Okay, we have decided on The Mythical Man Month by Frederick Brooks as the book we shall read for our next meeting (8pm Thursday 8 June at Far From The Madding Crowd). Please feel free to discuss what you think of the book as you go along under this post, or to join in the discussion here even if you are unable to attend the meeting.
The Oxtremists
Oxford-based tech discussion group.
One thing that has struck me reading through the first couple of chapters is that he talks about ‘debugging’ a lot and uses the term synoymously with ‘testing’. I think of debugging as fixing problems whereas testing is the activity where you show that the code is doing the right thing. One can lead to the other, but they are different. Is it just me?
I would say that they aren’t synchronised activities, but that I can’t debug successfully without testing.
I actually said “synonymously” i.e. he uses the 2 terms as if they are the same thing.
Heh, I understood, but what I meant was that I agree they are seperate processes (debug, then test), but perceive them as intertwined (I can’t do the former without the latter), so I can understand where he is coming from.
I’m most of the way through the book now. I think that the core question is the one in the “Surgical Team” chapter. Which is, we know that small “sharp” teams (I read agile here) can be much more productive than large ones. So how we do big projects effectively? If you’re developing something big like a system for national government, what do you do? A small number of really good people will obviously make a better job of it than a horde of mediocre ones, but it’ll still take them too long. I don’t know what people do when they try to scale up agile processes – I guess you try to split the thing up into several parallel projects with clearly defined interfaces?