## Not signed in

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

## Discussion Tag Cloud

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

• CommentRowNumber1.
• CommentAuthorMark C. Wilson
• CommentTimeJul 2nd 2012
• (edited Jul 2nd 2012)

In an attempt to resurrect this forum (anyone going to reply to my last post?) I note the launch of new gold OA journals by Cambridge University Press, featuring Tim Gowers, Terry Tao, Henry Cohn and other Elsevier boycotters as editors. See http://gowers.wordpress.com/2012/07/02/a-new-open-access-venture-from-cambridge-university-press/. I am sure Henry will have plenty to say about this venture, which looks very promising to me. Again, I will state that I find the fee to be much larger than necessary, but perhaps for a premium offering it is acceptable. Anyway, I hope it goes well.

1. Thanks for the information, this is very interesting. If everything goes well, I might announce another such initiative in a few months.

• CommentRowNumber3.
• CommentAuthorScott Morrison
• CommentTimeJul 5th 2012

I’m really excited about this one; in fact I’m holding onto an article for the October 1 opening of submissions.

2. That is planned to be non cheap :( After 3 years which are for free.
3. Concerning the price, there is several things I would like to stress. First, the price of a proper copy-editing is roughly 30 dollars per page even for cheap and efficient journals, so unless we abandon it, the proposed price (750 $per published article) seems a fair minimum. Second, right now I guess we pay far more by published paper via subscription (but who spends the money is different, so generalized Gold OA would at least disturb the current balance). Last and most importantly, the three years without fees gives time to, maybe, include FoM in a instution-pays paradigm, if people are wanting to do it. For example, I would welcome the CNRS to pay annually for papers published by its units in FoM, taking the money from subscription (and therefore increasing the pressure on subscription prices). This would be all the more efficient because CNRS currently develops a policy to secure subscriptions to cheap, not-for-profit, high quality math journals by national deals. If enough large institutions follow this path, we could expect fee waivers for all individuals not already covered by these institutions. This may be too optimistic, I don’t know. 4. I am sorry what means "copy-editing" ? Do you mean hard-copy ? if not why electronic journal should cost that much ? • CommentRowNumber7. • CommentAuthorDavidRoberts • CommentTimeJul 9th 2012 ’Copy-editing’ means the process of editing the manuscript to correct for sloppy grammar, typos, missing or superfluous bibliography entries and so on; in general, the actual process of editing the text, rather than what the journal editors do. For instance, Nature Publishing Group say that the people they use for copy-editing have PhDs in the field they look at papers in, so as to get people who understand the material reading and correcting the text. This means that hopefully they won’t ’correct’ and in the process introduce errors. To contrast, come big publishing houses contract out their copy-editing to the cheapest possible companies. In the case of Springer’s publication of Mumford’s collected works, this caused huge pain to those putting the collection together. (Elsevier also outsource to very cheap external editing houses in countries where English is not the primary language). Unless an author is highly proficient in written English (and this excludes many native English speakers) copy-editing, when done correctly, has a non-zero input to the publishing process. Whether this is always necessary is another question. People complain about badly written papers, but usually these are structural problems, not fine-grained problems of style (isn’t it said that the international language of mathematics broken english?). Tim Gowers points out in a comment to his post that for a large number of papers in mathematics, i.e. those that are published as a flag: “this paper is publishable”, copy-editing is probably not necessary due to the small and specialised number of eventual readers and costs can be saved in not spending time on copy-editing. 5. How to estimate the money for this "copy-editing" work ? 6. It seems likely that mathematical publishing will develop a stratification of the amount of work on the form of papers, correlated with the importance of journals. I do not see how Pub. Math. IHES or Acta Math. or Annals of Math. could stop print (at least in the not-too-far future) ; other top journals will probably continue to have copy-editing like FoM plans to do (hopefully even better than now for those who are at Springer or Elsevier, if they quit them), but the vast majority of journals could drop it to become really cheap ; this move makes it possible to have most journals run by institution (a university or, more probably, a national research institution) without subscription nor author fees. One big problem is for institutions (notably academic societies) whose funding rely on journal subscriptions, we should maybe devise a strategy to redirect part of the money saved from subscription to them. Why not have a kind of support subscription open to labs or universities? • CommentRowNumber10. • CommentAuthorAlexander Woo • CommentTimeJul 11th 2012 Let me try to put some kind of lower bound on the cost of decent copy-editing. For most of my papers, the last thing I do before posting on the ArXiv is copy-editing where I focus entirely on the language and ignore the mathematics (which i assume by then is fine). I can manage to do about 5 pages an hour. Counting benefits and overhead, I cost my employer over$100K a year. At 2000 hours a year, that comes to $50/hour, or$10/page.

Keep in mind that I am familiar with my writing style and know what kinds of grammatical mistakes I am more or less likely to make, so it is easier and faster for me to copy edit my own papers than for me to copy edit someone else's. I am also familiar with the mathematics in the paper, so copy-editing issues requiring judgement are easier for me. (The most common judgement required is determining whether a clause or phrase is 'essential' or 'parenthetical'; 'parenthetical' clauses or phrases should be set off by commas, but 'essential' ones should not.) Also, being essentially a native writer of English, I don't have too many mechanical problems to fix.

