Re: [icann-europe] Recommended Reading: Brad Templeton on ICANN and the DNS

["Griffini Giorgio" <grunz@tin.it>]

Marc Schneiders wrote:

> 
> > This is more or less exactly what millions of users actually do
> > when they use ICQ. Surely is not as friendly as using a name but
> > actually works well for zillions telephone device out there. 
> 
> Both ICQ and (mobile) phones have their own directory. You put in
> someone who talks to you without even looking at the number. You
> don't have to type (or remember) the number ever. 
> 

Yes, this exactly the point...local directories...but you had to type it almost 
once and probably you got the number directly from the counterpart o by 
using a global search....

> 
> It does not make sense to put a long number on a car, or advertize it
> on the radio (unless it is 1111111111). People will not remember it,
> when they get to their office.
> 

It depends on available tools... If people do not have safe means to find the 
counterpart other than that 'long' number you are absolutely right... If one can 
resort to a safe directory (which is not available or possibly incomplete 
(contribution driven search engines) with the current DNS) things becames a 
lot easier and consistent. Just to talk about imagine that when registering a 
domain one get a hierarchical unique opaque identifier  (technically 
corresponding to what actually is the 'name') and then is allowed to give 
under his own responsability 4 to 8 (we say)  'attributes'   to put into the 
'directory' entry rwhich must exist and is reserved for his domain.in order of 
what the registrant thinks to be more important to him for being found / 
identified / searched for.
When a user want to contact him use: a) direct opaque identifier as shown 
on business card, email, ad...  b) search the directory by keyword which 
likely identify the registrant (trade marks , company name, product name, 
family name)  c) resort to a local directory entry where the opaque identifier 
has been collected by method a,b  

With this kind approach all IP and trademark claims goes under the 
appropriate jurisdition and there is more freedom about setting/relaxing 
constraint that we actually have in the current model (and you can always 
make an attribute mandatory and just a multiple choice between a small set 
of keyword and you get the current gTLD scenario) 

> It is not so much a matter of how to contact your friends and
> relatives. They are in your private directory in your mobile, ICQ etc.
> It is all about how people think they can sell their stuff. Telling
> this is wrong does not make it go away, just as telling an alcoholic
> to stop drinking does not help.
> 

It is a matter of tools... until there are not safe directories (I mean that do not 
exclude/limit reachability for whatever reason) there will not be tools for user 
to access such directories. The only way to have safe directories is to make 
them part of the registration infrastructure. 

Trying to condense the whole thing in few words: ...it is time to switch from 
the 'hierarchical view' to the 'relational view'.  What I see right now is that we 
are just adding new tables to the old hierarchical 'database' to work around 
the lack of relational capabilities...and as already seen in real world of DBs it 
may work for a while... but there will be a moment when you cannot add 
more tables or the 'database' will became unmanageable-->useless.


> The DNS as we have it, will not go away, unless a real alternative
> comes up that satisfies everyone, including marketing managers. 
> 

I think that proposing them to apply same kind of abuse they like on DNS 
(forced uniqueness to save money about protecting/advertising a trade mark) 
to the phone numbering system (in a way truly unique not the ABC=2  
DEF=3 and so on way) will keep them a little bit busy.

Sorry for the emphasis...

Best regards
Giorgio Griffini



-- 
To unsubscribe, e-mail: icann-europe-unsubscribe@lists.fitug.de
For additional commands, e-mail: icann-europe-help@lists.fitug.de