Alternatives to browsers (was Re: Alternatives to the W3C)

[John Lam]

The way I see it is this:

1. No proprietary runtimes (e.g. $$$ or platform lock-in). 
2. No complex runtimes (e.g. ubiquity of XML and HTTP 'stacks') and their
ubiquity (e.g. ready availability of open source implementations) results in
relative ease of porting of these runtimes to new environments as they
appear.
3. Ease of understanding of the underlying wire protocols (you can use
telnet to issue an XML encoded RPC call!)

Tunneling (read: sneaking RPC packets past your / your customer's security
folks) are bound to result in a rapid and rather severe clamp-down on your
application in no time at all (firewalls are very capable of filtering based
on http payloads). The nice thing about XML encoded RPC calls is that all of
the information is "in the clear" and can be readily parsed by HTTP filters.
There are also other efforts afront (e.g. SOAP) that make the intent of the
HTTP packets even clearer by adding HTTP headers that declare the
"SOAP-ness" of the packet payload, thereby allowing firewall admins to more
readily filter your packets: a compromise rather than an end-run.

My $.02.

-John

-----Original Message-----
From: Miles Sabin [mailto:msabin at cromwellmedia.co.uk]
Sent: Tuesday, January 18, 2000 10:57 AM
To: xml-dev at ic.ac.uk
Subject: RE: Alternatives to browsers (was Re: Alternatives to the W3C)


David Hunter wrote,
> As am I.  And it's exactly the combination of XML and HTTP 
> which, to me, makes the browser unnecessary in many 
> situations.  If I can just put some kind of application on the 
> client, whether it be written in Java, or Visual Basic, or 
> C++, and have that application communicate with my servers via 
> XML and HTTP, I get all of the benefits of using the Internet 
> (or an Intranet or an Extranet, or any of the other names I 
> can't keep up with), PLUS, I get all of the advantages of 
> splitting my processing intelligently between server and 
> client.

I'm having trouble seeing why XML over HTTP is preferable to
eg. CORBA or Java RMI (maybe tunneled through HTTP if there's
a need to traverse firewalls) for application specific comms.
How is application specific markup better than an application
specific binary wire protocol?

Cheers,


Miles

-- 
Miles Sabin                       Cromwell Media
Internet Systems Architect        5/6 Glenthorne Mews
+44 (0)20 8817 4030               London, W6 0LJ, England
msabin at cromwellmedia.com          http://www.cromwellmedia.com/


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ or CD-ROM/ISBN
981-02-3594-1
Please note: New list subscriptions now closed in preparation for transfer
to OASIS.

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ or CD-ROM/ISBN 981-02-3594-1
Please note: New list subscriptions now closed in preparation for transfer to OASIS.