[ih] History of Naming on The Internet - is it still relevant?

[Patrik Fältström]

Personally, for reasons you mention and more, I do not classify URLs (as we know them) as long term stable identifiers. This is why I was working with others so much with URNs.

My point was only that for URLs to be well functional we must allow the creator of a URL to have the ability to choose the lifetime, and for that, control over the domain name is a minimum for what is needed.

I also, fwiw, believe ”register for 10 years” is a weird registry policy as I believe a registrar can offer such services for their customers and having registry and registrar policies mix is confusing.

Yes, completely agree on work done by Bruce and others is excellent. Also many libraries around the world do excellent job in preserving information for the future.

And for that, long term identifiers are really needed.

Patrik

> 21 juli 2025 kl. 20:39 skrev Karl Auerbach <karl at iwl.com>:
> 
> On 7/21/25 8:46 AM, Patrik Fältström via Internet-history wrote:
> 
>> - Who decides the lifetime of a "name"
>> 
>> Regarding lifetime, I think (influenced by a few discussions with Tim Berners Lee about lifetime of URLs) the most important thing with domain names is that the holder of a domain name can decide what the lifetime of a domain name is, because that is the key thing to a decision about the lifetime of a URL.
> 
> I am not sure I fully agree about the lifetime of a domain name. I agree that the holder of the name has a level of control.
> 
> However our world is ephemeral - people age, get bored, run out of funds, and die.  Same even for institutions, but usually on longer time scales.  (I work with astrophysicists who would quickly remind me that even the Earth and Sun are ephemeral. ;-)
> 
> One thing that came out in the early days of ICANN was a registration system driven by a calendar - one year increments, ten year maximum per renewal period.  It was never clear to me how or why this came about - it seems to me to be something in great need of review.
> 
> That calendar drive expiration of domain names pretty much guarantees that domain names (and the URLs based on them) will eventually expire and vanish.  Services have evolved (such as, I believe, Iron Mountain) that will manage a portfolio of domain names on behalf of a paying client to make sure that those names are not inadvertently lost to expiration.  That kind of service is not used by smaller business or individuals, thus essentially guaranteeing that a significant portion of our Internet URL namespace will eventually vanish, with potentially great loss
> 
> I fear that we are putting dry kindling around the Internet's horde of data - that we risk a reprise of the burning of the Library at Alexandria.
> 
> This is of concern to future historians.  (And as an intellectual property attorney I have concerns about the impact on the patent system as ideas once visible on the net become invisible and someone comes along to claim them again in a new patent.)  At this point I kinda thank the gods-of-the-net for people like Brewster Kahle and the Internet Archives. (I also make yearly contributions to the Internet Archives.)
> 
>         --karl--
>