Not signed in (Sign In)

Not signed in

Want to take part in these discussions? Sign in if you have an account, or apply for one below

  • Sign in using OpenID

Site Tag Cloud

2-category 2-category-theory abelian-categories adjoint algebra algebraic algebraic-geometry algebraic-topology analysis analytic-geometry arithmetic arithmetic-geometry book bundles calculus categorical categories category category-theory chern-weil-theory cohesion cohesive-homotopy-type-theory cohomology colimits combinatorics complex complex-geometry computable-mathematics computer-science constructive cosmology deformation-theory descent diagrams differential differential-cohomology differential-equations differential-geometry digraphs duality elliptic-cohomology enriched fibration finite foundation foundations functional-analysis functor gauge-theory gebra geometric-quantization geometry graph graphs gravity grothendieck group group-theory harmonic-analysis higher higher-algebra higher-category-theory higher-differential-geometry higher-geometry higher-lie-theory higher-topos-theory homological homological-algebra homotopy homotopy-theory homotopy-type-theory index-theory integration integration-theory k-theory lie-theory limits linear linear-algebra locale localization logic mathematics measure-theory modal modal-logic model model-category-theory monad monads monoidal monoidal-category-theory morphism motives motivic-cohomology nlab noncommutative noncommutative-geometry number-theory of operads operator operator-algebra order-theory pages pasting philosophy physics pro-object probability probability-theory quantization quantum quantum-field quantum-field-theory quantum-mechanics quantum-physics quantum-theory question representation representation-theory riemannian-geometry scheme schemes set set-theory sheaf sheaves simplicial space spin-geometry stable-homotopy-theory stack string string-theory superalgebra supergeometry svg symplectic-geometry synthetic-differential-geometry terminology theory topology topos topos-theory tqft type type-theory universal variational-calculus

Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.

Welcome to nForum
If you want to take part in these discussions either sign in now (if you have an account), apply for one now (if you don't).
    • CommentRowNumber1.
    • CommentAuthorEric
    • CommentTimeMay 29th 2009
    In my opinion, we should start thinking about how to migrate from Instiki to Mediawiki.

    It is becoming increasingly clear that Instiki is not scalable. There is some very good content going up there and Instiki does not have the developer support we need for a long term solution. Mediawiki has a large base that will ensure continued updates and improvements to the software.
    • CommentRowNumber2.
    • CommentAuthorTobyBartels
    • CommentTimeMay 30th 2009
    The main problem is the inferior math support from MediaWiki. There are alternatives (within MediaWiki, I mean), which were discussed on the Café at some point before the Lab was created. (If I can find the discussion, then I'll let you know.)
    • CommentRowNumber3.
    • CommentAuthorEric
    • CommentTimeMay 30th 2009
    I'm not suggesting a migration now. I'm simply pointing out that a migration is likely inevitable, so it might not be too early to start talking about it.

    What we really need is a way to incorporate itex2mml into Mediawiki as a plugin.

    Instiki is great, but I don't think it was designed for large-scale multi-decade projects involving 10s and maybe 100s of (sometimes simultaneous) contributors.
    • CommentRowNumber4.
    • CommentAuthorTobyBartels
    • CommentTimeMay 30th 2009
    I actually prefer a locked edit page to the old days on Wikipedia and the edit conflicts that this generated. (But MediaWiki's current sophisticated conflict resolution system does work pretty well.)

    How many decades and contributors can Instiki handle? Other than the slow server (which is not Instiki's fault), I haven't seen any performance problems. What have you seen, or what are you anticipating?

    • CommentRowNumber5.
    • CommentAuthorEric
    • CommentTimeMay 30th 2009
    I think a line on the instiki feature requests page sums things up:

    http://www.instiki.org/show/FeatureRequest

    "Simple

    - if you want mediawiki features, use mediawiki—i like instiki being simple"

    A very nice thing about instiki is its simplicity, but I think we actually do want many of the Mediawiki features, but with the super nice math capabilities of itex2mml.

    It might be worthwhile considering installing Mediawiki along side Instiki on the server so that we can start experimenting with it. I've never tried to create a Mediawiki plugin, but I think it would be a tremendous contribution to science on the web if itex2mml could be made to work with Mediawiki.

    I have experienced extremely poor performance when two or more people are modifying different pages simultaneously, but that could be an issue with the server. Plus, I think we all agree that having real redirects would be extremely nice.

