OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

legalxml-econtracts message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Minutes Draft Face to Face, August 9, 2003

Posting on behalf of Dave.



Subject: Minutes Draft Face to Face, August 9, 2003


Here are the draft minutes of the face to face meeting. They are a bit 
rough in part because there seems to be a lot more out-of-band 
communication  in a face top face than in the conference calls. (They 
may also be rough because I was so distracted by actually meeting so 
many of you.)

In any case, please pass any corrections and/or comments my way.


Dave Marvit
Special Projects Consultant
Fujitsu Labs of America

Draft Minutes (Version 1)
OASIS LegalXML eContracts Technical Committee
August 9, 2003 Face to Face

- Discussion of IP issues as related to participation in and observation 
of the TC
- Status report on the resources page, and the decision to open all of 
the content to the public as a default.
- Discussion of IP issues related to use of contracts for TC purposes
- Jason presents an Enterprise Contract management scenario.
- Discussion of clause model – options and some implications
- Dave presents Automated Negotiation scenario and technology demo.

12:40 Brought into session

Dr. Leff
Dave Marvit
Jim Keane
Jason Harrop
Dan Greenwood
John Trudenick (sp?)
Jane Winn
Dave Weitzel (sp?)
Rolly Chambers
John Messing
Jamie Clark

At 12:40 the meeting was brought to order.

It was noted that there were some people present in the capacity of 
observers. Their presence formally appreciated.

This did present an opportunity to clarify some issues associated with 
IP. Dan invited Jamie to provide a brief description of some of the 
issues associated with IP and OASIS, especially in the immediate context 
of this meeting.

Jamie’s comments are summarized as follows:
1. The ABA has an IP clause.  It is assumed that the ABA will own and 
use any IP disclosed ‘benignly’.
2. OASIS has an IPR policy. People are expected to declare their claims.
3. The legalXML TCs have a special condition. They can say ‘we have 
decided not to include in our work product anything that is of a 
proprietary nature’.

Dr. Leff: Moves to approve all prior minutes.
DanG. Seconds
Vote -- Unanimously in favor with 1 abstention.

DG: Dr. Leff, can you update us on the resources page?

Dr. Leff. We have been using the Kavi page. I have our minutes and 
agendae for upcoming meetings.

Jane: Where is it?

Dr. Leff: They are locked to members only right now. We need to decide f 
they should be public. ‘www.oasis-open.org’ will get you there.
www.oasis-open.org--> home page  left side menu  ‘legalXML-eContrcat’

DanG: Moves to make all info on the docs page to become public as a default.
John Messing: Seconds
All in favor – unanimous

Dr Leff: Rolly has submitted a benchmark contract.  I have also 
submitted one

Rolly: I submitted a design/build contract that I found on the Findlaw 
site. This is one of about a half dozen contracts that are frequently 
used in the construction industry. What I have NOT put up, although I 
hope that the group will be aware of them, is any form of contract that 
is copywrited. These are commonly used but I didn’t put up there because 
of the copyright of the contracts.

Looking at construction contracts generally, there are a number of 
aspects that would be useful to identify in the body of those 
agreements.  As with any contract there are lots of obligations. Having 
a way to track and identify those obligations would be very useful. Many 
times these are conditional. If something happens then one party or 
another has a certain obligation. Finding a way to extract these 
conditions (I call them events) is useful. In addition, finding ways to 
extract the names of the parties, their qualifications, who the contact 
people are (in terms of notice) is also valuable.

DG: I submitted (with Dr. Leff) is a class of contracts that you are 
likely to find when you click on a web site.  Here is an area of 
practice where it is already electronic. We don’t have to convince 
people to put them in electronic form because they already are. The only 
issue is getting them to mark them up in a new way.

Finally, as a matter of contract management at large companies it would 
be useful to have these ‘click through’ contracts represented 

People click on all kinds of things as part of their work process. They 
are entering into all kinds of agreements for the corporations. There is 
some logic to ignoring this, but in so far as people want to pay 
attention, it would make sense to capture this, load it onto a server, 
parse it, and decide if someone is in a position to make the commitments 
that the contract is likely to impose.

Jane: The acceptable use policy could address this.  There is an 
unstated premise that if they are given a computer and a suite of tools, 
the corp. believes that they have addressed it.

DM: That may be their best argument, but they KNOW it’s not true. 
Everyone knows that people download and use a variety of software and 
services from the net. Even just reading the New York Times requires an 

Dr. Leff. The AUP does not protect the firm. It is well established that 
apparent authority is binding.

