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.
    • CommentAuthorMike Shulman
    • CommentTimeJan 18th 2011

    Over at a discussion entitled sheafification there was a digression about what sorts of things should be put into the nJournal. Part of the disagreement seems to be about the “standalone-ness” of such things, and part about whether reviewing is author-driven or referee-driven. For concreteness, here’s a list of what I think the various things people have in mind are:

    1. Complete articles, either expository or original research or both, written for the purpose of publication here or elsewhere, by one or a group of people. Tom’s article is in this category.

    2. One or a group of existing complete nLab pages which someone suggests to referee as a submission, and freeze the refereed copy (with potential for re-refereeing and updating later).

    3. An extract from an existing nLab page, such as the statement and proof of one theorem.

    Please add, if you have some other possibility in mind.

    So, to what degree should “an article/submission/page-group/thingy in the nJournal” stand alone? The first type of submission certainly one expects to stand alone, although we may also hope that the material will be integrated into the corresponding pages on the main nLab. The others are unlikely to be as self-contained. I would, myself, tend towards wanting nJournal publications to be fairly self-contained as is reasonable for their contents. Certainly they can and will be interlinked with the nLab, but (for instance) I would be inclined against publishing “stub pages” (even those containing particular interesting facts) but rather only the sorts of pages which are well-organized, introduce the subject clearly, etc. On the other hand I can see wanting to publish a particular theorem, in which case it wouldn’t matter whether the page that theorem appears on is currently well-organized or not. But it would then be nice, I think, if the published theorem were supplied with a bit of introduction and discussion (with links) to make it into a coherent thing in its own right.

    A separate, but related, question is whether reviewing is author-driver or referee-driven. We have talked in the past about wanting a way to mark pages, or parts of pages, as “I think these are correct” or “this looks plausible” and sign our names, which is the same sort of statement that we want to get from “transparent refereeing” for nJournal submissions. However, it’s not a given that we should conflate the two; it sounds like Urs wants to and Zoran doesn’t.

    My initial feeling is that the nJournal is a little closer to the world of traditional publishing, in which submissions are sent to editors, who find referees, who at least have the option to remain anonymous, even if they also have the option to be nonymous and their reports are always more transparent than usual. The idea of “referee-driven” sort of “flagging” of pages and parts of pages seems to me like an extra step away, so that it might be good to have both, but separately and not conflated – sort of a “two-step bridge” between the traditional world and the more radical wiki-world.

    Urs’ argument seems to be that we don’t seem to be going to get software which will allow referee-driven “flagging” of pages and snippets in the near future, so let’s do that with the nJournal since that at least looks like it’s going to happen. (Is that right?) I’m a little worried that using the nJournal in that way might hinder its mainstream acceptance as a destination for author-driven standalone publications. One possibility might be to have another nJournal-like-thing existing in parallel which is referee-driven, maybe called something like “Extracts of the nLab”.

    Thoughts, anyone else?

    • CommentRowNumber2.
    • CommentAuthorDavidRoberts
    • CommentTimeJan 18th 2011

    Extracts is a nice term! Perhaps this can be a bit of a two-tier system, not unlike the AMS, where the Proceedings are at most 15 pages, but the Journal allows longer articles.

    • CommentRowNumber3.
    • CommentAuthorUrs
    • CommentTimeJan 18th 2011

    Is there a real need to make the distinction? Space is not a restriction. Would any traditional article that sits in the Publications on the nnLab suffer from there also being a reviewed version of geometric realization of simplicial topological spaces?

    My original idea had been as simple as this:

    we already have a unit of self-containedness established in practice: a single nnLab page. As we know, that can range from very short pages to very long ones, and it can contain units with sub-units that spread over several nnLab pages. We have a fairly well established tradition of ensuring a minimum of self-containedness of each page by writing it into the template page-pattern: Idea, Definition, Properties, Examples, References.

    Now all we do is that we assign to some of these pages a label: “has been peer reviewed in such and such a way”.

    To me this seems to be a simple, robust and flexible procedure, that we effectively already apply routinely on the nnLab: if a single theorem that is mentioned in some longer entry is important, we tend to give it a separate page. That could be reviewed separately, if anyone cares.

    In summry I think we ought to do two things:

    1. not create a lot of new structure with the nnPublications . We have a system that works well and can accomodate all sorts of publications: the nnLab. The nnPublications just consist of certain marked nnLab pages.

    2. get our hands on examples to sort out these issues in practice, as opposed to theoretically. We have now a referee report for geometric realization of simplicial topological spaces. Soon I’ll forward a public version that everyone can see and react to. Then we can bring bring that artcle through the refereeing process and put it into the nnPublications to get a feeling for how it works.

    • CommentRowNumber4.
    • CommentAuthorAndrew Stacey
    • CommentTimeJan 18th 2011

    I completely agree with Urs that we don’t need to make this separation formal. However, we could do well to signpost the more “snippety” entries as such since they will be a little different to what people expect to find. If we flag them as different then it should offset the worry that nPublications will be seen as too different. So “Extracts of the nLab” could be a “meta article”: in each “issue” we’d have an “Extracts of the nLab” listing those articles that were snippets rather than what might be deemed traditional articles.