    PS: It seems that someone has picked up the baton recently with blahtex (another mathml extension for Mediawiki)

    http://www.mediawiki.org/wiki/Extension:Blahtex

    Maybe we could combine efforts.
    • CommentRowNumber6.
    • CommentAuthorAndrew Stacey
    • CommentTimeMay 30th 2009

    My instant reaction to this is to suggest that we wait until the server issues have been solved before declaring Instiki inadequate. The host for the n-lab definitely has a few oddities (Urs mentioned elsewhere that once we were both logged in at the same time and got each other's output!) and some of the problems with the 'lab may be entirely due to the host.

    As for scalability, how big do you think that the 'lab is going to get? My guess would be that actually in terms of editors then the answer is "not many more". What is more likely is that other groups are going to emulate the 'lab when they see what fun we're all having, but they're not necessarily going to come along and join in.

    What specifically does Instiki lack that you feel we really need?

    • CommentRowNumber7.
    • CommentAuthorEric
    • CommentTimeMay 31st 2009
    What seems adequate today will likely be inadequate 10 years from now. Even less so 20 years from now. I don't see a dedication by developers to continue to improve Instiki or even to add basic features such as redirects.

    I am pretty confident that Wikipedia (and by extension Mediawiki) will be around 10 years from now. I am not so confident about Instiki. The content that is going onto the nLab should be fairly timeless. I can easily imagine it being a valuable reference 10 years, 20 years, or even longer from now.

    Most of the thought process so far has been about basic functionality and less about posterity. There is no urgency, but by bringing up the subject, I'm hoping to start thinking about posterity.

    Some of the unpleasant experiences lately could very well be simply related to the server, but they could also very well be due to flaws in Instiki itself.

    Regardless, I don't think it hurts to think about the long term. And when I think about the long term, I am doubtful that it will involve Instiki.

    As far as how big the nLab will get, I think it should get as big as the field of n-categories gets, which could be pretty big. At a minimum, it should be able to easily handle 10 people simultaneously editing possibly different (or even the same) pages. I also think it would be REALLY unfortunate if another group starts ANOTHER wiki to promote collaborative research on n-categories. In fact, the nCafe could and should ultimately encompass all of mathematics.

    Currently, we have serious researchers (me not included) putting serious material on the nLab. It could conceivably become a modern day version of Bourbaki, so we should not underestimate the potential that we're creating a lasting piece of science.

    Just to clarify something...

    I am not criticizing Instiki. I think Instiki is great for what it was designed to be, i.e. a lightweight, simple, transportable, collaborative research tool. I'm just beginning to think that the nLab could be outgrowing it. I'm not saying that it has already outgrown Instiki, but at the rate it is growing, the writing is on the wall.
    • CommentRowNumber8.
    • CommentAuthorEric
    • CommentTimeMay 31st 2009
    If anyone would like to carry out some experiments with me, I just managed to get Mediawiki installed on my local pc (Windows XP) and it miraculously works :)

    You just need to install XAMPP

    http://www.apachefriends.org/en/xampp-windows.html

    Then follow the instruction here:

    http://www.mediawiki.org/wiki/Manual:Installing_MediaWiki_on_Windows_XP

    1. Download and install Xampp from here.
    2. Use the Xampp control panel to make sure that both MySQL and Apache are running. Port 80 must be free.
    3. Download MediaWiki and unzip into the htdocs directory (eg, c:\xampp\htdocs\mediawiki)
    4. Connect to http://localhost/mediawiki/ and follow the instructions.

    I would imagine that analogous simple instructions exist for any operating system.

    My goal is to learn how to create Mediawiki extensions and then think about what needs to be done to get itex2mml (or something like it) to work as an extension.

    I would also like to write something to assist in porting things back and forth between Mediawiki and Instiki.

    Help appreciated!
    • CommentRowNumber9.
    • CommentAuthorEric
    • CommentTimeJun 1st 2009
    I'm going to assume that I am the only one interested in this project for the time being, so will not continue to update my progress here. However, I do want to have a diary of my experience, so I am putting it here:

    Porting nLab to Mediawiki
    • CommentRowNumber10.
    • CommentAuthorTobyBartels
    • CommentTimeJun 1st 2009
    I am interested in watching, so thanks for the diary.

    I think that you still have a case to make to the rest of us that using MediaWiki would be a good idea. A working model to point to will bolster that case.

    • CommentRowNumber11.
    • CommentAuthorAndrew Stacey
    • CommentTimeJun 1st 2009

    Hang on, Eric! You started this topic on a weekend so give us a little time to react. I gave my initial reaction and you've also gotten a comment or two from Toby. Both of us are hesitant, but I think that's understandable.