DG: We are going to have an implementation guide, and advice about 
entering into these agreements.

[At an institution where I worked] we didn’t have an agreement with 
Adobe, but we mandated that they had Acrobat.

Jamie: I’d like to suggest that R&D orgs and academic institutions 
represent one end of the spectrum. Many corps have a different 
perspective. They are very careful to know what people are clicking on.

Also, when you talk about the Adobe license it represents a class of 
phenomena.  There is probably a class of licenses that are sufficiently 
benign that you don’t really care.

Jane: Many multinationals hated UCITA for this reason.

Dave: This is analogous to the issue of downloading executables from 
some years ago. People needed to be educated that downloading (and 
running) executable files had risks. The same education needs to take 
place in the world of contracts now.

Jamie: Yes. They are legal executables.

John Messing: I met a guy from Google interested in electronic 
contrcats, and someone from MISMO. These folks might be interesting 
contacts for us.

Rolly: If they have benchmark contracts that’d be great.

Jamie: You need to know that it is reusable. Even if you do you may 
anonymize it to make people happier.

DG: I also want to submit the OASIS membership contracts. They are 
electronic, have good IPR. The other is Tom ?’s Web technical 
development contracts. They have a good model contract.

Jamie: I’ll confirm by email, but I’m pretty sure that you can use and 
do surgery on the OASIS contract. The ABA guys might be more difficult 
to deal with.

Jim: The IP guy Vince is pretty reasonable. I’ve published with them and 
they have been very reasonable.

Jane: The collection of ad hoc examples of contracts seems not to be 
adequate. There are more obvious ways of collecting contracts. You can 
go to a law library and look. Has that work been done, and if so, where 
is it.

DG: Yes, it has been done. It is difficult to be all things to all 
people because the contracts are so varied. So, in an effort for speed… 
Jason’s clause model seems to fit, and we have been spot checking.

We have, in an effort to support vocabulary and semantics, we have 
gotten together experts and were hoping to use the intuitive and expert 
sense of these people. Then we are hoping that they will be able to vet 
these ideas.  I don’t know that you could ever think your way to the end 
product. Our though was that you could try something... something where 
people are ready to adopt. Then, through interaction with actual users 
we can zero in on an appropriate solution.

We couldn’t see the pattern in advance.

Dr. Leff: There are some things that are missing. Sample PO, lease 
order, bond debenture. One source may be simply going to court. If a 
contract has gone to court it is part of the public record. Class action 
agreements (like credit card agreements) are also public.

Jane: SEC filings

Jamie: Findlaw had a practice of collecting these

Jason: By way of very basic background, there is a whole heap of things 
in a contact that can be marked up. The decision as to what should be 
marked up comes down to what use it might be. There is structural stuff 
that allows people to extract info and there are also heaps of non 
structural stuff, commencement dates, renewal dates all the way to 
things that are controversial – that may have significance by being 
marked up. So we had people submit contracts that describe the kinds of 

Another approach is managing the contract lifecycle.  In a round-about 
way the benchmark contracts tie back because they shed light on a 
particular scenario (like a contract that has an interesting dispute 
resolution clause) or, on the structural side, a contract that has an 
interesting structure.

I was going to suggest that when people submit a benchmark contract they 
should mention why it would be of interest to the committee. In some 
cases we merely need to look at the contract. A URL pointing to it is 
sufficient. This should help us avoid any rights issues.

DG: I have a dream that, at the end of the day, we would go back to the 
benchmark contracts and try our draft spec against them. Then we need to 
decide if something is out of scope, or if our work will need to be 

Jason: The effort in actually marking up a contract is much greater than 
looking at a contract and seeing if it is mark-up-able.

John Messing: Can’t you look at a contract and just have it marked up 
automatically? For certain types of contracts…

Jason: Yes, so it comes down to conversion.

DM: Structurally, yes they can be marked up automatically, but not 
semantically. That is still much too difficult.

(The group generally agrees on this point.)

DG: It is possible that all we will achieve in this group will be the 
clause model, and that would still be a very significant advancement. 
When we are done with the clause model it should be easy to have it 
allow the overlay of other semantic systems.

John Messing: You won’t know the value of the clause model until you 
have finished it. But then are you still within the charter?

DG: Doing the clause model is within the scope of the charter. Trying to 
do more and failing is also within the scope of the charter.