    • CommentRowNumber5.
    • CommentAuthorzskoda
    • CommentTimeJan 18th 2011
    • (edited Jan 18th 2011)

    Urs has a vision of one mode of organization of nLab activities which is OK mode. My idea of separation is that it gives chance to the others without compromising the others modes of work and other purposes of work. If we treat the extracts from nLab on equal footing as say original contributions which were never part of the nLab we are setting single standard which compromises both: original contributors will be judged on the same level as editing work of well known material so original contributor from outside of our community will be less motivated to join. On the other hand, self-appointed referees like me, will be also less motivated to do proof-reading because it will be harder for us: we then referee the whole article from all aspects from correctness to the exposition; while I think that if I just read 1/5 of the article but the one which contains some substantial material we can judge I just submit a flag that I read it and like it, without saying anything about the rest; so I will feel that I just bother the community with my small opinion and would not like to bother editors which such a small piece of refereeing work. Also editors may get bogged down with judging the final outcome of the whole bunch of small refereeing issues. Editors make sense only for top level decisions which should happen sufficiently rarely, otherwise they will soon resign the positions.

    The last, but not the least, I reject the idea to cite the “extracts” rather than the original nLab articles: I always cite the sources and that is an intellectual principle which I will not compromise when citing. The orginal nLab pages have the whole history besides. So, I will cite the nJournal only as far as it contains something essential not in nLab.

    not create a lot of new structure with the nPublications

    I argue that this way nJournal will get less real publications (those originating out of nLab), less referees (as they must be then full referees of full articles), less editors (as they will border both with extracts and real articles, so too much) and less citations (I believe I am not the only one who likes to cite the original reference, what is in this case the original nLab page in some cited version). By making multiple categories of the same orginal nLab article there will be actually more confusion about the versions than if we leave flags within the original articles (that is if we do not extract extracts but referee within nForum threads connected to nLab pages).

    • CommentRowNumber6.
    • CommentAuthorzskoda
    • CommentTimeJan 18th 2011

    I also think, if we compromise the level to very unequal standards at the beginning, it will be perceived as such in future. So if we want to attract the articles from outside community and new contributors, we may be interested in setting it properly from the start.

  1. I subscribe each single word from Zoran in 5 an 6 above.

    • CommentRowNumber8.
    • CommentAuthorTobyBartels
    • CommentTimeJan 18th 2011

    I reject the idea to cite the “extracts” rather than the original nLab articles

    I would hope that the extracts would have a prominent link to the original article from which they were taken. With such a link, I would prefer to cite the peer-reviewed extract. Nevertheless, I wouldn’t expect you to follow my choice.

    However, if you do cite the Lab (whether or not there is also an Extract), it’s important to cite a specific version of the original article. (Perhaps you already know this.) Otherwise, years later, you won’t be citing what you thought you were citing. A revision number would be handy, and a date is also essential, since sometimes pages are moved (although it’s hard to use the date to track this; in principle it could be done with help from Andrew).

    • CommentRowNumber9.
    • CommentAuthorUrs
    • CommentTimeJan 18th 2011
    • (edited Jan 18th 2011)

    I have to admit that I don’t understand the problem. It seems to me that all issues raised have simple evident solutions. Let’s not make things more complicated than they are. I suggest again to concentrate more energy on getting something concrete up and running than to lose ourselves in abstract discussion of potential problems.

    We have currently two articles submitted:

    1. geometric realization of simplicial topological spaces;

    2. Leinster2010

    The steering committee, acting in place of the board of editors to be put in place, has found an expert refeeres for both. A good report for the first is already available and will be posted publically here, soon. A report for the second is announced for in roughly a month’s time.

    Anyone who is worried that something that should work cannot work with the nnPublications I encourage to try it out now. If you think that it will be strictly impossible that your refereeing of 1/5th of some existing nLab entry will not work on the nnPublications, let us know what thatt 1/5th is and I am confident we can figure out a way.

    • CommentRowNumber10.
    • CommentAuthordomenico_fiorenza
    • CommentTimeJan 18th 2011
    • (edited Jan 18th 2011)

    Hi Urs,

    the problem raised by Zoran and maybe by Mike, I guess, is that geometric realization of simplicial topological spaces and Leinster2010 are objects of a very different nature. For us it is natural to consider them both as two instances of tha same medium, and I agree that this is the way we should consider them, but we are now also proposing to a wide audience and to a public this. And this public is potentially more interested in giving an a priori value to something marked nPublications that to really go and see what it is. Extremely sad, I agree, but I fear this is a realistic pictre of reality outside there. So mixing the two aspects of the nPublications can be dangerous, discouraging authors from submitting and making editors feel to be in a place they didn’t expect to be.

    I would rather suggest that each issue of nPublications should have a section with classical papers, and a section with refereed nLab pages. the two are th same medium, but we could and should distingush them that bit that allows recognizing what is an nPublication from outside.

    I think it is important we have a quality certification on nLab pages, so that one can cite them as reliable sources and all that, but this should be in parallel with the journal and not the same thing. or, better, for us that will be the same thing, and with time we can figure it will be the same thing for everyone. but presently we should distinguish them: I can imagine nothing less incouraging to an author who is not already involved in a wiki project than submitting a paper to a place where it could be evalued as “oh, just a contribution to a wiki page”.