    I think that the n-lab has survived long enough to be more than one of these good ideas that never gets off the ground. There are plenty of these and it's always worth waiting a bit before investing time and effort in a project, particularly in the non-portable parts. The actual pages of the n-lab are extremely portable so none of that will be lost if we eventually decide to shift to some other system.

    I don't know much about how Instiki actually works, but the connection issues you talk about may well not be anything to do with Instiki itself. They feel much more to do with the host machine, the web server (and how it deals with simultaneous requests), and the communication between Instiki and the database. Now that the n-lab has proved itself not a fly-by-night thing, we can hunt around for a better server (hunts are in progress) to see what that resolves. As for the web server, I'm not sure what the n-lab is using at the moment, but on a better host we can fine tune it a little more carefully, and for the database communications then this will not be so much Instiki as Ruby on Rails which does, as far as I'm aware, have massive developer support.

    So no, you've not yet convinced me. And with the exception of Jacques (who, I get the impression, is not taking too much of an active role in the n-lab (again, completely understandably)) I'm probably one of the most technically-minded people involved with the project.

    Let me consider your points.

    What seems adequate today will likely be inadequate 10 years from now. Even less so 20 years from now.

    I completely agree. But this applies just as much to mediawiki and Wikipedia as it does to the n-lab. In fact, looking at the brief history of software, the stuff that has survived and worked well is the stuff that originally had an extremely small group of people who were really dedicated to making it work. The Linux kernel, Perl, TeX, these all spring to mind as projects that have worked and survived. So Instiki and the n-lab fit this very well.

    Wikis are very much the current fad - everything must have a wiki these days. I'm not deriding this - indeed, I quite like the Wiki idea. However, I think that most wikis will not survive the next Big Thing and those that do will be the ones where the Wiki has a group of dedicated workers who have decided that it is worth pursuing with.

    And even if mediawiki does survive (which I fully expect it to do), there's a big difference between the core engine of mediawiki and individual plugins. MathML is always going to be a specialist area and Wikipedia is unlikely to switch over to MathML any time soon simply because of the lack of widespread use. They need to be accessible to all, and that includes school kids' grandparents who are trying to do their grandkids' homework. The n-lab and n-cafe, on the other hand, are only trying to reach a specialist audience and asking that audience to install a few fonts is a reasonable ask for the usefulness of the resource supplied.

    The content that is going onto the nLab should be fairly timeless. I can easily imagine it being a valuable reference 10 years, 20 years, or even longer from now.

    I can't. Very few textbooks written 20 years ago are still relevant today. We all know of some, but what we don't know of are the vast array consigned to the rejects pile in the publishers' bins. Some projects have lasted well, Bourbaki being a classic example, but although we should not do anything to hinder our future existence, making decisions now solely on that is a little premature.

    However, the content of Instiki is already timeless. The input format is very simple: Markdown+iTeX, and it can be exported as LaTeX. And while I have doubts about any other system, I am as sure as I can be that LaTeX will still be around in 10 or 20 years' time. Or at least, that whatever system we have in 10 or 20 years' time will be fully backward compatible with LaTeX.

    Regardless, I don't think it hurts to think about the long term. And when I think about the long term, I am doubtful that it will involve Instiki.

    Agreed. We have been quite involved with the day-to-day functionality. Partly because we've got to solve these issues in order to have a chance of having a long term. If it's too hard to do stuff in the n-lab then it will die off. But it doesn't hurt to think about the future and I agree, it probably won't involve Instiki. But it probably won't involve MediaWiki either. It will probably involve some completely new system that no-one has yet thought of. Google Wave maybe? (I hope not!)

    • CommentRowNumber12.
    • CommentAuthorAndrew Stacey
    • CommentTimeJun 1st 2009

    (Had to split into two comments!)

    As far as how big the nLab will get, I think it should get as big as the field of n-categories gets, which could be pretty big. ... I also think it would be REALLY unfortunate if another group starts ANOTHER wiki to promote collaborative research on n-categories. In fact, the nCafe could and should ultimately encompass all of mathematics.

    I have no wish to upset mine hosts, or anyone else involved in this project, with my opinions of the importance of n-categorical thinking in mathematics. It may well prove to revolutionise mathematics, but then again it may well not. Category theory still hasn't the impact that it ought to. Set theory is recognised as being of central importance in status, but not in practice. So saying that the n-cafe could and should ultimately encompass all of mathematics seems a little hyperbolic to me. Even if it could, that's a little much to lay on the shoulders of mine esteemed hosts. As for another group starting another wiki, well, I just don't see it. Mathematicians may have the social graces of cats in a sack, but they do recognise that they aren't a huge group so if there's already a reasonable project out there then it's better to join that than start a new one.

