🚀 go-pugleaf

>From JSOL@USC-ECLB Fri Jan 22 23:53:57 1982
Works Digest	        Saturday, 23 Jan 1982       Volume 2 : Issue 12

Today's Topics:		  Bell Labs 256K Ram
               Swap & Shop - Computer Equipment Wanted
                           Status Of WorkS?
----------------------------------------------------------------------

Date: 22 January 1982 1400-EST (Friday)
From: Bob.Colwell at CMU-10A
Subject:  256K RAM from Bell

A friend at Bell who worked on their 64K DRAMs told me in 1980 that
their 256K DRAMs were going to be ready in late 1981, so sometime 1982
sounds about right for their real release.  At the time, all that was
left to do was tweaking the myriad timing paths so that the chip
tended to reflect a constant load on the supply -- they were very
worried about noise margins and claimed that this was the major
contributor.

The 256K chip will use their redundant-rows technique that Bell
developed for their 64K DRAMs.  (In fact, this same person told me
that the 256K and larger chips were the real reason that technique was
devised in the first place).

Did Newsweek really get the scoop over all of those Electronics News
magazines?  They knew about Libya, too......

------------------------------

Date: 22 Jan 1982 1605-PST
From: Daul at OFFICE  
Subject: WANTED: Computer Equipment

 
WANTED:
   1. 3270 Emulator for DEC VT100   (supported by UNIX(tm))
   2 . VAX -> IBM Channel connection (supported by UNIX(tm))
   3. VAX -> IBM Bisync. connection, looks like a 3271
		 (supported by UNIX(tm)) 

For more information call Bob Herriot at 408 745 1300 Ext. 237.  I
will relay info to Bob if you would rather use me as a go-between.

------------------------------

Date: 22 January 1982 11:36-EST
From: Stavros M. Macrakis <MACRAK at MIT-MC>

WorkS appear to be degenerating.  There is no need for long discussion
on trivial and peripheral points:

1. Of course we'll want hardcopy.  Paper is a marvelous medium.

2. Of course we need OCR.  However, it will likely be hard to justify
OCR's for individual workstations: reliable reading of free text is
remarkably difficult and expensive.

3. Some things are serial-access, some are random-access.  Clearly the
logical interfaces to the two differ: the former is not a segment.

4. Few underlying mechanisms are strictly invisible.  VM, like any
other such, has its demands but remarkably little overhead for
core-resident applications.  Certainly, all interfaces should not be
forced into a Procrustean bed, requiring all file I/O to be through
demand paging.

Now, may we return the the discussion of workstations?

------------------------------

End of WorkS Digest
*******************
-------