Slashdot Mirror


Volume 4A of Knuth's TAOCP Finally In Print

jantangring writes "It's been 28 years since Volume 3 of Donald Knuth's The Art of Computer Programming was published. The book series is a classic work of computer science in spite of the fact that still more than half of the seven volume series is still to be finalized. In 1992 Donald Knuth retired to medieval monkness in order to finish his work. After many long years in draft, volume 4A now in print and you can get it in a boxed set if you don't mind admitting that you don't already own the first three volumes. They won't be checking if you read it."

49 of 173 comments (clear)

  1. Who else hasn't read his copy of volume three? by kriston · · Score: 4, Funny

    Who else hasn't read his copy of volume three?

    --

    Kriston

    1. Re:Who else hasn't read his copy of volume three? by Anonymous Coward · · Score: 5, Funny

      I have read his copy of volume three. The sheer walls of his retreat were quite a challenge but the rest of it was easy.

    2. Re:Who else hasn't read his copy of volume three? by arth1 · · Score: 2

      It's not a novel you have either read or not read.
      It's more like an encyclopaedia, where you read what you need or want, and quite often read some more to get the background.

      Now I wish they would sell 4A + the empty box, so I could upgrade my 1-3. But my guess is that the publishers haven't read the books, and didn't plan for upgrades.

    3. Re:Who else hasn't read his copy of volume three? by WillAdams · · Score: 4, Funny
      --
      Sphinx of black quartz, judge my vow.
    4. Re:Who else hasn't read his copy of volume three? by Xtifr · · Score: 2

      Volume 3's the good stuff! It's volume 2 I've never done more than skim. (Though I do have to admit that the analysis of balanced ternary in vol2 was fascinating, but not very practical in the real world where computers use binary.)

    5. Re:Who else hasn't read his copy of volume three? by haystor · · Score: 3, Interesting

      I came into programming from a math background. Every time I try to read his books, the programming stuff is wonderful and then he writes something about a math problem or two and I lose a week of my life.

      --
      t
    6. Re:Who else hasn't read his copy of volume three? by theaveng · · Score: 2

      I haven't. Too busy playing video games, programming VHDL or verilog, and working 70+ hour weeks. So is it worth the $200 pricetag? (Volumes 1-3 Boxset plus Volume 4, fasciles 1-5)

      No. You can download Volumes 1-3 off the net for free.
      70 hour weeks suck but hopefully only temporary.

      --
      FOX NEWS.com should be BANNED from television and internet. Have the Congress take it over and give us Truespeak.
    7. Re:Who else hasn't read his copy of volume three? by Hatta · · Score: 2

      Totally off topic. Have you checked out the 1541U-II? Great expansion device for the Commodore 64, with open source firmware written in VHDL.

      --
      Give me Classic Slashdot or give me death!
    8. Re:Who else hasn't read his copy of volume three? by Tolkien · · Score: 2

      No. You can download Volumes 1-3 off the net for free.

      Maybe, but this isn't the MPAA or RIAA you're dealing with. This man is a genius and an extraordinary teacher by all accounts. I have no doubt that many of the texts you've read about algorithms who weren't authored by him directly were still influenced by his works. By not paying for the books you're showing no gratitude for his life long works which have and will unquestionably benefit programmers including yourself (and by extension, society) for years, possibly even decades or centuries to come.

    9. Re:Who else hasn't read his copy of volume three? by larry+bagina · · Score: 2

      He should give his books away for free and make money by selling tshirts.

      --
      Do you even lift?

      These aren't the 'roids you're looking for.

    10. Re:Who else hasn't read his copy of volume three? by Mr+Z · · Score: 3, Interesting

      I've actually leaned on both volumes 2 and 3. The discussions about floating point and number systems are very useful in Volume 2. The sorting and searching networks in Volume 3 are must-have reading if you're trying to do complex sorting functions on highly parallel machines. (Think median filters, for an example, particularly in the context of sub-word SIMD.)

      I read the first couple draft fascicles for Volume 4A, and see a bunch in there I can directly apply to other work I do, such as exhaustively evaluating portions of a larger search space. Also, some of the Boolean logic properties are very interesting.

      Fun fact: It turns out that during the exact same month (March 2007?), both Knuth and I attacked essentially the same problem. We both set out to find minimal instruction sequences to implement all Boolean functions of 5 variables. I didn't find this out until well after the fact, while hunting through his website looking for a new Volume 4 fascicle. We actually had fairly similar results, but his approach was far more elegant (naturally). Also, mine was constrained to tree-like sequences and a specific target instruction set, whereas his permitted any DAG and used more generic Boolean operations. (For example, I had a "not-and" operation which does "A and not B", whereas I don't think he did.) Still, it was rather amusing to see we had both tackled the same problem at about the same time, and came up with similar overall results.

