Re: "The use of Metadata in URIs" and UK law

Pat Hayes wrote:
> ...
> I suggest that it is not, and cannot possibly be, either of these. Or 
> indeed any other English paraphrase of some communication act between 
> human beings. GET is not a conversation, it is a mechanical transfer 
> protocol. We can of course speak metaphorically using this language, 
> just as we speak of machine "instructions" and software "agents" and 
> so on: our technical vocabulary is riddled with these suggestive 
> usages. But sometimes it is vitally important to remind ourselves 
> that these really are only suggestive metaphors. Computer hardware 
> does not obey as humans obey orders; software does not act as humans 
> act; and GET does not request, assert, claim or suggest in any human 
> senses of these words. It simply initiates a process which results in 
> bytes being transferred from one place to another on a network.

OK, so you suggest that if, a few nights ago, I put out an oversize
mousetrap and finished up with a little neighbourhood kid dressed
as a witch flaying around on my front lawn with a few broken
bones I'd have a good defence in court that it was only a
suggestive metaphor that the mechanical actions of the trap were
obeying my orders?  It simply initiates a process which results
in a bar of metal being transferred from one place to another on
a wooden baseboard, perhaps?

Software does not run in a legal vacuum (or a social, or, if you
like, a moral one either).  The server acts on behalf of its
administrator and the browser on behalf of its user.  Communication
between the browser and the server is a proxy for communication
between the two people.  The exact meaning of that communication
is defined, in part, by a stack of relevant standards from various

Ed Davies.

Received on Monday, 13 November 2006 13:32:56 UTC