[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Excellent IETF BCP on XML

• To: 'Miles Sabin' <miles@m...>, XML Dev <xml-dev@l...>
• Subject: RE: Excellent IETF BCP on XML
• From: "Bullard, Claude L (Len)" <clbullar@i...>
• Date: Fri, 22 Nov 2002 14:45:11 -0600

Yep.  However, since packets are sniffable?  
I'm trying to envision how many ways it can 
be abused, for example, alteration en route 
to a target that believes what it is getting 
is safe to dereference.  Yes, they would be
nuts not to check the value, but do they 
always.   Gotta be paranoid given the stakes 
and the times.  This isn't 1993.

I have this nervous feeling that in the 
80/20, think small, dare to do less, philosophies,  
the URzedEverywhere implications aren't 
completely thought through and we have entered 
a historical timeframe where they have to be.

len

-----Original Message-----
From: Miles Sabin [mailto:miles@m...]

Bullard, Claude L (Len) wrote,
> So the issue is for the protocol designer to explicitly say ... what?

Don't dereference it unless you need to and your trust it. Same as any 
other URI.

• Follow-Ups:
  • Re: Excellent IETF BCP on XML
    • From: Miles Sabin <miles@m...>
  • RE: Excellent IETF BCP on XML
    • From: "Simon St.Laurent" <simonstl@s...>

• Prev by Date: Re: Excellent IETF BCP on XML
• Next by Date: RE: Excellent IETF BCP on XML
• Previous by thread: Re: Excellent IETF BCP on XML
• Next by thread: RE: Excellent IETF BCP on XML
• Index(es):
  • Date
  • Thread