December 1996 / Inbox

New Software: Dead or Alive?	Mark Schlack's statement in "Wanted: New Software" (October editorial) that "The search for the killer desktop app is dead" is parochial and wrong.
Which Bottleneck?	First I read Tom Halfhill's wide-ranging feature "Break the Bandwidth Barrier" (September cover story).
Fat Prospects	"Break the Bandwidth Barrier" offered the best explanation I have seen yet of the real-world prospects for ultrahigh-speed communications.
Shrink-Wrapped Software	Regarding "Wanted: New Software," all my recent applications provide everything including the kitchen sink, take up megabytes of disk space, and work slower than the previous versions.
Adventures i n JavaScript	"JavaScript Adventures" (August) offered a refreshingly real-world approach, but it contained some minor discrepancies, and I found it a bit too critical of Java-Script.
Praiseworthy Distortion	In "Beyond Benchmarking" (August) you state that one of the reasons SPEC92 gave distorted results was that the whole of a program could sometimes fit into the CPU's primary cache.
Death Spiral	"Push Me, Pull You" (September Web Project) is a pearl.
Visual FoxPro Revisited	Obviously BYTE and NSTL cannot be blamed for Microsoft's bizarre "marketing" of Visual FoxPro (September Inbox), but you should have been more careful when stating that VFP "is not on the same level as the products we e valuated" when it comes to building client/server front ends.
No-Mix MMX	To the impressive technical detail Tom Halfhill presented in reply to John Michael Williams' letter about MMX programming (September Inbox), I would like to add one point: Using the Empty MMX State (EMMS) instruction costs 100 cycles.
Fixes	The features table on page 129 of "Running on NT" (October) contained a typo.

      article     sidebar     table     illustration     photo     screen shot