- From: Noah Slater <nslater@bytesexual.org>
- Date: Tue, 15 Jan 2008 23:34:30 +0000
- To: Paul Prescod <paul@prescod.net>
- Cc: James M Snell <jasnell@gmail.com>, Erik Wilde <dret@berkeley.edu>, uri@w3.org
On Tue, Jan 15, 2008 at 02:47:35PM -0800, Paul Prescod wrote: > To be clear: My preference is the protocol scheme error message > because the problem is not that data is missing on a Web server but > that an entirely different way of accessing resources is > unimplemented. But there is only STDIN, STDOUT and STDERR. That's it, just three. The URI space is infinate which is why a it's important to denote protocol from the URI scheme when you otherwise have no idea how to dereference a URI. You are going to have a hard time convincing me that your application is going to attempt to dereference "std:in" in some abstract manner using the "std" protocol and the "in" resource. There is no such "STD" protocol that you application implements and no need to do abstract dereferencing. There are three unique and seperate concepts which your language /should/ provide to you as a read/write object. All you need is an identifier to determin which of the three to use. > By the time the app attempted to access the remote server, it was > already in error. The same applies for when the application attempts to access the "in" server using the "std" protocol. Unless of course your application uses the "std:in" string as a magic token to switch into "use the language's native STDIN interface" in which case it should be clear to see that any old name would do, you may as well choose "OxDEADBEEF" or (the standard) "-" character. -- Noah Slater <http://bytesexual.org/> "Creativity can be a social contribution, but only in so far as society is free to use the results." - R. Stallman
Received on Tuesday, 15 January 2008 23:34:43 UTC