Owning a Domain Increases Digital Self-Reliance
11 points by confusedalex
11 points by confusedalex
Owning a domain makes it a single point of failure if the US (ICANN) decides the country of your registrar is worthy of sanctions.
https://www.internetgovernance.org/2017/01/13/icanns-jurisdiction-sanctions-and-domain-names/
If only it was feasible to own a domain rather than renting one from a registrar!
I still feel like that would be a great use of cryptocurrency tech (like Namecoin!!), but I can understand why it hasn't gotten off the ground yet. Last time I tried to use Namecoin, I had a hell of a time even acquiring any NMC to begin with, and after that, the domains expired and were squatted by someone because I forgot to renew them, and there's no such thing as auto-renew in the "most usable" Namecoin software. But I'm still glad it exists, and maybe if it ever becomes truly necessary for enough people, they will start using it and start fixing its usability issues.
Owning something rather than renting it implies that a) nobody(ish) can take it away from you, but b) you pay for the infrastructure and maintenance costs. "Paying infrastructure and maintenance costs" is complicated when you have a massive shared resource such as a global namespace; it's really not a technical problem, it's a management and logistics problem. "What happens to a domain when the owner dies" as mentioned in sibling comments is a good example.
It would be kinda nice if you could just rent a domain for 100 years and never have to worry about it again though.
I will preface this by saying I know nothing about Namecoin.
How can the domain expire if you own it? Or were you referring to a traditional domain and not something related to Namecoin? It would seem to me that anything worth pursuing as an alternative to the current rental situation is one where your domain (or some form of digital identifier) doesn't expire and can't be squatted (but could be sold to someone else later if you wanted, I guess).
The namecoin name registration transactions are kind of like Bitcoin time lock transactions. In other words, the consensus rules enforce that only one person is allowed to modify the value attached to that name for a certain period of time, a certain number of blocks into the future.
It was set up like that instead of permanent ownership as a safety mechanism so that if you make a mistake or you lose your private key, it doesn't mean that the name is lost forever, like how it is if you send bitcoins to the wrong address. I don't think that's necessarily a bad design for namecoin, but I do think that the software should do a much better job of supporting auto renew.
I assume renewal is necessary because the domain owner can die or go out of business without having made arrangements for the domain, meaning Namecoin would gradually run out of domains as all the good ones are stuck pointing to dead websites.
In the old times, ISPs gave you email accounts.
I think when you get any kind of Internet connection, you should get a domain, some friendly tools to migrate your domain, and some basic services (minimally, email forwarding, static web hosting).
Working email is a basic necessity, and access to it should be somewhat regulated (e.g. I should be able to recover access to my email address under catastrophic conditions by going to a physical place and identifying myself). And being held hostage to your email provider I think has led to a bad place.
The problem with this is that not everyone has their own Internet connection. I never had an ISP email address because my first few years on the Internet were spent at my local library. I signed up for email addresses from places like Excite and Yahoo! instead.
What you describe feels like it ought to be handled by government rather than ISP (though maybe that's what you have in mind as well; I don't want to assume either way). Maybe everyone over a certain age is assigned a subdomain at some my.country domain that is theirs, and they can apply for a change or additional subdomain if they like. That has its own set of problems, but I'd rather it be paid for by taxes than corporate goodwill.
I mean as long as it is not the exclusive way to get one I'm not seeing the downside of the proposal. After all, you managed to get an email address without an ISP, right? :)
I'm not saying it's a great idea that needs pursuing, but I really don't see any large downsides.
Yeah, whenever I think about this, I think about other entities that could do it.
I suspect a lot of people are suspicious of the government, but it's another possibility I've considered.
Banks could also work. (And in Spain, many banks have an absurd amount of offices. They have closed a lot of them, but still...)
I had never checked their prices, but .nom.es (for people's names) domains are like 2€/year, which is close to free.
I think national post offices should provide basic email. It seems like it'd make sense.
Ah yes, back when the state-owned post and telecoms monopolies were often the same org, and when there were national mandates to support the OSI stack, and the future was going to be X.400 for email with a global X.500 directory.