    Most important: we have no hurry. Let us make every step carefully.

    Said this, great thatwe already have a report on geometric realization of simplicial topological spaces and a referee on Leinster2010. I’m working on the itex version of the second one. Not going so fast as I’d like to, but having an nPub version does not seem too far, now.

    • CommentRowNumber11.
    • CommentAuthorMike Shulman
    • CommentTimeJan 18th 2011

    Urs, the issues being raised are not ones that will come up in the process of dealing with one or two individual submissions. If you read what I wrote, I said

    I’m a little worried that using the nJournal in that way might hinder its mainstream acceptance as a destination for author-driven standalone publications.

    Likewise Zoran wrote

    original contributors will be judged on the same level as editing work of well known material so original contributor from outside of our community will be less motivated to join.

    and cited several other potential problems as well. These are NOT problems which can be solved by “trying it out” right now; if they are real worries then it will take a while for them to manifest. If you could please tell us in what way the solution to these problems is “simple and evident” I would be much obliged.

    • CommentRowNumber12.
    • CommentAuthorUrs
    • CommentTimeJan 19th 2011

    If you could please tell us in what way the solution to these problems is “simple and evident” I would be much obliged.

    We already said we indicate which kind of review a page got. If you think it helps, we add a sentence “This is a research article.” to a publication. Or whatever it is you think should be indicated.

    And I do think we can try it out, that’s precisely the point of having Tom’s article alongside with the one on geometric realization, to have two examples of different kind.

    • CommentRowNumber13.
    • CommentAuthorUrs
    • CommentTimeJan 19th 2011
    • (edited Jan 19th 2011)

    If you think it helps, we add a sentence “This is a research article.” to a publication.

    Or, if this is what you are suggesting, we put some submissions to the nnPublications into separate wiki webs that are titled something like “PL no 2011 vol. 2”, while other submissions are collectively put into the main nnPublications web. And then maybe we declare that only the former are handled by the editorial board, while the latter receive their “reliable”-status by enforced transparent refereeing. Or the like.

    I think as soon as the two submisions that we currently have have made it through the refereeing process (if they do) we create nn-Pub-pages for them and then play around with their appearance, their titles, their extra information boxes ec. until we are satisfied.

    • CommentRowNumber14.
    • CommentAuthorzskoda
    • CommentTimeJan 19th 2011

    Urs 13: I see that we are getting closer to the agreement, though this is not a complete solution yet. On the other hand, this discussion should not prevent with going with vol. 0 in a different way. As Mike pointed out the problems about the meaning, motivation and quantity which are central are not relevant with some small number of articles, but only later. While we should not wait for the scale problems to happen to discuss and define things correctly ASAP, in the same time, first several articles of both kinds can be handled right now in any way people handling this find appropriate now.

    • CommentRowNumber15.
    • CommentAuthorTim_Porter
    • CommentTimeJan 19th 2011

    One way out of some of the problems is to give a ’heading’ as you suggest, Urs. My preference would be to have a simple heading saying what ’category’ the entry is in, e.g. (nPub:Survey), (nPub: Article), etc.

    • CommentRowNumber16.
    • CommentAuthorUrs
    • CommentTimeJan 19th 2011
    • (edited Jan 19th 2011)

    My preference would be to have a simple heading saying what ’category’ the entry is in, e.g. (nPub:Survey), (nPub: Article), etc.

    Might be tricky. An nnLab entry may start out more a survey and later become more an article.

    My impression is we have two different applications for the nn-publications, and that the worries expressed here are about how explicitly they are being separated.

    1. add a review-functionality to nnLab articles similar to what they have at the Manifold Atlas whose main purpose is to improve the commonly and reliably available net knowledge in the scientific community.

    2. host a traditional style online journal whose main purpose is to further the submitting author’s career.

    I think we can easily ensure that the second functionality works well, too, by addding sufficient markers, whatever it takes. For the first purpose however I would suggest that we keep it a very simple add-on to the nnLab: no extra structure, just a place to keep reviewed versions of nnLab articles.

    • CommentRowNumber17.
    • CommentAuthorTim_Porter
    • CommentTimeJan 19th 2011
    • (edited Jan 19th 2011)

    I don’t think I made my idea clear. We have a page for a survey article and in the title area we have:

    Homotopic Incoherence

    prepared by: Joe Bloggs, Ernest MacShea, Dubious O’Callahan.

    (nPub: Survey Article)

    If its status changes later it is just a keystroke away from another form. (Nothing automatic, just used for the convenience of the user, i.e. a note to the reader on how this is seen at the time of acceptance by the editors.)

    We might also have (nPub: Book Review), (nPub: Refereed extract), etc. and we do not have to list all the possible categories initially, just keeping it informative and fairly informal.

    • CommentRowNumber18.
    • CommentAuthorUrs
    • CommentTimeJan 19th 2011
    • (edited Jan 19th 2011)

    Oh, I see. Sorry, I had misunderstood you. Yes, that certainly makes sense. We can supply detailed meta-information for every article. We should try out possible info-boxes once we have the first test-case articles gone through the peer-review.