    11. Re:Who else hasn't read his copy of volume three? by martin-boundary · · Score: 2
      Most academics who write books want them to be read as widely as possible. They'd much rather have someone read a pirated copy than not read it at all. In fact, pirating is the modern 21st century equivalent of borrowing the book from a library, nothing more.

      That said, I'm not sure where Knuth is likely to stand on this, not so much because of the people reading for free, but rather because pirated scans are ugly. Knuth spent ten years on a sidetrack just to build a world class mathematical typesetting system because he wanted his (and others') papers to be beautiful, and the hardback TAOCP volumes reflect that. He might well object to people reading scans on artistic grounds.

    12. Re:Who else hasn't read his copy of volume three? by mattack2 · · Score: 2

      Most academics who write books want them to be read as widely as possible. They'd much rather have someone read a pirated copy than not read it at all. In fact, pirating is the modern 21st century equivalent of borrowing the book from a library, nothing more.

      The library paid for the copy that is being borrowed. So they are not in any way similar.

  2. Just want to point out that .... by Anonymous Coward · · Score: 2, Insightful

    I just want to point out that these books are called "The Art of Computer Programming" and not:

    The Art of Software Engineering.

    The Art of Computer Science

    or The Art of [insert some pretentious title]

    And even without and pretentious term for what it is, it is still taken quite seriously and nobody disparages it.

    You don't have to be called an engineer or scientist to be taken seriously.

    1. Re:Just want to point out that .... by Anonymous Coward · · Score: 4, Insightful

      General public's view of:

      Programmer: A fat guy behind a PC making Facebook and Google.
      Engineer: An intelligent construction worker.
      Scientist: A guy doing pointless research just so he can say he does research.

  3. It must be admitted... by Sponge+Bath · · Score: 2

    From Vol 1: "However, it must be admitted that MIX is now quite obsolete. Therefore MIX will be replaced in subsequent editions of this book by a new machine called MMIX, the 2009.

    I take it Vol 4A is still MIX with 6-bit bytes?

    1. Re:It must be admitted... by RDW · · Score: 5, Informative

      ...and until Vol.1 is updated:

      http://www-cs-faculty.stanford.edu/~uno/taocp.html

      'As I continue to write Volumes 4 and 5, I'll need to refer to topics that belong logically in Volumes 1--3 but weren't invented yet when I wrote those books. Instead of putting such material artificially into Volumes 4 or 5, I'll put it into fascicle form. The first such fascicle is in fact ready now (see above): It describes MMIX, a RISC machine that is used in Volume 4A; MMIX will also take the place of MIX in all subsequent editions of Volumes 1, 2, and 3.

      Download the 16 Feb 2004 version of Volume 1 Fascicle 1 (583KB of compressed PostScript) (this old version is however no longer being maintained)':

      http://www-cs-faculty.stanford.edu/~uno/fasc1.ps.gz

    2. Re:It must be admitted... by C0vardeAn0nim0 · · Score: 4, Informative

      there's several erratas on his website, one of those is exactly about MMIX.

      site
      fascicle 1: MMIX (compressed postscript).

      on the site he tells which parts of volume 1 are replaced by the fascicle.

      --
      What ? Me, worry ?
  4. Wow by forgot_my_nick · · Score: 2

    Am I right my assumption that there may be as many as 8 volumes of Volume 4?

    --
    Cultist of the Average Middle-Aged Ones
  5. Most used by woboyle · · Score: 2

    Volume 3 is my most used one. I pretty much devoured it when I was doing research on databases and practical optimization of sort/search algorithms. I bought the first 3 volumes pretty much when they were all first available, back in 1983 near the beginning of my software engineering career in the Silicon Valley.

    --
    Sometimes, real fast is almost as good as real-time.
  6. Re:Wow --- volume 4BXz? by Anonymous Coward · · Score: 5, Funny

    Right now, tex (written by Knuth) is at version number 3.141592.
    Following the same pattern, we may get a boxed library of programming books from Knuth without ever reaching volume 5.

  7. Finally! by wandazulu · · Score: 4, Interesting

    Given the $$$ for the boxed set, which was way more than a poor college/post college programmer could afford, I promised myself I'd get these books when volume 4 came out. Over the years I've read through and copied, a lot of times by hand, his algorithms while sitting at B&N or someplace, and I always would finish by saying "Why don't I just buy this and save me the trouble?" Then suddenly everything was on the internet, and I could refer back to my notes, and then I didn't need to look at my notes any longer, but I kept wanting to buy the books, if anything to show gratitude. Now that the 4th is out, I'm going to do it.