I have on a shelf a manual for the PP MTA which was notorious among UK postmasters in the 1990s for being perfectly painful. (The manual’s only explanation for the name is that it does not stand for Postman Pat.) PP did protocol conversion between JANET grey book email, Internet email, and others such as X.400 which were less widely deployed. (I kept the manual to provoke old postmasters if they happened to visit my office in the university.) PP was rather brittle and inefficient: AIUI it liked to explode messages into multiple files on its queue, and sometimes became unable to reassemble them for onward delivery.
Internet MTAs were much less troublesome, even Sendmail with the special macro hacks to deal with endianness conversion between DNS and JANET NRS names such as <fanf2@uk.ac.cam.phx>. When JANET added support for IP in 1991 the network went from 100% X.25 to less than 50% in six months. Internet software was available for far more systems and much easier to get working.
This is still "This sucked 40 years ago when nobody knew how to do this for real".
Yeah, with a subtext along the lines of “… and why would we expect them to do better now?” especially since the old government-owned monopolies have been sold off and are now operated for profit not for collective benefit.
Pretty sure this was planned in Sweden in the 1990s. It didn't pan out.
Now, services like Kivra serve as "post boxes" for stuff like tax returns, receipts etc. In other words, if you want to receive "paperwork" you can sign up for the service and get it there.
Worth investigating further, but "didn't work in Sweden in the 1990's" seems only tangentially relevant to 2025 2026. :D
I think "this whole government regulation thing for digital comms not working out" can be found intertwined in the early years of many things related to phreaking and the formation of the Chaos Computer Club - I also kinda think it does not make sense, especially the parts about not tinkering with your hardware, free choice of hardware (with recent legal developments of allowing/disallowing that).
IMHO the problem is that the state usually delivers some sort of non-open solution. Maybe it's more realistic to hope for something build/rolled out in the 2020s than the 1980s, but I am pessimistic. We also had De-Mail and it was the worst possible implementation imaginable. OK, second worst after jsut being a WhatsApp account ;) But then again Germany is comically bad at stuff like this anyway.
Germany can do it when the stars align. I remember when the government managed to stomp out the contract tracing app for COVID within like a few months, implemented in a very privacy friendly way, for a dev cost of like 10 million € and ongoing annuals of under 1 million. We just suck once the government lets in cosultants.
I thought a bit more about this, and there are fundamental differences between postal/parcel delivery and email. (I know PTS is a generic term for postal and telecom services, but at least in Sweden post and telecoms have been separated as entities since at least the 1880s).
Post fundamentally just has to be concerned about physical addresses, not the actual inhabitants thereof. If I move into a new apartment, I just have to put up a nameplate on the door, and post addressed to me at that address will arrive. The postal delivery doesn't have to track my current legal or chosen name, and can deal with temporary residents (care of). It just has to keep track of postal endpoints, which are pretty limited.
Consider the difference with email. You need to keep track of different spellings of names. How do you localize Swedish (or German) characters? How do you handle legal or voluntary name changes?
Then there's spam. At least commercial mail is paid, spam isn't. You need to deal with it, which takes resources.
The current Swedish solution with web- and app-based digital postboxes tied to your legal identity[1] basically fulfills the need for the government to communicate securely and verifiably with citizens, in a way that SMTP-based email could never do.
[1] overview in Swedish here https://www.digg.se/for-privatpersoner/digital-post-for-dig-som-privatperson#h-sv-default-anchor-0
Fair enough.
Also worth mentioning that Sweden doesn't have a national post office anymore, it's a joint stock company (aktiebolag) owned jointly by the Swedish and Danish states. Also, the Danish part is no longer required to deliver private letters anymore.
In other words, the basic infrastructure proposed - you have a physical address, you should have an email address, therefore the local PTS is the most appropriate provider of such - is no longer the case in Sweden (or Denmark).
Or whether non-technical people will arbitrarily add an "m" at the end of an email address that ends in ".co" because they don't know any better?
This has bitten me several times. Once somebody even sent "me" money (via Paypal) to .com instead of .co.
I hoped .co would gain more traction with Twitter using it for t.co, but it seems t.co is no more.