hosers-talk: Fred Brooks, Turing Award Winner

["Jon Stewart" <stew1@xxxxxxxxxxx>]

So Amazon delivered _Brilliant Corners_ and _The Mythical Man Month_ (20th 
anniversary ed.) to my door yesterday and I finished TMMM around noon 
today (making it onto the very exclusive Books Jon Stewart Has Actually 
*Finished* list). 

It was a good book, a fast read. Not, perhaps, the revelatory screed 
everyone gushes about, but I'd read so many reviews about it that I knew a 
great deal about it beforehand. There were a number of very good ideas 
and, using very simple mathematics and some graphs (and very little Greek 
symbols; Brooks is my man), presented some stark evidence for why the 
software process can be so crappy.

One of the interesting things was how much detail he goes into about 
actual systems programming on crappy old IBM mainframes. Not to the extent 
that it's tedious, thankfully, but enough to convince you that writing 
software was a very different beast in the 60s and 70s. (Beyond just the 
language flavor of the month and the limitations of hardware; things were 
just _different_, in a way which I'm not sure I'll ever completely 
understand, lest, God forbid, I learn Cobol.)

It's also interesting because it goes into a fair amount of detail about 
how the federal government and the corporate world go about building large 
scale software systems. It's an experience that seems quite foreign to me 
and, I'd bet, any other college student.

Combine that with what's happened at my company the last couple days -- 
short story is we're getting sued, but I don't know that, and neither does 
anyone else, especially the president and the top breeder who are sorting 
through files downstairs at the moment to prepare a defense -- and I'm 
just struck by how surreal large personnel entities really are.

My thoughts for the evening.



Jon
-- 
Jon Stewart
stew1@xxxxxxxxxxx

"Survey says: it was a bad idea in the first place."

	-- The Dismemberment Plan