  8. Re:Boxed set by theaveng · · Score: 4, Insightful

    It's a new feature. You buy the book, read it, and then sell it on ebay or amazon as "like new/mint" condition to recoup ~90% of your money. --- or --- If you like the book you can keep it forever without having to fear someone will delete it from your kindle. You can even pass it on to your children! Personally I think this is an improvement.

    --
    FOX NEWS.com should be BANNED from television and internet. Have the Congress take it over and give us Truespeak.
  9. Pigs flying, hell freezing over by JamesD_UK · · Score: 5, Funny

    Donald Knuth has published a book and a date has been set for the release of Duke Nukem Forever? It's all too much.

    1. Re:Pigs flying, hell freezing over by will_die · · Score: 3, Funny

      heck there are even rumors that the next book in A song of Ice and Fire will be out this year.
      If the end of the earth is not this year then it is definatly next year.

    2. Re:Pigs flying, hell freezing over by Anne_Nonymous · · Score: 2

      >> Donald Knuth has published a book and a date has been set for the release of Duke Nukem Forever?

      I guess this pretty much proves that hexagons just aren't that much fun in Civ5.

    3. Re:Pigs flying, hell freezing over by butalearner · · Score: 3, Funny

      AND the Wheel of Time will end next year. If we find out that GNU Hurd is about to be released, I think Slashdot might spontaneously implode for lack of stalling jokes.

  10. Re:Dead Serious Question by hubie · · Score: 4, Informative

    Get off my lawn!

  11. So was Charlie Stross wrong? by JoshuaZ · · Score: 4, Interesting

    I thought that Knuth had a deal with a mysterious British intelligence agency that as long as he didn't publish volume four they would let him remain metabolically active. I hope he doesn't have some illness that made their threats moot.

    1. Re:So was Charlie Stross wrong? by RDW · · Score: 3, Funny

      It's only Volume 4A. Algorithms to generate Dho-Nha geometry curves in polynomial time aren't covered until Volume 4C, so he's safe for the moment.

  12. Re:Dead Serious Question by Anonymous Coward · · Score: 5, Informative

    The 3 first volumes of "The art of computer programming" were for many years the de-facto reference for learning about data structures and algorithms using a rigorous approach.

    The problems given at the end of each chapter are comprehensive and often very difficult, which make the series challenging and particularly interesting.

    Today there are much better textbooks if you simply want to learn about algorithms. The TAOCP series demonstrates implementations using MIX, a made-up assembly language which is quite frankly, horrible. However, this doesn't change the fact that the series was a huge contribution to the field, and still has its merits.

  13. Re:Boxed set by ron_ivi · · Score: 3, Insightful

    But more seriously - it would make far more sense to buy this as an e-book that I can search on my computer wherever I am.

    I do enough of my programming from home or from a coffee shop that it'd be rather useless to have these things stuck on my office bookshelf (except, perhaps to look a bit pretentious).

  14. Re:Wow --- volume 4BXz? by somersault · · Score: 4, Informative

    I had to check to see if you were kidding! And actually:

    ali@katamari:~$ tex --version
    TeX 3.1415926 (TeX Live 2009/Debian)

    --
    which is totally what she said
  15. Re:I have the first 3 boxed by TheRaven64 · · Score: 5, Funny

    He's a professor. I'm pretty sure he can procrastinate without the aid of email. He wrote an entire typesetting system as a procrastination exercise once! Most PhD students would envy that level of dedication to The Art Of Procrastination.

    --
    I am TheRaven on Soylent News
  16. Re:Dead Serious Question by Anonymous Coward · · Score: 3, Informative

    When you consider that the copyright date in my copies of V1 & V2 are 1968,72 and 1968,73, and 1980 for V3, it is amazing that these books are still of use.

    It is hardly surprising that MIX is a little "odd" by the standards of today - it would be like comparing Mercury autocode with C#

  17. Re:Boxed set by vlm · · Score: 2

    But more seriously - it would make far more sense to buy this as an e-book that I can search on my computer wherever I am.

    I do enough of my programming from home or from a coffee shop that it'd be rather useless to have these things stuck on my office bookshelf (except, perhaps to look a bit pretentious).

    Kindle / etc / generally seems to suffer under the load of large equations and diagrams. OK for the latest Tom Clancy or Gibbons decline and fall of the roman empire, not so hot for this application.