    Currently, we have serious researchers (...) putting serious material on the nLab.

    This is the strength of the n-lab and this is what we need to work to preserve.

    It could conceivably become a modern day version of Bourbaki, so we should not underestimate the potential that we're creating a lasting piece of science.

    A-ha! So you want to pick a fight with the iBourbaki project!

    But seriously, posterity can look after itself. I've enough to think about right now that I have little enough left for that.

    However, having just poured the entire contents of Trondheimsfjord (a vast quantity of cold water) on the idea, I will keep an eye on your diary and if there's any way I can help then I will. I have a mediawiki installation of my own buried somewhere on my website (I'm allowed PHP+MySQL without difficulty but have to ask for anything else, hence no Instiki) and so can try out anything you produce. I also have a project that may be of some use which I've been working on this week involving LaTeX and MathML. I'm hoping to be able to post something about it in a day or two. Also, I've adapted LaTeX plugins for this forum and for bbPress (a forum system from the same guys as Wordpress) so I've played with these ideas a little.

    In conclusion, don't be too hasty to assume that no-one else is interested, and don't take our hesitation as being all bad. But, for myself, I only have a limited amount of time to devote to this project (unlike many of the contributors, I'm not a fully paid-up member of the n-categorical club), and have to pick my projects.

    • CommentRowNumber13.
    • CommentAuthorMike Shulman
    • CommentTimeJun 1st 2009

    Andrew, I don't think anyone was saying that n-category theory will eventually encompass all of mathematics. But although the nLab currently contains a lot of category theory and higher category theory, it is not exclusively dedicated to those subjects.

    It sounds as though there are three reasons being proposed why one might want to use mediawiki:

    1. Response issues. We should probably wait until we have a more solid server before condemning instiki for these.
    2. Long-term support. On this one I would be inclined not to fix something that ain't broke yet. As has been said, the content of the nLab is stored in a very portable form, and if and when the Internet moves on from instiki, it shouldn't be difficult to move it to whatever format seems best then -- and what to move to will probably be more evident then than now.
    3. Redirects. Is there something preventing us from simply implementing these in instiki? Would Jacques accept a patch if we wrote one? It doesn't seem like a complicated feature.
    • CommentRowNumber14.
    • CommentAuthorAndrew Stacey
    • CommentTimeJun 1st 2009

    Mike, my comment on "encompassing" referred to Eric's use of the word. Both of us were talking specifically about the Cafe at that point, not the whole of n-category theory. So you're right to say that no-one was saying that n-category theory will eventually encompass all of mathematics, but that isn't what I was responding to.

    As for redirects, I'll start a new discussion on that.

    • CommentRowNumber15.
    • CommentAuthorMike Shulman
    • CommentTimeJun 2nd 2009

    Sorry. This passage:

    ... n-categorical thinking ... may well prove to revolutionise mathematics, but then again it may well not. ... So saying that the n-cafe could and should ultimately encompass all of mathematics seems a little hyperbolic to me.

    sounded to me like you were saying "n-categories may not revolutionize mathematics... and therefore the n-cafe may not encompass all of mathematics."

    • CommentRowNumber16.
    • CommentAuthorAndrew Stacey
    • CommentTimeJun 2nd 2009

    Mike, I see how that could have been read that way. I was trying to tone down the rhetoric a little. The "so" was not so much a logical inference, more of a weak connector! Sorry.

    Eric, looking at your comments on Jacques' blog, may I just correct one thing. The time when Urs and I got each others' output was when we were logged in to the server simultaneously and is due to the strange way that the virtual server is configured. It has nothing to do with Instiki.

    • CommentRowNumber17.
    • CommentAuthorEric
    • CommentTimeJun 2nd 2009
    Thanks for clarifying that Andrew. I was just about to point out that conversation in the hopes that someone could explain what really happened.

    http://golem.ph.utexas.edu/~distler/blog/archives/001083.html#c024171
    • CommentRowNumber18.
    • CommentAuthorEric
    • CommentTimeJun 3rd 2009
    Argh. The server is dead again. This reminds me of what the web was like in 1993 :)
    • CommentRowNumber19.
    • CommentAuthorEric
    • CommentTimeJun 3rd 2009
    By the way, Toby. Jacques implemented a new redirect feature that I see is causing you some confusion. If the server were actually working, I could demonstrate. For now, you can see his instructions and description here:

    http://golem.ph.utexas.edu/~distler/blog/archives/001083.html#c024203