Talk:SantaClausProblemV2

Beautiful concurrency

I am writing a chapter for a book called "Beautiful code", edited by Greg Wilson. The chapter is a tutorial about Software Transactional Memory in Haskell.

After a lot of useful feedback, I have now completed Version 2 of the paper. The Haskell code is also available.

I would welcome any comments or questions you have on the paper, or constructive suggestions for improving it; the more concrete the better. You can see what people said about version 1 on the version 1 talk page: Talk:SantaClausProblem .

The book is aimed at a general audience of programmers, not Haskell geeks, so I have tried to explain everything necessary as I go along. So if you are not a Haskell expert, your input would be particularly valuable to me.

You can email me directly (simonpj@microsoft.com), or add Wiki talk notes below.

If you give your real name somewhere in your text (or email it to me), I'll add you to the acknowledgements at the end of the chapter. Notably, I'd like to acknowledge Steven807, Tibbe, Fanf, Garious, Rycee, Brecknell, Mvanier, Gaal, Fernando, Gknauth, EngineerScotty, BoAdler.

---

Simonpj 14:26, 22 December 2006 (UTC) To add a note, begin with four tilde signs ~~~~; the Wiki will fill in your user name and date.

Brecknell 13:38, 11 January 2007 (UTC) In Figure 1, retry should be STM a, not STM ().

Sylvain 13:45, 11 January 2007 (UTC) small points: since you go through explaining that the do-notation is overloaded and works for STM as well as IO (page 8), you might want to tell the same for return on page 11 (return being given with a type a -> IO a on page 5). There is also a double "Here is the code" at the bottom of page 17.

Brecknell 13:59, 11 January 2007 (UTC) The rewrite of the paragraph at top of page 17 has left behind a fragment that I think just needs to be deleted: "just runs each of the actions in sequence".

Brecknell 14:15, 11 January 2007 (UTC) Last sentence of page 18: "Here is another way approach that..." Delete either "way" or "approach".

ArthurVanLeeuwen 14:27, 11 January 2007 (UTC) minor stuff, really:

• page 2, middle of page: syncrhonized -> synchronized
• page 5, top of page: For example, here are two Haskell functions -> For example, here are the type signatures for two Haskell functions (the functions themselves may be primitives, after all)
• page 5, second paragraph: do the parentheses around hPutStr serve more than to confuse? Especially given their absence for hEchoLine...
• page 5, footnote 3: comma after effects
• page 6: make discarding the return value of forkIO in the second example main more explicit, e.g. with a footnote
• page 7, after Atomicity: This ensured -> This ensures

Brecknell 14:32, 11 January 2007 (UTC) I like the way you build up to the definition of "choose" in this new revision. I think this helps to convey how STM and actions as first-class values can improve modularity and composability in concurrent applications. So, if you like, you can consider my lsat comment on the previous revision "done", as well.

Maeder 14:38, 11 January 2007 (UTC)

• page 11, use subtract (+ -> -) in limitedWithDraw
• page 22 (2nd line), the the -> the

Genneth 15:07, 11 January 2007 (UTC) p4: make analogy between () and void to help non-haskellers

ArthurVanLeeuwen 15:10, 11 January 2007 (UTC) The bottom of page 17 has 'Here is his code:' in duplicate.

ArthurVanLeeuwen 15:19, 11 January 2007 (UTC) The note 'The code for choose is brief, but a little mind-bending:' on page 19 is very short, for such a leap in required understanding. In just one single block of 7 lines you introduce both the concept of IO actions encapsulated inside STM actions as well as the concept of list comprehension. For a non-haskeller this is quite a lot.

Maeder 15:25, 11 January 2007 (UTC) maybe generalize nTimes on page 7 to type: Int -> (Int -> IO ()) -> IO () and use nTimes on page 16 instead of sequence_ and list comprehensions. (The function choose on page 19 is also higher order.)

Malcolm 15:27, 11 January 2007 (UTC)

• Intro, para 2, "Sadly, parallel program[s] are" - missing plural
• sec 2.1, para 3, "if two thread[s] call" - missing plural
• sec 3.2, Atomicity, "This ensured that" -> "This ensures that"