    • CommentRowNumber19.
    • CommentAuthorTim_Porter
    • CommentTimeJan 19th 2011

    I frel that that goves a lot more freedom of movement once the machine is rolling more smoothly. Some of the other (more technological) ideas could be good but might tie us down as to future developments too soon.

    • CommentRowNumber20.
    • CommentAuthorMike Shulman
    • CommentTimeJan 20th 2011

    Okay, I think this is moving closer to something good. If the main nPub web page explains that it contains different types of publications that are submitted and refereed in different ways, and each publication has a header saying something about how it was submitted and refereed, then probably that is good enough. And yes, I agree that that the best way to present that is something we can experiment with. Sorry if I got a little heated.

    • CommentRowNumber21.
    • CommentAuthorMike Shulman
    • CommentTimeJan 23rd 2011

    Sorry to reopen this can of worms, but I thought of a reason why headers for each publication may not be as good as separately named “journals.” Suppose some bean-counting mathematician first encounters the name Publications of the nLab as a bibliographic reference to an extract-type publication. Curious, or interested in the subject, he looks up the URL, and finds that it is a refereed wiki page. Then some time later he is evaluating someone’s CV and sees a publication cited as in Publications of the nLab. “Oh,” he thinks, “that’s just a refereed wiki page, not an actual published article” — not bothering to follow the URL and therefore see the header on it saying that this is a different sort of publication.

    • CommentRowNumber22.
    • CommentAuthorTim_Porter
    • CommentTimeJan 23rd 2011

    That point suggests that it would be a good thing to think of the style of citation in other papers, and CVs etc of an article in the nPub. My long term idea was that eventually when the journal was up and running there might be a need to spawn other separately named versions and the header idea is a means to make that feasible with little effort later on, thus leaving our otpions more open.

    Here is a citation model

    J.Bloggs, Theory of everything, Publ.nLab(cat:article), 2011, no.64, p. 1-1000.

    No hassle for us yet for citations it should confuse the automatic counting software enough!!!! (:-)<-Evil luddite grin!)

    • CommentRowNumber23.
    • CommentAuthorUrs
    • CommentTimeJan 23rd 2011
    • (edited Jan 23rd 2011)

    "Oh," he thinks, "that’s just a refereed wiki page, not an actual published article"

    Why necessarily "just a refereed wiki page"? and why "not an actual published article"? My idea of having peer review on a wiki was to turn this around: to get to a point where people say

    "Oh, that’s even a peer-reviewed collection of entries on the nnLab, not just a traditionally published article."

    Of course the moment that the project starts, this won’t be what people think. But if we do it right, I see a good chance that we can get to this point. And this would be something interesting.

    If we are very scared that people might notice the wiki-flavor and the relation to the nnLab of the project, then, sure, we could set up a traditional journal. But what I am not sure about is: why would we care about the project then in the frist place? There is no lack of traditional journals.

    Me, personally, I won’t have much energy to help set up a traditional-style journal with explicitly no relation to the nnLab. But of course that need not stop you and others from setting one up. Aside to the Publications of the nnLab there is room for plenty of further instki-webs on the server.

    But I have a vision: let’s be bold and not be shy about promoting wiki-pages as a good source of stable scientific information. If only we ensure that the peer review is good and the quality of the refereed pages is really good, then I see a lot of advantages of a refereed wiki page over a traditionally published article.

    I think we should try it.

    I am not so keen about tons of submitted isolated research articles anyway: who is going to take care of all the work? After all this is run by a community of volunteers. Me, I am willing to volunteer as long as I see that the process contributes to a greater good, such as the development of the nnLab to a good general source of information. If somebody does not care about the nnLab and just wants the official credit for his 80 page-writeup of his work, without caring about integrating it into the nnLab to make it worth more than the pieces it consists of, then why would I be motivated to help find referees, convert it to instki format and all that trouble? Why would you be motivated? And even if you were, how would you find the time?

    • CommentRowNumber24.
    • CommentAuthorMike Shulman
    • CommentTimeJan 24th 2011

    Urs, I think you’re still missing the distinction between evaluating something as a mathematical resource and as a means to evaluate its author(s). Reviewed wiki pages have numerous advantages for the former, and we can hope to convince people of that over time. But they are trickier for the latter, as we have discussed in the past. So whether something merits a deprecating “just” or “only” may depend on what one is looking for.

    Also, I hate to say this, but please don’t put words in my mouth. I am not “scared” that people will “notice” the wiki-flavor and connection to the nLab. And I am all in favor of boldly promoting wiki pages as a good stable source of scientific information.

    But I also think that all of the other advantages we have discussed, and described in detail over here, have the potential to attract mathematicians whose primary interest is not (yet) the welfare of the nLab — at least, if they can have an expectation of receiving some credit for it. That doesn’t mean dissociating the nPublications from the nLab! We may hope, or expect, to draw such a person into the nLab at least a little bit through the process of community refereeing and integration of their submission. But even if we failed at that, I do not regard the welfare of mathematics as synonymous with the welfare of the nLab, and I hope and expect that the future of mathematical publication is not simply that everything will eventually become part of the nLab! We have always been very explicit that the nLab is the lab notebook of one research group with one particular POV.

