Slashdot Mirror
The History of UNIX
Tucros writes "There is a nice article over at Bell-Labs.com detailing the History of UNIX." This is a somewhat lengthy bit with lots of entertaining stuff that normally would just be sorta anecdotal. I enjoyed this one a lot.
It seems to me that the biggest change in Unix in the past decade is that people are tring hard to get away from the core philosophy. One of thouse cores is "Write programs that do one thing and do it well."
/bin/mail as an interface after the first version of mailx. /bin/mail knows how to do everything it needs involving getting a message and sending mail and the "difficult problem" of properly locking mailboxes.
As McIlroy quoted "Write programs that do one thing and do it well. Write programs to work together. Write programs that handle text streams, because that is a universal interface.'"
About the only place I see this going on in modern program design is the mp3 players and that is a result of the people who do good UI work generaly don't have the skill set to do the MP3 decoding so they link to something like mpg123.
Another quote: "Cognitive engineering" is what Condon called it, "...that the black box should be simple enough such that when you form the model of what's going on in the black box, that's in fact what is going on in the black box."
Based on one of the major ideas in Unix, why does every program grow till it can read mail? I don't think I've ever seen a program that uses
I also like the bit about fixing the code so they didn't have to document the uglyness. Now that might be the best reason I've ever heard to properly document a program.
Hey, as far as I'm concerned, anything that reminds us of our roots -- computing or otherwise -- is great.
Interesting, but don't the "roots" at some point become so distant as to be irrelevant? Isn't it better to look forwards, solving the problems of today and tomorrow, than to stay in the past constantly trying to re-solve old and obsolete problems?
An "easy-to-use" OS is a fine solution to a real-world problem, but imagine the ideal: everyone is able to use a real OS because everyone has a grounding in computers and how they work. At the very least, entertaining and interesting histories such as this are a step in the right direction.
I have to disagree with this. Why would it be ideal for everyone to have the detailed grounding in computers necessary to use a complex and powerful OS? Would it be a better world if everyone spent years learning enough about car mechanics to be able to strip down and rebuild an engine? No, it would be an enormous waste of time; time that could be better spent (for most people) learning other knowledge. We need well trained car mechanics, but only a fairly limited number. Similarly, I don't think 3-5 years of intensive training in the basics of computer science would be a sensible use of most people's limited time.
Sailing over the event horizon