35 comments

[ 1.6 ms ] story [ 88.1 ms ] thread
I've been following this saga for a while. but always based on The Register's coverage. Is the current management really as entirely screwed up and corrupt as they appear?
Can someone please give a little bit more context for people out of the loop? It’s confusing to understand who stands for what (board of .uk, that company, the ex BBC dude) and what is/was going on.
Not sure if this helps, but it gives a little background. This piece [1] from The Telegraph was published before the chiefs were voted out in the EGM. It says that Google, a Nominet member, was expected by activists to abstain in the EGM vote.

The Nominet saga seems to be, roughly, split between small UK members (activists unhappy with Nominet's leadership/direction), and larger overseas members (GoDaddy etc) who seem to back the status quo.

I don't think that it is known whether Google voted or abstained, but their vote would have had some clout, given the way voting rights are weighted to size of members. Perhaps Nominet's continued deafness to the activists suggests that they are confident of big-hitter support in the event of a second EGM.

Re the ex-BBC dude, another Telegraph piece [3] gives info on his stance and his background.

The Telegraph link will hit a paywall unless you turn off JS. Alternatively, the first piece was archived four weeks ago [2], and the second is available at [4].

[1] https://www.telegraph.co.uk/technology/2021/03/14/google-dra...

[2] https://archive.is/ZrNO2

[3] https://www.telegraph.co.uk/business/2021/01/30/former-bbc-c...

[4] https://archive.is/ySSQG

EDIT TO ADD: As per child comment by ohashi, this link to the vote outcome on the campaign site is informative: https://publicbenefit.uk/#support Also, as ohashi helpfully points out, "You can see Google abstained. GoDaddy and 1&1 support the status quo. Tucows, NameCheap supported the removal of the board."

Thanks for that.

You should link the actual campaign: https://publicbenefit.uk/#support

You can see Google abstained. GoDaddy and 1&1 support the status quo. Tucows, NameCheap supported the removal of the board.

Thanks for that, you are right. I will amend.
Have to say I was surprised when I saw Michael Lyons's name pop up as a candidate for Nominet chairman. I went searching and found this The Register article[1] about a report he did for Nominet on their future governance and operations[2] which allowed the pieces to click into place.

Sir Michael has form for not delivering the answers he was hired to deliver, so the outcome of his report for Nominet is no surprise to me. I think the man is allergic to whitewash[3].

[1] - https://www.theregister.com/2016/02/05/michael_lyons_tells_n...

[2] - https://publicbenefit.uk/resources/Sir-Michael-Lyons-Report-...

[3] - I worked with Sir Michael on his Lyons Inquiry into Local Government, so this admittedly anecdotal evidence is at least first-hand anecdotal evidence.

Nominet (.UK) is a membership organization, basically anyone registering more than a few domains would be interested in becoming one. There are some 2,000+ members. The voting is determined by number of domains registered but capped at 3% voting power (so large players like GoDaddy have a large voice - and board seat). Nominet has been trying to go from a non profit entity which provides registry services into a startup under the recent management (many of whom were just fired by the members in the latest vote). They also started paying themselves ridiculous amounts of money (see: the complaints https://publicbenefit.uk/)

The campaign proposed a vote (EGM) on removing 5 board members and putting in two well respected veterans (Sir Michael Lyons and Axel Pawlik) to help steward the organization back to its core mission of running a registry and giving profits to charity instead of raising executive salaries and trying to be a tech company (Nominet tried to do things like self-driving cars and spectrum management, which all failed and lost money).

The campaign was a success despite Nominet spending (millions?) spreading FUD about how government would intervene and it would be a disaster. 5 board members were removed including the CEO (who resigned one day before - what a pathetic way to go). The remaining board then decided to put one removed board member as the interim ceo (yeah, wtf? members literally just voted her out). And now they are ignoring the second half of what the whole campaign was about, reforming the organization, they aren't bringing in Sir Michael Lyons and Axel Pahlik, instead they feel capable to recruit their own.

I'd bet a second EGM coming to remove the rest of the board which continues to spit in the face of its members and ignore their wishes.

Thank you so much! Man I cannot believe I missed all that drama.
Yes. A couple more points worth noting:

Sir Michael Lyons wrote a report on Nominet some years ago. The Nominet board basically ignored all his recommendations, though his report was widely supported by the membership.

Axel Pawlik was head of RIPE, which is an Internet membership organisation with a good relationship with its members and a reputation for doing worthwhile things for the benefit of the Internet as a whole.

ICANN has failed us, Nominet has failed us, Public Interest Registry has failed us. We definitely need some better system where a single entity going rogue can't affect millions of domain owners. I can see two approaches.