    • CommentRowNumber25.
    • CommentAuthorUrs
    • CommentTimeJan 24th 2011

    Hi Mike,

    my apologies if I put words in your mouth.

    I don’t want to cause trouble. But I feel I need to try to reply. I think maybe it would help (me) if instead of you trying to make me grasp the problem more properly you would concretely propose solutions, as you envision them. I don’t expect to have any real objections, rather I seem to have somewhat different inclinations.

    I can roughly imagine myself trying to help the welfare of the nnLab. (And I see potential for it way beyond the little community currently active on it.) But the welfare of mathematics as a whole? I feel overwhelmed. I never intended to run a traditional journal. I only intended to add peer-review to the nnLab!

    But we don’t all need to work on the same project. As I said, if you and others feel more attracted to running something that is more a traditional journal than an add-on to the nnLab, then please do. Maybe that’s what all this discussion here is about: that we should split the projcect currently titled “Publications” into two really different projects, run by different (if maybe intersecing) sets of people. One a simple add-on to the nnLab, the other an online journal that just happens to run on some wiki software.

    • CommentRowNumber26.
    • CommentAuthorMike Shulman
    • CommentTimeJan 24th 2011

    I don’t know that I have a definite view of what I think we should be trying to accomplish yet, which is probably part of my problem. I didn’t intend “the welfare of mathematics as a whole” to sound grandiose—I think we all contribute to “the welfare of mathematics as a whole” when we do little things like referee other people’s articles for traditional journals, write blog posts, go to conferences, and answer MO questions. I don’t want to run a traditional journal either.

    I only intended to add peer-review to the nLab!

    I’m confused. How does publishing Tom’s expository article on topos theory fall under the heading of “adding peer-review to the nLab”? I think if I understand that, I’ll have a better idea of what you want to accomplish, and then maybe I can articulate some concrete suggestion.

    • CommentRowNumber27.
    • CommentAuthorUrs
    • CommentTimeJan 24th 2011

    I’m confused. How does publishing Tom’s expository article on topos theory fall under the heading of “adding peer-review to the nLab”? I think if I understand that, I’ll have a better idea of what you want to accomplish, and then maybe I can articulate some concrete suggestion.

    Right, so that’s why I had originally emphasized that clause that anyone who wants to submit anything for peer-review with us first has to put the material into the nnLab. As soon as that happens, it is part of the nnLab and hence subject to peer review of nnLab pages.

    In my world view the best possible situation would have been the following:

    Tom feels the energy to write an exposition on topos theory. He goes to the relevant nnLab pages and starts expanding and polishing them, adds new pages as necessary and links to them. This way all the material he ads benefits from all the other material that is already there, and vice versa. For instance where he says “I don’t know how to directly see the relation between subtoposes and Grothendieck topologies” it could instead say, “We don’t discuss the relation between subtoposes and Grothendieck topologies here on this page, for more on that see the page reflective subcategory”. Conversely, the existing topos-theory on the nnLab – which is by now quite a remarkable, i think, if certainly still very imperfect fusion of the classical textbook material on topos theory and higher topos theory which in this unified form is not available anywhere else in the literature – would profit from having more introductory and expository material coming with it, for the general benefit of nnLab readers.

    The next best thing to such an approach where “we all jointly work on decent online material on topos theory” is that Tom writes his exposition in an isolated piece of text and we put it just into parallel nnLab pages. This is better than not having the expositional material at all on the nnLab, and serves as a compromise between the goal of jointly creating the best possible resource on topos theory – eventually – and Tom’s desire to officially record individual credit for his work, instead of opting for credit as part of a Bourbaki-like team.

    So in summary, the approach is: you have to contribute to the nnLab in order to profit from nnLab peer review, but if you insist and don’t want to go through the trouble of properly working your material into the existing material, then, well, there is nothing in the nnLab policy to stop you from creating parallel pages. So please do, and as a bonus for your efforts we see if we find volunteers to peer-review your material.

    • CommentRowNumber28.
    • CommentAuthorMike Shulman
    • CommentTimeJan 26th 2011

    Thanks! If anyone else is still paying attention to this discussion, I’d like to hear what they thought, because this is quite different from the impression I got from the original discussions about the nJournal (this time around, not previous discussions that were specifically about refereeing nLab pages). But if I was the only one who had a different impression, then I must have been misinterpreting.

    • CommentRowNumber29.
    • CommentAuthorzskoda
    • CommentTimeJan 26th 2011
    • (edited Jan 26th 2011)

    Right, so that’s why I had originally emphasized that clause that anyone who wants to submit anything for peer-review with us first has to put the material into the nLab.

    I find this both surprising and disappointing. I mean I understood mention of such things a while ago as a matter of wiki format, not as a matter of purpose/content classification, as it is here more clearly delimited. Asking people to join nLab in the narrow sense, in order to have the access to the formatting, editorial services and content connection to the circle of labs including joyalsCatlab, nLab and others, is an additional turn off for many who did not join nLab in last 2 years. Asking editors to do something just for the purposes of enhancing nLab and not for the extension of the wide benefits for the community from existing software, servers and community, is possibly also disappointing to the editors. nLab is a known format, copyright model, POV etc. I would hope to have extension of this with nJournal, not limited to nLab community and purpose (reread John Baez’s hope in his comments about the openess of possibilities for nJournal, which we now shrink to a nLab marking device, which I was hoping to be just for the “class snipet” so to say). Do you really think, one will attract more people with nJournal this way ?

