Good books in computer science?

[Phil Runciman]

FWIW I actually dislike this book! Gasp...

Much of the material is excellent but IBM got into the huge mess with the 360. Brooks observed failure from the inside and IMHO did a great job of it.

Project managers can never rescue stuffed concepts especially if a lot of money has been spent! Such projects have momentum and roll over anyone who gets in the way.  

Brilliant architects are worth their weight in gold. I believe that ICL's VME/B OS began as a skunk works project.* It had such an architect. The latter was the official OS and was pretty good too. I think Warboys took over later once VME/B became official... if anyone out there knows better then please let us know and correct Wikipedia too. The Wikipedia item on VME is too sanitised for my taste. The "truth" is generally far more interesting.

If the software you are developing is going to be used by many people then remaining sharp and on top of your game is so important. Do not program if you are tired or you will spend your life debugging. ;-) I stop coding at 3pm for this reason. I come right again around 10pm!

Yes, despite the above, do read the book, but remember that among the content is a cautionary tale! 

Ooops, the above is a bit away from Python. ;-) 


Phil


*I was told this by the leader an ICL research team, no less than Alan Sutcliffe himself... many years ago now. (c. May/June 1970)


-----Original Message-----
From: Roy Smith [mailto:roy at panix.com] 
Sent: Sunday, 14 June 2009 2:21 p.m.
Subject: Re: Good books in computer science?

In article <mailman.1534.1244926333.8015.python-list at python.org>,
 "Rhodri James" <rhodri at wildebst.demon.co.uk> wrote:

> The Mythical Man-Month (Brooks) is a must.

What's amazing about this book is just how relevant it is today, 35 years 
after it was written.  Some of the technical details have changed (how many 
of us still keep our project notes on microfiche?), but cross out 
"microfiche" and write in "wiki" and what he's saying is just as valid 
today.  It's not about computer science.  It's not really even about 
software engineering.  It's more about general project management than 
anything else.

In the same vein, Death March, by Ed Yourdon.