From:	LUKE::GEUSER        5-JUL-1989 14:02
To:	ARISIA::EVERHART
Subj:	Wermit

Glenn,
   I have placed the source Wermit files in DRC1:[KER].  The implementation
for sliding windows is based on Jan A. van der Eyk's PC version.  In 
addition to sliding windows, I have tried to fix many of the VMS specific
bugs.  The DIR command works 99% of the time but it may get hung in a 
qio wait in the zxcmd procedure of ckfio.c.  I am still tracking this.

   I have been testing Wermit on and off between VMS and ULTRIX Vaxes
for the past two months and it appears to work as expected.
Also I have used telephone line impairment test equipment to check timing
statistics over a range of time delays and SNRs.  My interest in this
is because at OKIDATA in Mt. Laurel, we sent 1000s of blocks of data to 
OKI in Japan.  Regardless of the modem speed (1200 - 9600), the effective
baud rate with Kermit never exceeded 450 bps due to satellite delays.  
Using Wermit, the effective baud rate is always close to the modem speed
and this is a great savings to me in time and Ma Bell bills.  When I 
bring Wermit up, I usually "set window 10" (or put it in kermc.ini) and 
when I bring up the other Wermit, it is smart enough to figure out the 
window protocol.  The maximum window size is 30 but I believe 10 is 
pretty efficient (and easy to remember).
   I hope others who are in a similar situation of long distance file
transfers will be able to use this.  I caution you, I have only tested
with Ultrix but tried to put in the necessary changes for other UNIX 
flavors.  I would appreciate it if you would let me know it anyone 
else is interested in trying it and what bugs they are finding.
Thank you.
                                
                                 Colette Yanosov
                                 (ex. RCAer - Camden
                                  and current consultant)
                                 Okidata Corp.
                                 (609) 235-2600 ex. 7368