    • CommentRowNumber30.
    • CommentAuthorTobyBartels
    • CommentTimeJan 26th 2011

    You’re not the only one, Mike. At least now I understand why Urs

    had originally emphasized that clause that anyone who wants to submit anything for peer-review with us first has to put the material into the nLab

    The scope of the nJournal now sounds rather more limited than I had thought. Any particular limit can in principle be overcome, by putting the material on the nLab, but I didn’t grasp that aspect.

    • CommentRowNumber31.
    • CommentAuthorTobyBartels
    • CommentTimeJan 26th 2011

    I’d be interested to know what the people who’ve agreed to be editors thought about this part!

    • CommentRowNumber32.
    • CommentAuthorMike Shulman
    • CommentTimeJan 26th 2011

    @Toby #31: Ditto! How was the project presented to them when they were asked?

    • CommentRowNumber33.
    • CommentAuthorTom Leinster
    • CommentTimeJan 27th 2011

    @Mike (32): here's part of what I wrote to Carlos Simpson:

    --

    The people active on the lab now want to set up a way to stamp some of this content with the approval of peer review, thus marking it as stable and reliable. This should increase its usefulness to everyone. So a frozen, write-protected, citeable version of the approved page would be created, along with an evolving one that could accommodate changes and improvements. If the evolving page became significantly different, it too could later be submitted for review, etc.

    There's been a lot of discussion over the last couple of days about how best to achieve this, and as a result of that we're experimenting with the idea of calling it a journal. Being part of the nLab, it would be based on wiki technology, with all the advantages this has for interconnecting mathematical content (e.g. click on a word to see its definition). But otherwise it's supposed to work much like a traditional journal: people may submit content to it, and an editorial board chooses referees, who review the material and provide reports, etc. Material submitted might be a page that's evolved on the Lab over time (perhaps involving many contributors), or it could be a Latexed paper submitted in the classic way (and then converted with the aid of existing software).

    • CommentRowNumber34.
    • CommentAuthorUrs
    • CommentTimeJan 27th 2011
    • (edited Jan 27th 2011)

    I don’t have the energy nor intention to run a stand-alone math journal. If you do, please go ahead!

    I have energy and intention to add peer review to the nnLab, I want something like they have at The Manifold Atlas . This is already a very wide field, given that you can open about any page on the Lab.

    If you want more or something else, go ahead. The things we are talking about exist only to the extent that somebody actually does something. I am running already out of steam with the long discussions we have about things that do not yet exist. I need to concentrate the little spare time I have on looking at things that do exist.

    I have started pushing the entry on geometric realization through the system, and if nobody else looks into reacting to the referee’s requests that we have there, then I will do that some time. Andrew is thankfully pushing the entry on Tom’s writeup. Once that is done, I trust that you will find some way to label and present it as “genuine publication” and “not just a wiki page”. I think that should be easily possible.

    Concerning what the editors we have emailed think: I have intentionally stopped us from asking more would-be referees for the very reason that we first need to figure out ourselves what it is they will be editors of. I stiil think once we have our first two test-case publications refereed and up online and sorted out how exactly to present and label and title and reference them, we’ll all have a better idea. Then we should email the candidate referees again, show them what we have, and ask them whether or not they think this is something they want to act as editors for.

    • CommentRowNumber35.
    • CommentAuthorUrs
    • CommentTimeJan 27th 2011
    • (edited Jan 27th 2011)

    Ditto! How was the project presented to them when they were asked?

    I have pointed the people that I emailed to the project’s proposal web page Proceedings preparation (nlabmeta).

    This page, by the way, is or was intended to be the Sandbox for the project. Anyone who has a proposal should, I think, either edit that page or branch off another version, for us all to see.

    • CommentRowNumber36.
    • CommentAuthorTim_Porter
    • CommentTimeJan 27th 2011

    I suspect that much of the heat in the above discussion results from misunderstandings resulting from reading between the lines, jumping to conclusions, sometimes slight awkwardness in wordings, etc. I do think that Proceedings preparation (nlabmeta) gives a good working model for a start on the nJournal. The interpretation of the wording there will evolve, but, as I said, is a good start.

    • CommentRowNumber37.
    • CommentAuthorzskoda
    • CommentTimeJan 27th 2011
    • (edited Jan 27th 2011)

    I don’t have the energy nor intention to run a stand-alone math journal. If you do, please go ahead!

    Once you establish the editorial board, with the help of community you really do not need to take care of all the modes of work of the nCommunity. If the software and rules are done at the beginning with more than one mode in mind, you can really just enjoying hearing that not only your main mode but many others give fruit. As did Joyal’s independent catlab…

    If you want more or something else, go ahead.

    I also wanted less i.e. the consensus to mark the pages even if they are not submitted. Submition is a threshold which requires very complete refereeing, hence I am not likely to contribute. With less flexible system one will have smaller number of people helping so you (and remaning enthusiasts of single-tier system) will have to do more on your own.

