>Subject: U Illinois Chicago computer center replies to Thomas
>
>Please keep  in the  back of your  mind the fact  that Eric  Thomas (who
>wrote  the  excruciatingly  detailed   technical  story  that  that  Dr.
>Bledstein  forwarded  to ACADEMY)  has  a  personal financial  stake  in
>BITNET. He is the founder and owner of owner of L-SOFT Inc., the company
>that markets  LISTSERV software.  LISTSERV is  the package  behind UIC's
>highly successful  "H-NET" and thousands of  other worldwide "discussion
>groups". Eric does a very good  job of explaining BITNET's problems, but
>at the  same time  he is using  fear, paranoia and  his vast  network of
>contacts (including  those who run  H-NET and other LISTSERV  groups) as
>marketing tools to promote his new products.
 
Jim,
 
I do not wish  to enter into a public name  calling contest. However, for
the record,  I do take  objection to  these gratuitous accusations  on my
integrity. You  are free to  have your own opinion  of me, why  I started
LISTSERV, and  why I decided  to make it a  commercial product. But  I do
have a problem when these opinions are presented in a public statement as
though they were demonstrable facts.
 
To take  an uncontroversial (if trifling)  example, you assert that  I am
the sole owner of L-Soft. While the fact that L-Soft is privately held is
public knowledge, the list of shareholders and their respective stakes in
the  corporation has  never been  published.  Unless you  broke into  our
office  (or our  lawyers'  office),  you simply  do  not  have access  to
information that would allow you to state  that I am or am not sole owner
of L-Soft. Not that a partial  ownership changes anything to your opening
remarks; I am merely pointing out that  you took it upon yourself to make
this statement,  when in fact  I believe I  have said several  times that
there are other people behind L-Soft.
 
You also appear to think that I  have somehow used "contacts" at H-NET as
"marketing tools" to promote LISTSERV. In fact I don't even know who runs
the H-NET  list, although obviously  I know how to  find out. I  have not
contacted anyone at H-NET and I can't  see any financial reason for me to
attempt to  put pressure on a  site that has already  decided to purchase
the  products  I am  recommending,  and  that  is the  largest  worldwide
LISTSERV user. Why on earth would I want to hit one of L-Soft's very best
customers? I did  not know that H-NET had forwarded  my message to others
at UIC and I am sorry if this has caused you any trouble.
 
All in all I think you are dismissing my technical remarks all too easily
with simple attacks on my character. My message was based on facts. I was
not trying  to point the  finger at  UIC and I  apologize if I  gave this
impression. I  know from private discussions  that you did the  best with
the  resources you  have  at your  disposal, and  perhaps  I should  have
mentioned this  in my  public message.  The technical  problems, however,
remain. I have stated that  UGA's load will increase significantly within
the next  2 weeks or so  due to the disconnection  of the UICBIT-INTERBIT
link. I stand by this statement. In  2-3 weeks we will know whether I was
warning the network of a serious upcoming problem, or just using fear and
paranoia for  personal profit. People aren't  stupid, and as you  know it
takes a lot longer than 2 weeks to get a PO through Purchasing. If I just
made up  an imaginary problem in  the hope of scaring  people into buying
L-Soft products, and if I predicted that  the storm would hit in 2 weeks,
I would just  be making a fool  out of myself and  hurting L-Soft. People
would  wait 2  weeks, see  that nothing  happened, and  they would  never
believe  me again.  If you  won't grant  me integrity,  then at  least do
consider conceding me  a minimum of intelligence. As you  know from UIC's
own purchase decisions, people don't buy L-Soft's workstation products in
order to help the BITNET core, they buy them because of a local policy to
migrate to  unix, or  then because  they realize they  can save  money by
delivering their mail  on a cheaper system. On top  of that, universities
(which form  the bulk  of BITNET)  can't buy  anything until  this summer
unless they had already budgeted it last year when the budgets were being
made.  The only  thing  I could  get  by  scaring people  off  is a  mass
migration to Majordomo or other free  utility, which is clearly not in my
interest. In  fact, one  of the  reasons I did  not mention  this problem
before is that I did not have  a solution for the *short-term*, and I did
not want to scare people off. Above all, I did not want to say "We have a
problem here,  and the only way  to solve it  is for you to  buy L-Soft's
unix products", because,  even if it were  the case, I know  just how bad
this  would  have  sounded.  I  was confident  that  I  could  develop  a
short-term  solution,  just  as  I  developed  the  "second  generation",
distributed  INTERBIT scheme  several years  ago, but  convictions aren't
worth much  in the face of  concrete problems. Now I  have algorithms and
code, I know where I'm headed, and  I know it's something I can finish in
a week  by pushing aside all  other projects should an  emergency present
itself. Naturally, it will be a trade  off. If there were a simple, magic
solution with no drawbacks, I would have implemented it years ago. I know
that the changes will cause problems for  a few sites, that in some cases
they will be blamed  on me, and that the sites in  question may decide to
abandon LISTSERV as a result. I am  willing to take that risk because the
changes will solve the problem on a  global scale. Maybe I'm naive, but I
believe in solving  problems, not in abusing them  for commercial profit.
All I have  to do right now to  make big bucks is keep my  big mouth shut
and wait for the  storm to hit. These load problems  are caused by recent
changes that were made by others, either without consulting me or against
my recommendations.  As the edifice collapse,  all I'd have to  do is say
"Gee, what was it  I kept telling you guys, but  you wouldn't listen, now
look what  you've done". Then our  salesmen would be calling  the victims
and letting them  know how much a unix or  TCP/IP LISTSERV license costs.
And with thousands of people not getting their mail, I don't think they'd
have much trouble closing the sales.
 
The reason I  mentioned all these problems  on the list is  that they are
real, and so are the complaints about delivery delays and why problems in
the  BITNET core  are  affecting  LISTSERV subscribers  that  are not  on
BITNET. There are many examples in the archives of LSTSRV-L and LSTOWN-L.
Just as your users demanded an explanation  and you had to give them one,
L-Soft's customers wanted to know what was  going on and I had to explain
it. And I  did not make insinuations about the  character of other people
or organizations. I wouldn't do that unless I could prove what I said.
 
Again, I am working on algorithmic changes which will be available within
a few  weeks and  which will solve  the problems you  and the  other core
sites are now  facing, without requiring anyone to buy  any unix product.
If somehow there is a way to use these changes to force people to buy our
unix products  anyway, I would  be most grateful  if you would  share the
knowledge :-)
 
  Eric