Listing affiliations (Re: OPEN ISSUE: WG Chair Selection (in general))

[John C Klensin]

--On Wednesday, 28 May, 2003 08:06 -0400 Margaret Wasserman 
<mrw at windriver.com> wrote:

> At 08:16 AM 5/28/2003 +0200, Harald Tveit Alvestrand wrote:
>> On the other hand, including the corporate information
>> *everywhere* the  names get listed seems like overkill, and
>> against our principle of  considering the
>> person-as-an-individual (as well as increasing overhead).
>> Where would you suggest we list them?
>
> Perhaps it would make sense to list them on the IETF website
> where you list the names of the current IESG members:
>
> http://www.ietf.org/iesg.html#members
>
> I don't think that anyone is actively trying to hide their
> affiliation, but it can be hard to figure them out...  I
> know that people do work in the IETF as individuals, but
> all individuals' perspectives _are_ affected by their
> background/experience and self-interests.

Note that "current organizational affiliation" tells one very 
little about background and experience, and may not tell much 
about self-interest.    Let me give a personal example... In 
what might be thought of as different lives, I did considerable 
work in the penetration and influence of mass media that 
originated in one country into a different country or set of 
countries.  Separately, I did considerable work on nutritional 
databases and international interchange of food composition data 
and on some specific types of architectural image databases. 
Now, none of that "background and experience" would show up if 
you asked for my current organizational affiliation, or even for 
my last few organizational affiliations.  But it probably had 
more influence on my sensitivities, and some of the positions I 
took, while on the IESG and IAB than any of those organizational 
affiliations: I learned a lot about the importance of low-tech, 
low-bandwidth, intermittent connections and that it was 
undesirable to have only protocols that operate well in 
resource-rich environments.  I learned to think of the Internet 
as a  two-way and multiway communications medium, not just a 
mechanism for moving bits.  I learned a fair amount about 
internationalization and character sets in practice, not just in 
laboratory theory.  Is knowing that background important?  Only, 
IMO, if people are interested in theorizing about how I'm going 
to react to something rather than just watching what I do.  But 
that same comment applies to current organizational affiliations 
alone.

If we need to go down this path, then we perhaps ought to bite 
the bullet, require conflict of interest disclosures from IESG 
members (and maybe IAB members and ISOC Board members) and post 
those on web sites.  Company affiliations alone are interesting 
(and shouldn't be a secret), but often don't provide very much 
information.

I'm not convinced that it is really important except to fuel 
paranoid fantasies. Those with the most serious of those 
fantasies will always assume that the smoking gun is being 
hidden, no matter how much disclosure is required and supplied.

The reality is that, if we are worried about affiliations as a 
sign of conflicts of interest, they often don't tell us much. 
Examples:

	* Someone with an academic affiliation may have
	consulting activities, or serve on corporate boards or
	advisory committees, and may be far more influenced --in
	ways we would be concerned about-- than by anything
	having to do with the academic institution.  That is
	especially true because academic institutions are
	normally extremely hands-off on the outside positions of
	people with academic or research appointments.  But, to
	complicate things further, people who are part of IT
	departments that serve universities are, for many
	institutions, much more like their peers in corporate IT
	departments with regard to pressures and constraints
	than they are like anyone with an academic appointment
	in the same institution.
	
	* Some of the companies that support IETF work take a
	strongly "hands off" view of IETF participation by some,
	or all, of their employees who participate.  In them,
	the expectation of IETF participants is, at most, trip
	reports to keep others in the company informed as to
	what the IETF is doing that might be relevant (in the
	participant's area of interest/involvement or
	otherwise).  In other companies, and sometimes in other
	divisions or with other people in the same company,
	there is at least an attempt at fairly tight management
	of what people do and say in the IETF.  To untangle
	this, one would need to know, not only "which company",
	but a good deal about internal company policies,
	divisional structures, and, sometimes, individual
	employment agreements.
	
	* Some of the companies who support IETF work are fairly
	large and complex.   As a trivial and obvious example,
	we usually think of Cisco as a "router company" with
	certain interests, but I imagine there are people who
	work for Cisco who go for very long periods in their day
	jobs without ever having a conversation about routing or
	talking with anyone who actually works on a router.
	And some companies have research departments that are
	deliberately isolated from product groups and vice versa
	-- a product group seeking to influence a researcher's
	IETF actions would have a lot of trouble finding the
	researcher and would be violating corporate policies by
	trying to exert any influence.  Other companies have the
	split, but the relationships are considerably different.
	And these things sometimes change over time --
	two-year-old information about company structure and
	relationships may be obsolete and misleading.
	
	* Whether as a primary job or a secondary one, if
	someone is carrying on a consulting business, it may be
	important to know who the clients are, and exactly what
	they think they are paying for, in order to judge
	conflicts of interest.  Many consultants are far more
	subject to pressures from important paying customers
	than the typical large-company employee is to
	internal-to-the-company pressures, if only because of
	corporate inertia.

A comprehensive conflict of interest statement requirement might 
tease these distinctions out in enough detail to be useful.  It 
might also reduce the pool of people willing to serve on the 
IESG: Preparation of such statements can be onerous and 
unpleasant, even (especially when someone has nothing to hide 
but wants to comply with the spirit of things and disclose all 
entanglements.   Worse, many companies (and consulting clients) 
insist that some or all of departmental structures, employment 
agreements, reporting relationships, projects and even areas of 
work are valuable corporate trade secrets.  If those rules 
conflict with IESG service, it is likely that at least some 
otherwise-qualified and reasonable people will decline to serve 
on, or be considered for, the IESG.  We might even see 
resignations if IESG members chose to take jobs with those 
companies.   That is a fairly high price for the IETF to pay, 
given the comments/ complaints we have had already sbout the 
small size of the credible volunteer pool.

So, sure, let's be sure we get organizational affiliations and 
list them on a web page somewhere.  It can't hurt, and may help 
us notice if "too much" of the leadership is shifting into a 
single business sector or single company... which could be 
issues in and of themselves.   But let's try to be very clear 
that there is much less information in such listings than some 
people might like to read into them.

      john