    • CommentRowNumber38.
    • CommentAuthorTobyBartels
    • CommentTimeJan 27th 2011

    @ Urs

    I don’t have the energy nor intention to run a stand-alone math journal.

    You wouldn’t be running it anyway; the editors would be. You (together with others) would be setting it up, and (together with, especially, Andrew) running the overarching wiki framework on which it sits (most of which will still be the nnLab, just as it is now).

    Reading Tom #33, it seems that Carlos at least has been given an understanding similar to your, Urs. However, Proceedings preparation (nlabmeta) is much less clear to me; there is nothing there to suggest that this wouldn’t take traditional journal articles very much like a traditional journal would. (Several differences are highlighted, but not the stuff that confused me and Mike in comment #25 et al.)

    • CommentRowNumber39.
    • CommentAuthorUrs
    • CommentTimeJan 27th 2011
    • (edited Jan 27th 2011)

    You wouldn’t be running it anyway; the editors would be.

    is that clear?

    The first thing one of the editors asked me was if he had to take care of any of the work of adding hyperlinks and info boxes.

    And do you see any of the volunteer editors lend a hand in the work of setting this up? Aren’t they mostly lending their names to the project, and later their ability to select a referee?

    I think traditional journals aren’t just run by the board of editors, either. There is good supply of staff. Where is our staff?

    there is nothing there to suggest that this wouldn’t take traditional journal articles

    It says it will, if they are put on the nnLab.

    The discussion here started when Mike voiced the wish to publish articles in a way that indicates them distinctly not be associated to the nnLab. That made me nervous. But of course, as I said, if this is run by somebody, I won’t stop this project. As I sugested before, we may want to split this explicitly into two projects, one for refereeing nnLab pages, the other for something else.

    • CommentRowNumber40.
    • CommentAuthorTobyBartels
    • CommentTimeJan 27th 2011

    The first thing one of the editors asked me was if he had to take care of any of the work of adding hyperlinks and info boxes. [etc]

    OK, you are right.

    It says it will, if they are put on the nnLab.

    Sure, but neither Mike nor I ever understood why you wanted to make that a requirement. We never really understood your intentions with this project! Do the editors?

    • CommentRowNumber41.
    • CommentAuthorUrs
    • CommentTimeJan 27th 2011

    We never really understood your intentions with this project! Do the editors?

    Hm, the thread which started this was called “new web: nLab (reviewed)”. But okay, good that we talked about it! :-)

    • CommentRowNumber42.
    • CommentAuthorTim_Porter
    • CommentTimeJan 27th 2011

    I think that for submitted Lab pages there is no problem. If someone wants to submit and article to the nProc then there is one problem and that is that it would be better if they did, at least partially, put it themselves on the nLab system first, as an ordinary article is not optimised in its structure for here. They will need to try things out and may need help from more established ’oldhands’. Apart from me and André, the editors have little no experience of the nLab system and will only be able to direct submitted material to the referees, (who may need advice on style etc.).

    I think that when the nProc is floated then we should advise potential authors to discuss their ’project’ with an nLab expert (here on the forum) first to get some idea on how to proceed. (We are all novices in the actual production of the nProc!!!!)

    • CommentRowNumber43.
    • CommentAuthorUrs
    • CommentTimeJan 27th 2011

    I think that when the nProc is floated then we should advise potential authors to discuss their ’project’ with an nLab expert (here on the forum) first to get some idea on how to proceed. (We are all novices in the actual production of the nProc!!!!)

    Yes! For that reason i think it would in fact be good to have more such “test-case” submissions as soon as possible. There is no reason to wait until “it starts”. Because nothing will start by itself. I think everyone with a vision for what kind of publication sould be possible in the end would do good to try to find an author willing to serve as a test case, have a submission refereed and then put it up online in the intended form.

    • CommentRowNumber44.
    • CommentAuthorAndrew Stacey
    • CommentTimeJan 27th 2011

    For what it’s worth, here’s my thoughts on some of the above. I wouldn’t want to prohibit people submitting stuff to the nWhatever that they’d written elsewhere. I would be surprised if anyone would want to do that and would not want to have it wiki-linked and so forth. I would expect that the only hurdle would be the actual doing of it, and hopefully my script together with what we’ve learnt from doing Tom’s article will go a long way to lowering the workload of that. And I wouldn’t expect the editors to do that.

    I would like to see a large range of possibilities, from small nLab snippets to full blown articles. I think that at first it’ll be weighted towards the “article” end but as it develops we should be open to broadening the scope. Of course, we will need to have some way of differentiating the levels, but I don’t think that we can guess ahead of time what will be necessary there.

    • CommentRowNumber45.
    • CommentAuthorzskoda
    • CommentTimeJan 28th 2011
    • (edited Jan 28th 2011)

    The first thing one of the editors asked me was if he had to take care of any of the work of adding hyperlinks and info boxes.

    And do you see any of the volunteer editors lend a hand in the work of setting this up? Aren’t they mostly lending their names to the project, and later their ability to select a referee?