DM: There is still value is having a semantic representation. Just don’t 
expect to convert pre-existing contracts automatically. [Also generally 
agreed with…]

Jason presents an Enterprise Contract management scenario.
Jason: At one end of the spectrum is a telecom company that has a 
million cell phone contracts that are all identical except for a few 
fields (name, address, plan, renewal date, etc). At the other end are 
companies that have rich and varied contracts that threy are trying to 

Today all of the contract management systems have their own formats. 
Getting info in and out is a proprietary matter (often involving 
retyping).  The question becomes, what info do we need to capture in 
order to map onto the contract management systems.

So we split things up into structural requirements, and non-structural 

Jane: What is the relationship between this and the UBL work?

Jason: Party name is used in both. The UBL guys are working on the 
business process first. Nowhere in that do they go down to the level of 
the terms of the contract.

One of the things we’d like to be able to do is search – for example 
“Show me all of the contracts under the law of China.”

DG: At an earlier meeting in the cyberlaw committee we presented and got 
some feedback. The consensus on the ‘why capture this’ – they felt that 
the contract management phase was the most relevant – the biggest 
benefit now [of having an XML representation for contracts]. The 
management of changes are also important.

Jason: You will have people tell you that we have contracts that live 
for 20 years, and the people who made it are long gone, as are the 
people who made the changes. If you can capture the changes and the 
reason the changes were made you can save your company a lot of money.

[Digression into whether or not the information Jason is capturing would 
potentially be evidentiary.]

[Some discussion about whether or not it is required that there be at 
least 2 parties to the contract, and that they be represented in the 

There are two things we might be trying to do with the info we capture. 
One is to do the same kinds of things we currently do (name, expiry, 
governing law etc..) I call that a current generation contract 
management system.

It is also possible that you might parameterize the entire contract. 
Then you might be able to say to the system: if a party does X, what 
happens? There is that promise of having a complete model of the contract.

John Messing: Is everything structural or non-structural?

Dan G: Yes.

Dave: I stay awake at night worrying about this. I guess that if we find 
a counter-example we can add a rule. You can always add a rule to deal 
with any counter-example.

JimK: We are groping now, but there are those who argue that we may miss 
entire classes of things by thinking about it this way.

Dan G: Yes.  Part of our process is to vet our work publicly.
Can we shift the schedule… do the clause model review first (before the 
[group agrees]

Zoran: joins the meeting by phone

DG: We are going to do a quick review of where we are with the 
structural model status…

Jason: When you look at clauses there are a bunch of structures that you 
see. We need to be able to represent them.

One structure is where you have a heading , and then text underneath..

[A great deal of work at the whiteboard transpires that is difficult to 
represent in the minutes.]

Dr. Leff: The issue of 7.1 being dependant upon 7 is a type of 
constraint. We simply need a way of representing that constraint.

Jim Keane: In reality lawyers will use all of those representations. It 
is a many to many relationship and you are trying to make it an 
either/or relationship. If you chose to use a recursive model what 
happens to lawyers who are non-recursive.

Jason: Lawyers will never look at this [the underlying XML structure]. 
They will look at some WYSIWYG interface based upon the indentation and 
so on.

John McClure joins (by Phone)

J Messing: What is the responsibility of the drafter? I don’t want to be 
distracted by the system, worrying about structure. I want the ideas to 
flow directly into the document.

Jane: This is a point on a spectrum. The point just past you is an oral 
contract. There is nothing in the law that says contracts need to be 
written down.  We are addressing the contracts where we have agreed that 
there is a structure. You are saying it [providing structural 
information] is a distraction. Yes, that’s true. But if that’s a 
problem, use oral agreements. The point is that most of us have agreed 
to suffer with the distractions of form.

Dave: Dealing with any form you have to deal with some distractions 
until they are internalized.

Dan G: Dr. Leff, do you mind allowing Dave to present before you? We are 
running low on time and his demo could benefit from the live presentation.

Dr. Leff: No problem.

Dave M presents Fujitsu’s work on automated negotiation. (Discussion and 
technology demonstration)
The key requirement it imposes on our specification is that contracts 
can be generated from parameterized data. The scenarios discussed, 
Jason’s enterprise contract management scenario for example, merely 
require that parameterized data can be extracted FROM the contract. 
Automated negotiation – at least as envisioned in Dave’s scenario – 
demands that paramaterized data can build the contract.

Some discussion about automated negotiation…

Meeting Adjourned.

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]