Talk:The Art of Computer Programming

This is an old revision of this page, as edited by RossPatterson (talk | contribs) at 22:28, 7 June 2007 (Ambiguous re. typeset: See TeX). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Latest comment: 18 years ago by RossPatterson in topic Ambiguous re. typeset

The first paragraph mentions upcoming chapters 9 and 10 but gives them different topics then they are given in the table of contents.

maybe on the site of Dr. Knuth you can find the correct information, as this is a wiki, your are allowed to change the wrong entry.
Ppareit 19:44, 7 May 2004 (UTC)Reply

Vol 4, fasc 4.

Addison-Wesley says that Vol 4, fasc 4 was published on 2/6/06. Amazon says 2/7, B&N says 2/10, but both of them say that it isn't available yet. Bubba73 (talk), 22:10, 10 February 2006 (UTC)Reply

83 pages

I checked and 83 is now the correct number of pages for the Boolean Basics fascile. Sorry for thinking it was vandalism. Vandals do like to change one digit. Bubba73 (talk), 02:35, 18 February 2006 (UTC)Reply

Ambiguous re. typeset

'In 1976 Knuth prepared a second edition of Volume 2, requiring it to be type set again....'

No inidcation is given to the relevance of the typeset, for instance, why couldn't Knuth use the same type-set as everybody else? Some explanation should be given. — Preceding unsigned comment added by 86.129.218.204 (talk)

I'm not sure enough of this to state in the article, but I think it had to do with the technology of the time - thirty years ago. A new edition of the book had to be typeset again. An old technology was called hot type, where they poured molten lead into a form to make a plate that was used to print. Later they used photo offset which produces a low qualty product. Since the book was going to have to be typeset again, Knuth decided to work on typesetting by computer. (He thought it would take six months, it took 10 years.)
I'm not sure enough of these details to put them in the article. Perhaps someone can verify or correct them. Bubba73 (talk), 00:21, 7 October 2006 (UTC)Reply
You'll find them repeated and cited in the TeX article. RossPatterson 22:28, 7 June 2007 (UTC)Reply

New publishments

The third edition of volume I and volume II have already been published! — Preceding unsigned comment added by 205.209.103.230 (talk)

That happened years ago, and it is noted under "current editions". Bubba73 (talk), 15:15, 18 November 2006 (UTC)Reply

How many books were planned?

I read here that, "Knuth began the project, which was originally planned to be one book, in 1962". Yet on the page about Knuth it says, "...where he became a professor and began work on The Art of Computer Programming, originally planned as a seven-volume series". Can somebody resolve if it was planned as one book or a seven-volume series?

That is confusing. It was originally planned to be one book. When he started turning in the manuscript, it was clear that one book wouldn't do it, so it was planned to be seven books. Now volume 4 itself will be at least three books. Bubba73 (talk), 02:31, 8 February 2007 (UTC)Reply

Regarding the paragraph on Knuth's use of assembly language:

It is not clear to me why the author thinks that readers of these volumes would consider translating the MIXAL implementations to a high-level language. The algorithms themselves are the best source of the logic to be implemented. They are written in English so have no innate preference for the idiosyncracies of any particular HLL (or low-level language).

Beginning in 2000, I regularly taught a senior course in data structures using Volumes 1 and 3 of TAoCP. The implementations were in IBM assembler. I couldn't imagine having the students translate the MIXAL examples to IBM assembler; on the other hand, the algorithms themselves are readily coded frequently with one machine instruction for each logic statement.

Kcats 14:12, 13 March 2007 (UTC)Reply