One is where we make it technically impossible for DNS operators higher in the tree to mess with us. That's what the GNU Name System proposes with the "hyper hyper local root", a petname system which supports delegation, to defeat censorship.

Another one is a domain name cooperative like CCOR.org attempted to build. So the organization serving our TLD is in fact operating transparently in the interests of its members paying administrative fees (think <1$/domain/year)... not a business selling trying to make profits.

In my view we need both! Replace the corrupt pro-business ICANN with a democratic non-profit foundation, and at the same time make it technically harder for bad actors to mess with everyone.

Having a (domain) name on the network should be a free and accessible public service for all. They're half a dozen lines of text (counting glue records and DNSSEC anchors) in a database, it's OUTRAGEOUS we have to pay for that at all. In the early nineties, it used to cost money to host just a few kilobytes of information, and that's why we introduced administrative fees for domain names. Why are we still paying >10$/year for less than a kilobyte of information hosting just so that shady businesses can make a profit?

But how would that work in practice ?

Like letsencrypt ?: what's needed to incite enough people to move on so that it becomes feasible and a reality (while keeping the old system alive for legacy reasons and for the time it takes for thing to settle down) ?

If we're talking about GNS to replace the DNS protocol, it's in fact backwards-compatible and intends to recognize ICANN as root zone, as was the case previously.

If we're talking about a domain name cooperative, that would act as a registrar and potentially as a TLD in the future, all we need is a non-profit organization, some bylaws, and enough good-will people to make it work. Maybe OpenNIC would be it?

> One is where we make it technically impossible for DNS operators higher in the tree to mess with us. That's what the GNU Name System proposes with the "hyper hyper local root", a petname system which supports delegation, to defeat censorship.

DNS and our existing registries are too embedded in too many critical systems, businesses and workflows for it to be replaced by a niche unproven system, over the relatively small cost of domains.

Improving the governance of existing structures is feasible and useful.

Those new systems retrofit into the existing DNS system very easily. There is no excuse not to try new approaches.
Quote>>>

They're half a dozen lines of text (counting glue records and DNSSEC anchors) in a database, it's OUTRAGEOUS we have to pay for that at all.

This is not what you are paying for. Every time this argument comes up it hides the real problem...what you are paying for is the legal wrangling, security overhead and general compliance that comes with the internet.

Phishing, trademarks, spam, business disputes,and valid legal requests will all still exist no matter who runs things. Those costs are what your domain fees really have to cover.

It’s been a very long time since domain costs were highly dependent on the technical requirements of the data they represent.

> Phishing, trademarks, spam, business disputes,and valid legal requests will all still exist no matter who runs things.

Of this list, only "valid legal requests" is something a TLD should be concerned with in my opinion. And even that is questionable, given that law is often unjust, favoring the powerful and abusing the powerless.

We already have strong regulations holding internet hosts accountable for what they're hosting, and many countries (like France) have censorship at the DNS level. Isn't that more than enough to deal with abuse?

> It’s been a very long time since domain costs were highly dependent on the technical requirements of the data they represent.

Yes, and that's the problem. When you pay 1$/month for web hosting, you're not paying a thousand times the price because of hypothetical legal concerns. Why would it be different with DNS? Why is it normal that a line in a zone file on some TLDs costs more than an actual VM in a datacenter?!

TLD's have various compliance requirements around abuse, which does include the afore mentioned. There is significant volume related to this compliance as well as ensuring registrars maintain compliance. It's the nature of the existing arrangement so it's not hypothetical.

Of course the individual cost is low for a single item, however, businesses have never operated on that principle. They aggregate the profit and the risk across a pool, no surprise here.

> Of course the individual cost is low for a single item (...) They aggregate the profit and the risk across a pool

It's not the only way ever to deal with that. We could imagine a cross-TLD insurance which would take care of legal costs, for instance. Or have for-profit entities pay the current registration price (more?) so that others can have domain names for free. Or have a donation-based model with transparent accounting so people can see whether they need to donate more money. Or crowdfunding campaigns when a specific issue arises and introduces legal costs.

I'm not saying i know the best way, i'm just saying we know for a fact the current system is only making TLD execs pocket 6-digit salaries out of a public service, and that's a scandal. If at all, they should be paid the same salary as anyone else working for the TLD. Yes, the same as even the cleaners/cooks if such things apply. This way, they'll have incentives to either make things cheaper for the rest of us, or share some of the wealth they accumulate off our backs with less-privileged folks.

>a democratic non-profit foundation

Isn't that what Nominet basically was and here we are?

>Why are we still paying >10$/year for less than a kilobyte of information hosting just so that shady businesses can make a profit?

To provide a minimum barrier to people simply writing a script and registering literally every domain in existence. Either to resell or troll. Probably also to provide a financial paper trail in case of abuse, illegal activity or other reasons.

