the time than waiting for it!
Yes, that’s why I always wait for a dozen pages at once. That’s time spent more efficiently. ;-)
I timed a call to the Home Page and it took 2 minutes, but earlier my connection was very slow. (Probably someone in the neighbourhood downloading a film!)
Later I tried to get Recently Revised and gave up after 4 minutes. (AH! It wa s you starting up the server !)
Thanks for the info, Tim. I have just restarted the server. I guess that’s what brought you the page. Anyway, right now it’s back. Thanks for all the info. (I’ll let you all know when we play this international game again. ;-)
I’ve now created the missing nLab pages Goldman-Millson quasi-abelianity theorem and Cartan homotopy. I also spotted and corrected a couple of typos in the doriah version and added the links to L-infinity-algebra and to gauge transformation where needed. I’m now done with that: if it can be uploaded on the nPublication page I’ll ask Elena to have a final look at it (I’d like to avoid shocking her by coloured hyperlinks :) ) and then I think it can be published
Great, thanks.
Here is the final version FiorenzaMartinengo2012 (publications)
Everbody who can please have a last look!
Domenico, while it is fresh in your mind, I’d appreciate any thoughts you have on the conversion process.
Domenico, while it is fresh in your mind, I’d appreciate any thoughts you have on the conversion process.
I found everything extremely smooth. The only part which I really think could be improved (but I don’t know whether this can be automatized somehow) is putting in evidence if there are parts from the original latex that have been removed in order for the latex-to-itex conversion to work. I’m saying this since when an equation is removed it is not always easy even for the author overlooking the result of the conversion to notice there is something missing there and one could just think to be in presence of some badly composed sentence to be fixed. If there could be a way of keeping track of the removed parts, that would be great.
Apart this, everything has worked like a charm :)
Everbody who can please have a last look!
I’m now asking Elena to have a last look
According to Elena everything is ok: from now on every moment is fine for us for the article to be published.
And a voice came from the heavens, saying
So be it published!
And so it was.
Seriously:
Thanks everyone who helped for helping. Special thanks to Andrew and his scripts! And to the referee and, not to forget, to the author, who had quite a bit of editorial workload here beyond writing a good scientific article.
What needs to be done now is what we have been waiting for for about two years now:
we need to constitute an nPublications steering committee,
we need to contact the editors-to-be, point them two our two test-case-publications and wait if what they see makes them want to become editors-that-are.
we need to post announcements to the world that a new kind of journal has seen the day of light, the Publication of the Lab.
finally, we need to pray that somebody out there finds this fascinating enough to volunteer as secretary.
1 to 83 of 83