ive been reading up on LISP machines a little but cant seem to find a specific anylysis of why they didint become popular.<p>is it because people are turned off by LISP?<p>i mean the machines seemed to have decent interfaces so it didnt have to be for only AI but could be general purpose machines and a lot of stuff that we now use was created on them:<p>garbage collection, laser printing, windowing systems, computer mice, high-resolution bit-mapped graphics, computer graphic rendering, networking innovations and protocols like CHAOSNet — were commercially pioneered on Lisp machines<p>anyone using open Genera? it seems from the wikipedia article it is still sold.
Also, Symbolics, LMI and TI all faced the big problem that their machines were in the $100K range, which was too much except for well-funded ARPA projects or big-company research groups. (E.g., I was one of the original system hackers at the Schlumberger/Fairchild AI Lab in 1980, and, with Schlumberger's deep pockets, we bought several Symbolics machines.)<p>The newly-appearing general-purpose workstations (more like $10-15K) from Sun and Apollo ate their lunch, since you "can't bet against the desktop." I.e., no one ultimately wants to share hardware (which is also what killed the DEC VAX and special-purpose minisupercomputers like the Multiflow VLIW machines we built in the early 80's).
Here's a good overview from one of the Symbolics folks, DLW (Dan Weinreb): <a href="http://dlweinreb.wordpress.com/category/symbolics/" rel="nofollow">http://dlweinreb.wordpress.com/category/symbolics/</a> .
Because commercial LISP organizations spent too much time feuding amongst themselves and not enough time competing in the market, is the simple answer. Many commercial Unix organizations went the same way.