    If you're talking about PDFs of Knuths work to read on the laptop, those are, uh, available, from the usual sources.

    Personally I was more excited to receive the final (?) eighth collection of Knuths papers that being the "fun and games". Got it in the mail last week, barely cracked open yet. Has his "famous" MAD magazine articles.

    Now an off the wall question, do people really program in coffee shops (other than ron ivi) ? I have the impression locally that they are bar / meat market equivalent for non-drinkers, people whom want to re-enact "friends" episodes, and homeless / urban campers ala the recent one zillion "hacker public radio" series. Do you just leave your $$$$$$ laptop and confidential corporate papers on the desk when you go to the can? Does the staff tolerate your presence?

    I have worked in parks, at least a couple hours until the battery dies, its not like I'm locked in my office.

    --
    "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
  18. Re:Dead Serious Question by vlm · · Score: 2

    You can always identify people whom have never read Knuth because they get hung up on it not using the hot new language of the month. "Now with Erlang!"

    On the other hand, the people that "read" Knuth but didn't understand any of it, get even by complaining about the magnetic tape merge scenarios, and "doesn't everyone just use the qsort routine anyway?"

    The folks whom "get it" understand its not training, but education.

    --
    "Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
  19. Re:Getting this Right Away(tm) by e9th · · Score: 3, Funny

    I'm preordaining this

    Cool! Now we can call it the Reverend Volume 4A.

  20. Re:Dead Serious Question by siride · · Score: 3, Funny

    I had trouble following your post because every reading of one of your flagrant misuses of "whom" resulted in a cascade of aneurysms. These caused me to pause and foam at the mouth a little bit. I later regained consciousness and was able to continue reading. And then bam, another misuse of "whom"! I don't know if I will ever recover.

  21. Re:I bought Vol 1 in 1969. by Existential+Wombat · · Score: 2

    He needs to buck up and get a move on. He was born in 1938, at this rate he'll be long dead before he finishes...

  22. Re:Boxed set by camperdave · · Score: 2

    Hmm... How would you fold the cover of Mad magazine on an e-reader?

    --
    When our name is on the back of your car, we're behind you all the way!
  23. Re:Dead Serious Question by Bigbutt · · Score: 2

    The final "its" is what caused the stroke though.

    [John]

    --
    Shit better not happen!
  24. Re:Boxed set by chill · · Score: 4, Funny

    Hack the e-reader to install Linux.

    Install Perl.

    The rest is left as an exercise to the student.

    --
    Learning HOW to think is more important than learning WHAT to think.
  25. Re:Does he even INTEND to finish the series? by WillAdams · · Score: 2

    Yes, he does intend to finish.

    You have to understand that when he was first asked to write this ``book'' he wrote out longhand ~600 pages and submitted that as the first chapter --- when his editor received this manuscript he asked in response, ``Don, just how long is this book going to be?'' --- after a bit of back-and-forth they worked out that the first submission would be the bulk of Vol. 1 and planned out the balance of the volumes. Then the Monotype casting machines were retired and he took a bit of time off to write TeX (see his book _Digital Typography_ for the full story).

    TeX was declared finished in 1982 and since then he's caught up on bug reports for his other books &c.

    William

    --
    Sphinx of black quartz, judge my vow.
  26. Re:I bought Vol 1 in 1969. by SnarfQuest · · Score: 2

    You now have 4 data points, so can someone draw a plot to see when the next volume is coming out?

    --
    Who would win this election: Andrew Weiner vs Andrew Weiner's weiner.
  27. Re:Dead Serious Question by not-my-real-name · · Score: 2

    He's a time-travelling anti-software patent crusader from the distant future. He travelled to the dawn of the computer age to publish a set of books containing all the algorithms known throughout time to establish prior art. Unfortunately, most people don't read his books and thus we're still troubled by software patents.

    --
    un-ALTERED reproduction and dissimination of this IMPORTANT information is ENCOURAGED
  28. Re:Wow --- volume 4BXz? by Samantha+Wright · · Score: 3, Insightful

    The answer is that if that was done, then the next version, 3.14159265, would have a lower number. The whole point of the irrational number thing is that it's still technically increasing—thus the truncation.

    --
    Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
  29. Re:Wow --- volume 4BXz? by narcc · · Score: 2

    You fail math forever.

    3.1415927 is indeed greater than 3.14159265

    Consequently, the poster is correct. If the current version number were rounded instead of truncated, the next version would indeed have a lower number than the previous version. This is why the poster (correctly) pointed out that rounding would have been a mistake.

  30. Re:Dead Serious Question by gnapster · · Score: 2

    Who is Knuth? He's my homeboy.