Grouping Out Of Order

  • Ken McKelvey (10/7/2010)


    I am glad the order difference technique worked but unfortunately I cannot claim credit for the idea. The technique was used to win a competition but I cannot remember the details or who won it. (It may have been a databasejournal.com competition with either Joe Celko or Itzik Ben-Gan as one of the judges.)

    It is an extremely elegant solution to work out when state changes – in this case who has borrowed a book. Like all good ideas, the underlying concept is simple – work out two orders so that the difference is the same while the state is the same.

    Greetings,

    You may not be the author of the idea, but you certainly have taken the lesson to heart and used the idea in new situations. The use and adaptation of a good idea speaks far louder to the original author than just saying its a good idea and then its never used again. It is also great to see such honesty in a group of people who will admit they learned their lessons from another and are just passing them on. Then again, isn't that how most kids learn in school? :hehe:

    Such ideas that show others how to view data from a different perspective and then allow the data to be processed much faster are a boon to the entire community. People may use different words to describe the problem (different states, grouping by alternate order, etc.) but the logic is still the same and the solution is beneficial to all. Some ideas, possibly even this one, can also be dated way back to mainframe SQL where they had to use every trick in the book, and make some new ones, to get all the speed they could.

    Some of those ideas are later forgotten, and then re-invented or found again. Either way, they are still useful - when used.

    Terry Steadman

Viewing post 16 (of 15 total)

You must be logged in to reply to this topic. Login to reply