Lambda the Ultimate

First a question — could you say more about the "Think pieces". From the CQ writing page, Think pieces – Talk about a not-strictly-technical issue that hackers would care about. For example, should the code behind scientific research be released and if so, why isn’t it? Or, how can computers be best used in kids’ education; would a somewhat theoretical piece, say a part-tutorial, part-opinion piece about the usefulness of SSA in compiler implementation, be within scope?

Second — I'd like to encourage you to seek works of less than 8,000 words long. Shorter, more quickly digestible pieces are good for academic journals (think about the early "gems" pieces in the Journal of Functional Programming), and I think they would be for yours as well.

Third, about your business model — I think pay-per-pdf is hard to get to work: are you going to demand copyright assignment from your authors? Have you considered other models, such as, say, The Hackademic Society, whose members regularly get beautifully typeset editions of the CQ, with the PDFs of individual articles being freely available online? It works for academic organisations such as the EATCS. Subscriptions are probably tax-deductible and are more likely to be covered by an employer.

Actually I think something about SSA could qualify as a plain old technical article, even if it contains partly opinion. At any rate, sounds like a good topic that should fit in somewhere.

Regarding length: yes, absolutely. I'll be writing something soon (probably on my blog) about how I've been rethinking that length requirement. We'll still publish long things (even up to and including serializing full books, if anyone wants to write them) but will also publish shorter things as well. The real measure is, how long does a piece need to be to be interesting and not superficial. (My friend came up with the tag line, "Depth first writing" which I like a lot.)

The business model is in flux. We are not planning to demand literal copyright assignment but will need fairly broad and exclusive publication rights in order to maximize our chances of being to figure out a way to make money from our articles. The authors' contract is almost done so if you're interested in contributing email me at editor@codequarterly.com and I'll send it to you when it's ready. Subscriptions are definitely planned.

It would be interesting to use technology and couple the idea of purchasing individual pdfs with various options for getting more substantial copies. For example, an option to get a selection of purchased pdfs on a dvd, printed in a simple binder, various e-book formats, etc.

I read a lot and have an enormous technical library. For example, I have 11 books on C++, 20 or so on Lisp, 20 books on compilers, etc.

I generally don't splurge for a particular title unless I think it will saite some insane learning hunger I have for a particular topic (like I spent on software configuration management monographs recently), or I really want a hardcover copy of a book. For example, I paid $150 for Quinniec's LiSP book in hardcover form, because I suspected from the reviews it would be on my shelf for the next 10-20 years. Same for Kiczales Art of Metaobject Programming. Actually, books on Lisp are by far the most expensive used books in my experience. Monographs are usually way cheaper, because libraries discard them when people don't check them out, because if they can dump them on Amazon Marketplace then they can use that to fund books that may or may not be read; it's a vicious cycle.

For trade magazines, my recommendation to get me interested in consumption would be to make it really easy for me to get all historical issues in PDF format in a single zip for a single price, and subscribe for one year forward going from that. You could use some sort of power law curve for depreciating the value of old issues over time so that old issues still made money while getting the reader interested in having the whole collection.

I have access to ACM and IEEE and still see value for CQ. The niches are different.

You pay $100/yr or whatever to the ACM, and they don't even send you the JACM...

We've discussed the ACM/IEEE before.

Way back in the day, once Lisp Pointers disappeared, I soon thereafter let my ACM membership lapse.

What about a simple advertising model?

Hard to say. As someone else pointed out, people do buy books. And I know from my own experience and from talking to other tech book authors, that modern tech book publishers often skimp on the editing which is, I suspect, why tech books are of such mixed quality. I'm hoping that CQ, by putting more effort into the editing side, will be a source of more uniformly high quality stuff. If we can pull that off, I'm hoping that people will pay for it. Anyway, it seems something worth trying.

Discussing business models for publishing is, of course, way off topic for LtU... I think the issue is rather serious, and affects the issues we do discuss here, which is why I allow it to come up from time to time. I think there may be solutions and approaches than can work (I have my own ideas on just how this can be done, of course), but talking about the specifics is not on topic for LtU.

Absolutely right. Apologies.

Your reply was totally appropriate. I didn't mean to say otherwise. Sorry about that.