RE: [draft-ietf-ltans-reqs-03.txt] Questions & Remarks

[Peter Sylvester <Peter.Sylvester@xxxxxxxxxx>]

> [LH]I my view, a LTA was an trusted authority, then could be able to deliver signed-notification. (analogy whith TSA). Am I really wrong ? If it's the case, and if you have the time, could you clear my mind ?


That was also my position (water below), i.e. an service where you don't split
between the backend of archiving and a front end that delivers
an attestation. 

I don't thino you are wrong, you just use TWO services one behind
each other in a particular configuration, hydrogen+oxygen 
but you can have oxygen+carbon, this gives something else, also good for
plants.  The analogy stops here, because I am not saying that you need
boths, since otherwise plants won't grow. :-)

but you can have different type of USERS and SERVICES, it depends
on where you put responsibilities. And sometime you need a 'NOTARY'
in between. 


> [LH] I just had a problem with this. A the end of the archivation period, (if it's 30 years e.g.) how the user know that it is the end ?
> Must only the user deal with that ?

Well, a user can count? Whether a user is informed about a deletion, seems to
me an independant service, informing *THE USER* after 30 years
may be difficult anyway. what seems important for the protocol is
that the user can be identified, i.e. one has enough redundant information
about who owns or may own or may retrieve the info, this may be for example
the court of city, etc. 


> [LH] As you can read, my english is poor compare to yours. What I wanted to say is that an LTA could not provide means for confidentiality.
> Confidentiality could be part of the front end of the service, no?

Or the user: As far as I know, confidentiality with crypto means is never
supposed to last TOO long. An example: Would you think that a company that
hold a patent and recipe for a well know brown coloured drink would store
the information at an external archiver and transfer the information using
the Internet? 

regards
Peter