A $30/page estimate for a mathematically trained copy editor dealing with a substantial number of authors whose native language is not English seems entirely reasonable. (Keep in mind that even a natively German author with fluent English tends to create a good deal more copy editing work than I would because German requires commas in some places where English forbids them.) • CommentRowNumber11. • CommentAuthorMark C. Wilson • CommentTimeJul 12th 2012 I have read that large publishers such as Elsevier outsource copy-editing to places where the hourly rate is much lower. Of course their end product is not very good, in my experience. I think the key point about FoM is that the costs are promised to be transparent. Once we know what we are paying for, we can make a decision as to whether we think it is worth it. I can see benefits to the cost being borne more by the authors. However this will discriminate against those whose native language is not that used in the journal. I propose (seriously) that Esperanto would make a lot more sense for mathematics journals. It is much fairer and easier to learn for most people than English, for example. • CommentRowNumber12. • CommentAuthorDavid Speyer • CommentTimeJul 12th 2012 The editorial free lancers association suggests the following as typical rates for nonmathematical prose. Note that they specify that a page is 250 words which suggests that they are double spacing or using large fonts; I just counted a page from one of my papers and came up with 450 words. Standard copyediting 1 hr = 5-10 pages = 30-40 dollars Heavy copyediting 1 hr = 2-5 pages = 40-50 dollars Undoing the double spaced page count, that comes to a range of 8-50 dollars a page; I find it quite reasonable to believe that math is on the high end of that. I also have to say that 5 pages per hour is fast in my experience; I always find it depressing how long it takes me to copy-edit my work. Alex, it looks like you might have a reasonable alternate career as a copy-editor. :) • CommentRowNumber13. • CommentAuthorNoah Snyder • CommentTimeJul 13th 2012 No formatting or copy editing is clearly preferable to cheap outsourced low-quality copy editing. (It costs nothing and doesn't introduce errors.) High quality copy-editing at least adds some value, but personally I do not think it's worth the (large) cost for the vast majority of papers. • CommentRowNumber14. • CommentAuthorMark C. Wilson • CommentTimeJul 13th 2012 Agreed. I very much doubt that any math publisher is paying anything near what David listed. And I don’t know of any that does high quality copyediting. As suggested by many people, I think it makes sense to separate this function out from the usual refereeing process and put the onus on the author to use it if and only if necessary (and pay for it too). Or if the result is excellent but the writer can’t write properly in English for some reason, someone else might subsidize it. But my guess is that a very small fraction (1%?) would need the full treatment in any case. With respect, I think worrying about fonts or different types of clauses is missing the point, though I am myself a pedant. I very much like Paul Halmos’ essay Think It Gooder <www.springerlink.com/index/K6234P56771J0075.pdf> on this topic. Benoit’s comment makes sense to me, and I think we will soon see a two-tier system where only the premium outlets (e.g. top general journals/magazines) cost much to publish in, and the rest is essentially arXiv overlay. 7. There are publishers who do a good job at copyediting: the academic ones. I have notably good experiences with the LMS and the AMS. All commercial publishers where far below the quality of all academic publisher with this respect for my articles, as far as I remember. This at least shows that we could pay significantly less for better service in the subscription system if Topology’s example was followed en masse. • CommentRowNumber16. • CommentAuthorScott Morrison • CommentTimeJul 21st 2012 My understanding of MSP’s costs (from conversations with Rob Kirby) is that$20/page is a strong lower bound relative to the quality they provide (which in my experience is quite high).

Personally, I’d echo Noah — I don’t think that high quality copy-editing at $20-$50 per page is good use of “our” money. On the other hand I’m very happy to accommodate the widespread desire for this while thinking about improving publishing; the scandal is spending money on shockingly bad copy-editing that actually damages the papers.

• CommentRowNumber17.
• CommentAuthorDmitri Pavlov
• CommentTimeJul 23rd 2012

Copy-editing could be something done by universities. Just as before TeX days mathematical departments employed typists to type faculty’s papers, why can’t they also employ copy-editors to edit them? After all, universities (unlike “publishers”) have a direct interest in improving the quality of faculty’s papers.

• CommentRowNumber18.
• CommentAuthorDavidRoberts
• CommentTimeJul 24th 2012

And then the university can find the price/value balance that they are happy with, rather than having a) cheapest possible foreign outsourced copyediting or b) PhD-holding in-house publisher’s employees who are paid comensurate with their qualifications.

• CommentRowNumber19.
• CommentAuthorzskoda
• CommentTimeAug 8th 2012
• (edited Aug 8th 2012)

17: good idea;

18: I do not see why university would not outsource in some cases as well. I do not understand word foreign, do you think that this discussion is relevant only to the universities in primary English speaking countries ?

I should point out that the commercial publishers fuss a lot about copy editing costs, spend order of 1 year to accept a paper and then they give you the order of 2-4 (e.g. TWO) days or so for “proofs” )what makes large costs while on vacation etc. In proofs phase all kinds of minor improvements can be made, and in lack of time (imagine you typically teach and have other scheduled duties, even conference when receiving proofs version) we clearly do much suboptimal and many true errors can slip as one can not do really tough proof reading. If the publishers really cared about content quality the proofs would have more flexibility. Their copyediting often focuses more on uniformity of the font and bibliography format, where the image titles are and alike, than on the quality of the content presentation,

• CommentRowNumber20.
• CommentAuthorDavidRoberts
• CommentTimeAug 9th 2012

Zoran, I meant foreign in the sense that Elsevier outsource to India, and others to various eastern European countries. It’s not so much that it is these specific countries, but that it is the cheapest possible in a place where the publishing language is not the primary language, and I doubt the copyeditors are necessarily experts in the subject matter.