Re: [xml-dev] Caught napping!

[John Cowan <jcowan@reutershealth.com>]

W. E. Perry wrote:

 > It has taken principal ownership of data--some
 > of which comes from you and some of which may not--and takes principal
 > responsibility for how that data might be manipulated in light of
 > what my process, by its autonomous standards, regards as a useful
 > outcome.

How can a process take ownership of data that it was not responsible
for generating, in the absence of an agreement (public or private)
between producer and consumer?  This is the point I just don't
grasp.

Here's a message for you:  &*JFKLWEFMnasksdf 2j213rks vfjrekwndjf.
What does it mean to you?  What can you, absent any private agreement
between us, possibly do with it?  This is a theoretical question,
so "None of your business" is not an answer.

 > I was an interested party to your original data,

Then you need an agreement with me about the semantics of the
data.  Otherwise, I am entirely free to change it at any time,
and there is no way for you to catch up.

 > The
 > semantics which you expect are bound to the whole of the data as
 > you understand it

I either don't understand this, or I deny it.

 > In a
 > financial arbitrage I can buy securities from you at slightly more
 > than the going rate in your market, because I know, as you do not,
 > that I can sell them simultaneously elsewhere at a significantly
 > higher price.

Under your definitions of what principals have to share (i.e
nothing but syntax) you cannot buy securities from me.  If you send
me a message "Offer to sell 50 shares IBM at 32.5", and you later claim
that it meant (according to your understanding of the vocabulary)
"Request to buy 100 shares T at 42.7" a court will not listen
to you, and will very likely order you to make me whole.
Even if you send the message "Jwjvjes 100 jsjvjd T wjvlsk 42.7"
you will likewise get nowhere in court, absent proof of a private
agreement to employ Slobbovian.

 > Here you are again supposing how my process operates. The context in
 > which I use the textual content "John Cowan" which I found in your
 > data instance may be one in which the appearance of a matching
 > string is by itself significant, regardless of the 'role' (patient,
 > insurance company, document format creator) which got that
 > matching string into that context in the first place.

That may be so, but your example below refers directly to roles:

 > I have often
 > designed basic tests for fraud which are *very* interested when a
 > match turns up in a role where it should not have appeared (payer is
 > also payee, agent is also principal, etc.).

But you cannot tell one role from another in documents that
I issue.  Is VJWEJF the payer or the agent?  Is MMNFNWJE the principal
or the format designer? Perhaps the payer is sometimes UVJVJKWE,
sometimes VMWMEF, and sometimes HJBHWEJFF, and perhaps in turn
the payee is sometimes VMWMEF and sometimes NTKJEKWJ, depending on
random shifts of the cosmos.  (Note the collision.)  Since you are not
party to any metadata agreement with me, *you just can't tell*.  All you
can say is that "John Cowan" appears as character data.

 > The very different assumption of the principal transaction mechanism
 > is that I can do something outside the scope of agent processes
 > which may be available to you by virtue of a priori agreements that
 > you have accepted. I do not offer myself for hire to perform a service
 > as your agent, but instead offer the product which is the outcome
 > of my processing. You do not supervise nor employ me, instruct me
 > to create a process, nor dictate how I should proceed, for I am not
 > your agent. I buy data from you.

Fair enough, but unless I tell you what the semantics of the
elements are, you have no more hope of understanding it than
you do of understanding the Slobbovian message above.
If any element and attribute could represent anything, you have
neither theoretical nor practical basis for proceeding.

 > After I have bought data from you a number of times, I know from
 > experience what to expect of it.

Not at all, because I change the model every time I send the data.
This does not impact people (well, in practice it does, due to
development hysteresis) who have agreements with me, because I
update the agreements too.  But you have no such agreements
with me.

 > I will know if you change its model,

How?  You can't understand the model in the first place.
How does having a stream of transactions of the form

<b8u2j kwke="SBC" fwefv="5.2"/>
<JEKF kwke="SBC" RWfefs="12.2"/>
<b8u2j kwke="T" RWfefs="22.24"/>
<BJWJE kwke="MCI" fwefv="6.6"/>

tell you anything at all?  You might guess that these are stock
transactions about telecom companies, but in fact they are telephone
call records, and if you interpret them as stock transactions, you
are hosed; but since the data you are buying from me comes with
no guarantees, it could indeed be stock transactions one day
and telephone records the next.

 > and if you do I will know how such change affects my particular
 > use of it

How?  Without a high-level agreement, you don't know what the
change is about.  Indeed, perhaps I continue to use the same
attribute and element names and only change what I mean by them.
Previously "busjw" meant "payee" and "rj2k3" meant "payer",
but now I have swapped these.  You can't detect any change at all.

 > and can make a precisely informed decision whether it is
 > worth continuing to buy. You, in turn, are free to change the model
 > in response to your own needs or to respond to someone who is a better
 > customer for your product than I. In either case you are not
 > obliged to secure my agreement before making a change.

In the same sense that I am not required to secure your agreement before
cutting you off altogether, yes.  But I may be mulcted in damages
if we have a contract.

 > Clearly my answer is yes, there are significant new choices. To break
 > free of the schema, the model, and other a priori expectations and to
 > manage XML on its own terms--instance first--

To break free of higher-level interpretative material is to replace
meaningful utterances with uninterpreted gibberish.

 > is the revolution
 > necessary for supplanting the agency model of transaction
 > processing with the principal mechanism.

Not at all.  The principal mechanism is a very reasonable way of
doing business, but it has nothing to do with disinterpretation.
If your language is English, you do not want to buy data from
me -- far less rely on it -- that may at random be written in
Greek or Gujarati.  *You* want to bind *me* to use a consistent
data model which I am required to explain to you in detail
sufficient for your purposes.

I apologize for seeming to beat the dead equid, but all this is
derived from your original claim that it's possible to make
sense of other people's documents when all you know about them
is that they are well-formed XML.  That claim is just silly,
and since you are not a silly person, I conclude that you
do not mean what you appear to be saying.  I would therefore
like to know what, in fact, you do mean.

-- 
Not to perambulate             || John Cowan <jcowan@reutershealth.com>
    the corridors               || http://www.reutershealth.com
during the hours of repose     || http://www.ccil.org/~cowan
    in the boots of ascension.  \\ Sign in Austrian ski-resort hotel