On Sat, 13 Jun 1998 07:02:18 -0400 Rich Kulawiec <[EMAIL PROTECTED]> said:
>I find the time to read the mailing lists relevant to my work on the
>Internet
I read the mailing lists relevant to my work. This includes the
list-managers list, but not the IETF mailing list, which only has remote
relevance (and huge volume). This RFC should have been announced on
list-managers and other related lists. It does stand for "Request For
Comments" right?
>Whether you endorse it or not doesn't matter. You're expected to abide
>by it.
You still live in the 80s. Here is the 90's perspective:
1. A group of people who know little about mailing lists started an IETF
working group about e-mail addresses. They did not invite or even
notify the major industry players.
2. The group made a RFC requiring a change in behaviour for the mailing
list industry, which again was not invited to comment.
3. This change is incompatible with current behaviour and 15+ years'
accepted practice. Worse, it eliminates a method whereby a human list
owner could be contacted for assistance.
4. Given #3, major industry players decide not to implement the change.
Actually, it took over a year for the industry to realize that the RFC
even existed, which gives you an idea of the level of market demand
for the functionality!
5. The RFC becomes one of many "lame duck" RFCs, which are not worth the
paper they are printed on because they are not widely implemented.
Everybody loses - the users, the IETF, the industry.
>I find your attitude incredibly arrogant.
Excuse me, but where is the arrogance exactly? Just look at the facts! I
have often seen the level of arrogance that the IETF has displayed with
this RFC, but only with sovereign legislative bodies. For instance, the
French government passed a law requiring employers to pay people for 39h
of work, but only make them work 35h (and magically find the money
somewhere, this is not the government's problem). The industry was not
invited to participate, and was outraged. The French government, however,
has the sovereign authority to pass whatever idiotic law it wants
whenever it wants, and to send people to jail if they do not follow these
laws, as long as the democratic process is respected (people only have
themselves to blame if they elect a goverment that abuses this power).
The IETF, however, has no such authority. Its goal is to foster
interoperability through the definition of vendor-neutral standards. This
means working with the relevant organizations and individuals towards the
development of standards for which "rough consensus" can be reached. This
does not mean passing a law without involving the relevant parties and
then expecting them to abide just because it says IETF at the top.
Call me arrogant if you must, but I do not think the people involved in
the discussion that led to the formulation of this standard have the
necessary experience with running mailing lists to be qualified to make
this decision, on a purely technical basis and setting aside all
political considerations. I am confident that if they had, their decision
would have been different. I did not oppose the decision because I was
not aware of it until a few days ago, and of course it has been over a
year now and it is too late.
>How dare you willfully distribute a product which you *know* violates
>one of the most relevant RFCs?
Simply because I know better, on purely technical grounds. We also have a
higher degree of commitment to the users than the people in the working
group do. If we make the product harder to use, we will lose money. As a
rule, we do not think that mailing list software development should go in
the direction of making it harder to use. Removing the human contact
address would be preposterous, so we will not do it.
>But this kind of deliberately disruptive response is a huge disservice
>to the Internet community and makes you look like children stamping your
>little feet in petulant anger.
I am sorry, but it is the working group who made a disservice to the
community, in two ways. First, by passing a RFC that requires the removal
of a de facto standard method for getting help from a human, in an age
where things need to get easier to use and fast! This cannot possibly be
a service to the community. Second, by putting the major industry players
in a situation where they have no options but to demonstrate the flaws of
the IETF process. There will be public accusations such as yours, and
public responses. There will probably have to be a press release in which
we explain what happened, that we were not involved or even informed, why
this proposal is obviously not in the users' interest, and that in over a
year there was not a single request for any of our customers to implement
this change, ie the market does not want this. The IETF will look
incompetent, childish, arrogant and useless. In the real world, no
standardization body would even consider the possibility of discussing a
standard without involving the relevant parties, it would simply be
counter-productive to the point of being preposterous.
Our difference in point of view stems from a simple fundamental
disagreement. You think the IETF is sovereign and has a mandate to set
standards for anything related to the Internet, and the industry must
follow blindly whether it makes sense or not. I think the industry
defines the standards (at least for new developments, I don't think
anyone would disagree that TCP is defined by the IETF), and the IETF's
role is to act as some kind of neutral ground or mediator and develop
standards that the industry agrees with or is at least likely to
implement. I think Microsoft and Netscape control HTML, and the IETF's
job is to make them talk to each other and work together. I don't think
the IETF can set up a group of people who have successfully created a
home page, mandate a change that just about any technically qualified
people will agree is only going to make the system harder to use, not
involve MS or NS or even notify them, and expect that MS or NS will
gladly shoot themselves in the foot and introduce this change. For every
person who thinks the IETF is sovereign and people just have to shut up
and obey, there are hundreds who believe like me, and they are the ones
who send checks to the industry players. Companies generally have a
higher degree of commitment to their customers than to organizations like
the IETF, and this is normal. Many spam filters are formally in violation
of the relevant RFCs, but people implement them anyway because this is
what their customers want and they solve a real problem.
Anyway I don't have the time to start a flame war about this. The bottom
line is that we will not change LISTSERV, and we will take whatever steps
may be necessary to fight any bad mouthing campaign launched against us
as a result of this RFC. This is hardly the first time this sort of
nonsense happens, back in the 80s there was even a consensus that
LISTSERV violated RFCs and should be terminated because it is illegal, or
at least modified to emulate a sendmail alias. It's been several years I
haven't heard a single complaint in that sense, I think the last one was
in 92 or so. Users are simply not interested in these issues, their
buying criteria are totally unrelated, and nowadays ease of use always
comes in the top three. Anything that decreases ease of use is guaranteed
to be ignored by the industry.
Eric
