[Home] [By Thread] [By Date] [Recent Entries]


In a message dated 17/04/02 21:51:54 GMT Daylight Time, Doug.Ransom@p... writes:



I think big companies like IBM and etrade would be motiviated to use a feature like this so one knows "a page is from IBM" or "a page is from etrade" without having to got through the expensive and uncacheable SSL method.  They in turn would need to influence browser vendors.


-----Original Message-----
From: AndrewWatt2000@a... [mailto:AndrewWatt2000@a...]
Sent: Tuesday, April 16, 2002 1:38 AM
To: xml-dev@l...
Subject: Re: XHTML adoption curve


In a message dated 15/04/02 22:07:03 GMT Daylight Time, Doug.Ransom@p... writes:


I thought of a little carrot that might help improve the HTML developer
affinity for XHTML:  Signed XHTML.  If XHTML pages were signed in a specific
manner with using x.509 based PKI, browsers could inform users they can
trust the content of the page. 




Doug,

Perhaps you are confusing the notion of being certain (or do I mean confident?) of the origin of a page with the notion of trusting the content.

If, hypothetically, a page was certified as being produced by George W. Bush, it wouldn't help me trust the content because I would still feel that he didn't win "that election" and I would still view his "veracity quotient" as being defective in the extreme. Maybe I am unique in having such unworthy doubts.

It illustrates the more general point that knowing the origin of content is of limited value in knowing whether the content can be trusted or not.

It seems to me that authentication (if we can use that term in this context) and trustworthiness are two separate (and arguably orthogonal) processes.

Andrew Watt

Site Map | Privacy Policy | Terms of Use | Trademarks
Free Stylus Studio XML Training:
W3C Member