[Gllug] An article for you from an Economist.com reader.

[Daniel Ahrens]

Christopher Hunter wrote:
> On Friday 18 Jun 2004 1:04 pm, Jack Bertram wrote:
> 
> 
>>Wasn't Unix commercial?
> 
> 
> Certainly was!  The first Unix system I saw was in a telephone exchange in the 
> USA, back in the '70s.  Much of the early Unix work was done by AT&T.
> 
> Chris
> 

Apologies for rudely interrupting ur party. But I don't think this is 
entirely accurate, meaning, that the implication here is that the OS was 
  originally developed as a commercial product by AT&T.  Or perhaps, if 
you didn't intend this meaning, it certainly suggests to my humble 
powers of cognition something along those lines.

Of course the beauty of the thang is that this wasn't the case at all. 
The OS was born in 1969 as a result of various research interests and 
ideas Ken Thomson and Dennis Ritchie were pursuing at Bell Labs, the 
research and development arm of AT&T and Western Electric, and a need by 
the lab for an OS which would meet their requirements as programmers. At 
this stage of it's development it wasn't a commercial project. In fact, 
there was no direct financial support even at this stage from any of the 
cos involved.

Ironically, in order to get further development funded (they wanted a 
PDP-11), the lab had to promise to develop a text formatting program, to 
be used for processing patent applications (!) at Bell.

AT&T later granted universities and commercial companies rights to 
utilise the software under a licence 'royalty free' and the licensee had 
to pay a 'service charge'.  But the point is that UNIX started out and 
continued to be a research project, rather than a commercial piece of 
software. It was even distributed in source code form.

It was only around 1979 that AT&T set out to develop a commercial 
version, which was Unix System III and which was based on System 7. This 
commercial version was first released in 1982 (distributed without 
source code). From the beg. of the early 80's onwards many other 
commercial versions by other companies (including u know who) were 
developed. So Chris, the version you saw in the 70's must have been the 
non-commercial 'licenced' version. And I guess it is fair to say that 
Unix was both commercial and non-commercial from the 80's onwards. So 
the question should have been: wasn't Unix commercial as well? or : 
weren't there commercial Unix versions available as well? :)

Dunno, for many this might be splitting hairs, but I think this 
distinction between research projects and commercial products is pretty 
important for me with regards to the whole Unix issue. Incidentally, the 
fact that there were distros available which weren't commercial led to 
institutions like Berkley and ARPA to contribute their research and code 
to it eg. possibly the most important bit of code contributed in this 
manner which everyone must have heard of is TCP/IP.  That this code in 
turn wasn't commercial led to its widespread adoption, which of course 
is an understatement.

A similar genesis and development happened with Linux (the kernel), or 
rather, Linux continued in the same tradition which was already 
established. It even started out as a research project.

It is perhaps also interesting to note that during it's long history 
Bell Labs produced a number of discoveries which weren't then directly 
pursued commercially by either of the companies involved.

Chris, can you remember which Telco this was where u saw your first Unix at?

Disclaimer: Pologies for the length of the post and if I didn't 
represent somebody's pet fact with the accuracy it deserves or omitted 
an important variation of the story, please don't flame me, I'm only 
human myself. :)

Finally, read it from someone who was actually there: 
http://cm.bell-labs.com/cm/cs/who/dmr/hist.html

regards


roger d rockstun


























-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug