Subject: Re: Returns to service professionals (was Re: New ESR paper: The Magic Cauldron)
From: craig@jcb-sc.com
Date: 6 Jul 1999 01:29:06 -0000

>Well, part of my job as a publisher and all around loudmouth is
>to say "there's danger over there" and so I prefer not to let
>them "choose that path at their own peril" but instead try to
>warn them about it, and as a result, to get them to add something
>back for their own long term good.

Just wanted to add my own note of appreciation for your recent posts.

One small, but perhaps-effective, way we can encourage the "big players"
to open up their up-and-coming infrastructures (at least at the external,
or "interface", level) is to consistently ask them, and encourage others
(such as reporters) to ask them, questions like:

  If your new system to do XYZZY is as important to the future of this
  industry as you say, why isn't it an open protocol?

  As long as your XYZZY system isn't published as an open protocol, why
  should the industry view it as being worthy of long-term investments?

Wholesale replacement of existing versions of these sorts of questions
("How do you expect to succeed competing against Microsoft?"  "If your
product doesn't interoperate perfectly with Microsoft's evolving
proprietary standards, why would anyone buy it?") is probably impossible.

But the more people -- especially *leaders* -- who start asking the *new*
questions, the better.

The idea is to get more "major players" to consider "protocol not
openly published" as more and more of a non-starter for anything new
they deploy.

(There are plenty of other angles to this, as well, such as "How can
anyone who interoperates with you via your new protocol trust your
organization if you won't publish that protocol, so they can see for
themselves what sorts of information is being exchanged?", which are
gaining a bit of credibility these days as well.)

The ORA-documenting-qmail example illustrates this at least somewhat:
qmail is not as "successful" as sendmail, and it might never be, but
*email* is still viable *because* certain classes of people -- for
various reasons -- choose qmail (and others) over sendmail *and* because
the protocols sendmail implements are, in essence, open protocols.

Replace that set of open protocols with closed ones (such as closed
file formats) and replace sendmail with the MS app annointed to implement
those protocols for everyone (total world domination, y'see ;-), and
suddenly there is *little* likelihood of alternative, robust, lean,
whatever alternatives like qmail.

That, in turn, leaves entire classes of users out of the equation,
which further encourages them to come up with an entirely *different*
protocol, which, these days, is more likely to be open.

Once that new (open?) protocol becomes a legit concern, interoperability
becomes paramount.  But users of the new protocol have already given
that up to various degrees, so it's users of the *proprietary* protocol
that suddenly start needing interoperability.  But the One True App
they rely upon for that *job* is likely to only get *less* reliable
when (and if) support for the new protocol is added.

Anyway, the result is an industry with far more problems on its hands
than if it had insisted on an open protocol in the first place.

        tq vm, (burley)