> To provide a minimum barrier to people simply writing a script and registering literally every domain in existence. Either to resell or troll. Probably also to provide a financial paper trail in case of abuse, illegal activity or other reasons.

But that happens anyway, what has it discouraged?

>But that happens anyway, what has it discouraged?

Just because something isn't great doesn't mean it can't become worse or that the current solution isn't making it better than it could be.

Sure, it might get worse. But it may as well get better. Implying that changing paradigm can only make things worse is a fallacy.
> Implying that changing paradigm can only make things worse is a fallacy.

So is saying that changing the paradigm will make things better or stay the same. I'm not making a data driven argument for not changing things, I'm simply pointing out that you're not making one either.

No because i don't pretend i have the best solution to every problem. However, i'm convinced not doing anything is the worst possible answer.

https://ttm.sh/uzm.jpg

>>a democratic non-profit foundation > Isn't that what Nominet basically was and here we are?

Nominet was a non-profit, but not a democratic one. Democracy is not "choose your rulers", but "choose your rules". Having a board/parliament is by definition the opposite of government by/for the people. The simple fact the community has to fight against the board is proof of that.

> a minimum barrier to people simply writing a script and registering literally every domain in existence

That's just not true. First, because that was definitely not the argument presented when administrative fees for domain registrations were brought up. Second, because bad actors usually have a lot more resources than good actors... Sure you may stop a random teenager from registering many domains, but random teens is not who i'm afraid of personally. I don't know of a single time where economic barriers have stopped bad actors from engaging in malicious/hurtful activities. Third, because some TLDs are actually already fully free-as-in-free-beer because they're operated as a public service and they are not facing the problems you mention (to my knowledge), if only because their terms and conditions disallow parking spaces.

> Probably also to provide a financial paper trail in case of abuse, illegal activity or other reasons.

Not sure about that either. In most jurisdictions responsibility for illegal content falls on the hosting provider, not the registrar/TLD. In the rare cases that your TLD is going to seize your domain for whatever reason, whether there's a paper trail or not changes very little because the content is effectively taken down as was requested. Whether that's a feature or a bug depends on what kind of stuff is being censored and for what kind of reasons.

Personally, i'd be interested in a more democratic take on domain seizures. I'm pretty sure most people are opposed to wikileaks/scihub having their domains seized, despite what a board may think.

> or other reasons

Yes, to grow a crazy lucrative business out of infinite resources nobody should be competing for. I'm calling that an anti-feature.

> Nominet was a non-profit, but not a democratic one. Democracy is not "choose your rulers", but "choose your rules". Having a board/parliament is by definition the opposite of government by/for the people. The simple fact the community has to fight against the board is proof of that.

That's a No True Scotsman argument since by your definition there is no large scale democracy in existence. Most people would consider a representative democracy to be a democracy.

> by your definition there is no large scale democracy in existence

Take a look at Chiapas or Rojava. They are large-scale governance systems that most people would agree fit the definition of a democracy, if only because local assemblies and women's assemblies have a lot of power. They both have some forms of representation but nothing similar to what we have here in terms of accountability. You can for example watch "Viva Zapata" documentary about autonomy in Chiapas.

Also, if it hypothetically did not exist yet, would it be a reason to give another meaning to the word? If we can admit a God does not exist, should we call an AI a God because its ways are ineffable?

> Most people would consider a representative democracy to be a democracy.

In some places, most people would consider vaccines a trojan horse of the Illuminati and homosexuality to be a sin. Does the majority always have to be right? Isn't the tyranny of the majority precisely one of the problems with our pretend-democracies? To be fair, tyranny of the majority would be better than the current tyranny of the minority elite elected by the majority (of people who have a right to vote, which is not everyone) in a process that resembles more reality TV shows than a political argument.

Tor's Onion Services are pretty much ideal for this, and provides additional anti-censorship guarantees by providing anonymization for both parties.
Onion names are a very good global crypto-secure naming scheme, but they don't solve the human-meaningful part of Zooko's triangle. GNS has some similar design properties, but "solves" the triangle by tying two different naming systems: petnames for locally-meaningful/secure names, root names for global/meaningful names.

Also as good as tor's design is, it requires some level of centralization with hidden service directories, which GNS doesn't. Disclaimer: i'm a happy tor user for many years, criticism here is in good faith (posting this from the tor network).

If I could I would give you 50 up votes just because you wrote 5 paragraphs about what you see as problems with the current system without suggesting that it be replaced with some kind of blockchain.
We probably need a blockchain-based commenting system to make sure you can afford 50 upvotes as long as you have enough miners running ;-)

Who cares about sybil attacks when you can run a proper ponzi scheme and get rich! /s