Book talk:M. Ben-Ari/Mathematical Logic for Computer Science

The third edition (which I laid my hands on today) is very different from the second (which I already had) which in turn mentions to be "completely rewritten" WRT the first. The third edition is really good for the basic and thorough coverage of PropCalc and PredCalc and their models that we need (although there is a quirk that Ben-Ari only defines models for finitely many relation symbols and constants at once, this can easily be overcome) so I'd like to get started covering it (before proceeding to abstractions I can't even formulate here presently). But do we need to separate the editions in some way? --Lord_Farin (talk) 23:36, 14 February 2013 (UTC)

There is a precedent for multiple editions - I think the Book:Peter Barlow/New Mathematical Tables entry is the most comprehensive instance, but a more contemporary example is Book:Roland E. Larson/Calculus. Suggest this is entirely appropriate. My own copy is the first edition (1993). An excellent example of a work which between us we have all three editions and can therefore complete it. --prime mover (talk) 06:09, 15 February 2013 (UTC)

I note that the technique for doing multiple editions needs further work. As it is, the edition is included in the title, which seems suboptimal. Maybe we should add a further parameter to the template - or even build a new template for "edition" which will then be a transcluded entry into a main page (where only the publishing details would be in the onlyinclude, the main bulk of the contents in the fully expanded page). --prime mover (talk) 07:21, 15 February 2013 (UTC)

What occurs to me as the simplest solution is sort of an amalgamation of what you suggest. Namely, to add a parameter ed or edition to the BookReference, which will render e.g. Calculus (7. ed., 2001) and link to the subpage Calculus/Edition 7 which can be referred to as is currently done on the Calculus page. As an aside, I think "Further editions" should precede the contents. In the case of the Math.Tables (where multiple first authors occur) I think redirects to these subpages should resolve any issues with the template linking to the wrong page. Since the stuff is presently only relevant in a small amount of cases I think we can afford ad hoc resolution of anything not fitting this new mould. --Lord_Farin (talk) 08:51, 15 February 2013 (UTC)

That's very much the sort of solution I was thinking of. "ed" suits me. As to whether "further editions" should precede the contents - not sure I agree with that, IMO the contents themselves are of more importance and relevance to the reader than the details of the edition (maybe that's just me, I've never been particularly interested in keeping track of editions, it's just an awkward detail that needs to be kept track of sometimes).

Making sure that technique can be used when we only report on one edition (maybe we can't find out when a work was first published) needs to be taken care of, of course. --prime mover (talk) 10:22, 15 February 2013 (UTC)

Introduced the ed parameter. Haven't bothered with changing the links to the book pages yet as I have to contemplate the precise setup of the book pages in the new suite first. It is probably impossible to make this compatible with the current approach, so that we will enter sort of a transition phase. Main issue here is that presently, the year refers to the year of first publishing, while for an edition it is far more natural to give the publishing year of that particular edition. I highly suspect that there is already considerable discrepancy in this, so it won't make matters much worse - at least not to the extent of necessitating a mind-dazzling marathon to go through all source sections.

I will post a suggested approach to the edition subpages in my sandbox. --Lord_Farin (talk) 11:21, 15 February 2013 (UTC)

Techniques exist for global search-and-replace, so the mind-dazzling marathon may not be as daunting as we might otherwise expect. This technique was invoked when the "author2" parameter was introduced, so we know it works. It's documented somewhere on a talk page, it's just a matter of digging it out. --prime mover (talk) 12:40, 15 February 2013 (UTC)

An example:

1993: M. Ben-Ari: Mathematical Logic for Computer Science

Of course, if you dislike part of it, we can change stuff. --Lord_Farin (talk) 15:22, 15 February 2013 (UTC)

Works for me. --prime mover (talk) 20:28, 15 February 2013 (UTC)

't Should all be done now. Off we go. --Lord_Farin (talk) 22:02, 15 February 2013 (UTC)

Okay, it's possible that I may be in a position to start processing the flow of the 1st edition soon. --prime mover (talk) 23:29, 15 February 2013 (UTC)