    Urs, we are talking the difference between having a original-article and snippet sections. The difference between the two is exactly that the first needs more complete and appointed refereeing: that is by definition what the editorial board does. So the all the difference which comes up with more traditional articles is the more regular (non-self-appointed) refereeing process. Hence the thing which you do complain is exactly the thing which by the definition editors do. Now you bring issues which we anyway have to do, having two tier-system or one, and argue on the basis of the fact that those issues are not handled by the editors, that one should not count on editors when setting up the specific section of a more of a real journal. This is clearly a false argument.

    Finally your complaining that you do not have help, is turning off much help, for example mine, which is with insistence on a single tier system annihilated. As I said early in the discussion: I referee regular papers in journals, as a service to people who do good work and want their work recognized. Regarding that a regular journals count a lot in career I consider my work worthy. But if it were for some less original and less counted forms I would not spend that effort. On the other hand, on snippet side, I would sometimes just put some comments if we organize that part without board, just in a linked section of nForum – comments on quality of nLab articles. For that I do not see any need of editorial boards, it can be self-appointed, self-initiated commenting, not necessarily complete, does not need any new software (just agreement that we will use a section of nForum with backlinks for it) and is not much work as it does not require complete refereeing, no deadlines, and no official suibmission procedure. In your system, refereeing original articles has too little value as it is in the same section as the minor expository writing; refereeing snippets is to difficult as it requires complete, appointed, editored, etc. action which is too much for a casual worker and commenter. Hence I am not to help in either of the two, if the single tier system is to be official.

    Edit: I also think that one does not need to have stuff from nLab separated into special copies to be either cited or commented/approved for quality. Wikipedia has a system of labelling quality without forming multiple copies (they have several levels of quality mark for entire articles + they have a talk page related to every page where they leave comments on the quality, exposition, validity, questions to other writers and alike). However, I think that creative new combinations and extracts based partly or fully on parts of or multiple nLab articles should be creatively encouraged.

    • CommentRowNumber46.
    • CommentAuthorTobyBartels
    • CommentTimeJan 28th 2011

    Hm, the thread which started this was called “new web: nLab (reviewed)”.

    Yeah, but it quickly got more general than that … or so I thought!

  2. I agree with Zoran. And I think we can easily handle all different kind of nPublications in the same “container” while at the same time preserving their specificities. in geometric realization of simplicial topological spaces – Submission Entry we are indicating that its type is survey/encyclopedia article. So why not be clear from the beginning on which will be the types in nPublications, and on how different types will be handled. For instance I agree we should not bother an editorial board for snippets, since we can most conveniently hande that by nforum. maybe can make this a bit more formal having a specific editorial board of nPublications(type:survey/encyclopedia article) which could consist of people from the steering comittee with eventual adds (I would happily join the editorial board for this type).

    I think there are many different aspects involved here, and that developing just one will depress them all, while developing all them will improve them all at the same time.

    the types that come to my mind are:

    • encyclopedia article
    • journal article (this could be either a survey or an original reserach article, but I would not distinguish between them: where does the survey ends and originality begins?)
    • monography

    and I would think of an editorial board for the first type, and another for the other two types (jointly).

    • CommentRowNumber48.
    • CommentAuthorUrs
    • CommentTimeJan 29th 2011
    • (edited Jan 29th 2011)

    I have done the following:

    • created a new web nLab (reviewed);

      (the password is again the system password, if you want to edit this and don’t have the password make a copy of it at nLab meta, and edit there)

    • moved the single refereeing-in-process-page that we currently have from the web nPublications to there.

    • CommentRowNumber49.
    • CommentAuthorTobyBartels
    • CommentTimeJan 29th 2011

    So what’s the difference between these two webs now?

    • CommentRowNumber50.
    • CommentAuthorUrs
    • CommentTimeJan 29th 2011

    I think nLab reviewed should simply record all nnLab pages that have received peer review in any way, for one reason or other. Just extra information for those using the information on the nnLab. Where these reviews will come from we’ll see. Probably from the same source that nnLab entries come from: from people that happen to think it would be useful to have them.

    With the nnPublications web you (some of you) should see what you want to do. I’ll bow out of that project, though. I should not have made my original desire to have review of nnLab pages be pushed into this larger direction, it has caused a lot of confusion.

    • CommentRowNumber51.
    • CommentAuthorTobyBartels
    • CommentTimeJan 29th 2011

    Urs, I fear that I’ve pushed you out of the Publications (and I don’t even feel very much involved in that, anyway, I mostly just commented on the name and asked John Baez to comment), simply because I (well, and Mike, apparently) was confused over what you wanted.

    So one way or another, I hope that you are getting what you wanted!

  3. whichever form things will take, I think anything done with care will be good, and the interactions between the variuos aspects will be a resource. it seems we are at the point where the original project is splitting in two subprojects, each of which is more adapted to its scope than the original project as a whole (they call this evolution..). so in nPublications we are going to have reviewed material from the nLab and an nJournal part which will be very close to a classical journal, but with the addiional value we are describing in the nPublications home page.

    No one should be forced in a project, things works if we all do what we are interested in doing. so, for instance, Urs seems to be more interested in nLab reviewed and Zoran and me in the journal part. this is not bad, it is good: let’s go on with both projects, and let us keep them close and interacting.