One person catching a lot...

[Retired_Member38]

They wouldn't stop him straight away if they wanted someones abuse of it to be a good reason to implement changes though.

 

[Retired_Member38]

I'm not saying he is or isn't abusing anything as I don't know.

But if he is, the obvious answer to that question is that he's abusing it in a way nobody else has discovered.

 

[Retired_Member42]

Perhaps people can't afford to drop £4k on 10 Nominet tags. This has only come about in the past few weeks. Before it was more of a level playing field, almost a lottery.

 

[Retired_Member42]

I'm sure there are 24 people on this thread who could form a consortium to catch an hour daily and split the profits

 

[Retired_Member38]

Something to do with the randomness not actually being random?

Though if he has worked that out, I would say thats less 'abuse' than teaming up different allowances.

 

[Retired_Member42]

I don't buy into this whole calculating drop time or drop intervals. Purely from a programming perspective, it'd be easier to generate a random timestamp within a 24 hour range for each drop than assign each domain with a specific drop date based on a variable such as total to drop etc. I'm not saying it's impossible that they operate in this way, I just don't see the advantage in doing that. It's asking for someone to figure it out.

If there was a list of every domain dropping on a given day, a complete list and you knew the drop time of each of the domains in that list. Perhaps you could work out an interval and save quota in between those intervals. But still, so many domains are renewed at the last minute, some even after 12 it seems, you'd need to account for that. On top of that you've got domains like src.co.uk which don't drop as intended. More variables to account for. Any of these would throw your drop interval calculation out massively.

 

[Retired_Member38]

How is it abuse by him at all? If he worked something out that nobody else had, assuming he didn't work it out because he had access to a larger pool of resources, I cannot see how that would qualify as abuse by him. Should he build something in as not to take advantage of it or should he contact Nominet and declare what he may have discovered so they can "fix" it for everyone else who hadn't yet discovered whatever it was.


(from iPhone)

It wouldn't be abuse if he worked out the 'random' thing... but it would be a loophole that Nominet would probably want to fix.

If he is using multiple allowances from other peoples accounts then I would say that comes under abuse though...

 

[Edwin]

What about if the names all drop within a particular fraction of a second before/after each second? Say the "random" drop routine is tuned to 1/10th of a second accuracy or something.

Then putting in the DAC requests across the right range of ms within each second could mean you can hit it "more" than someone querying equally, and still not break the 1,000/60s limit. In other words, sleep for fractions of each second then hammer it for other fractions of each second...

NOTE: I don't do the drops, nor do I have DAC access. Just speculating idly...

 

[Retired_Member42]

What about if the names all drop within a particular fraction of a second before/after each second? Say the "random" drop routine is tuned to 1/10th of a second accuracy or something.

Then putting in the DAC requests across the right range of ms within each second could mean you can hit it "more" than someone querying equally, and still not break the 1,000/60s limit. In other words, sleep for fractions of each second then hammer it for other fractions of each second...

NOTE: I don't do the drops, nor do I have DAC access. Just speculating idly...

Idle speculation is good. It's a lot better than saying he's abusing the system, not abusing the system or simply that he's a programming genius with the ability to bend time

Knowing the exact time would not be enough, even to the millisecond as you're still going to have people going after it at that time. Then you've got to account for network traffic between your script and the DAC. It's never going to be a solid figure when you drill down to microtime. It would yield an advantage though. Perhaps the advantage is enough to be consistent in comparison to other catchers. As people have said, he doesn't catch everything under the sun, just the majority.

 

[Stephen]

Maybe there is something else?

When catching, I found that by omitting name servers in my request to register the domain that my success rate improved, this was done when I noticed a long time ago, that is what other catchers were doing.

Maybe there is something else about registering with epp code and the code language used and it is not just down to the Nominet drop times or bursts or checking at all but about the code registering the domain?

Sorry if somebody has already mentioned this, I have not followed the thread fully.

 

[Retired_Member42]

When catching, I found that by omitting name servers in my request to register the domain that my success rate improved, this was done when I noticed a long time ago, that is what other catchers were doing.

Maybe there is something else about registering with epp code and the code language used and it is not just down to the Nominet drop times or bursts or checking at all but about the code registering the domain?

Sorry if somebody has already mentioned this, I have not followed the thread fully.

Code changes only help so much. You've still got so much competition around the same second for the given domain. A good script will allow you to compete but not really give you a consistent edge in the lottery which is drop catching.

 

[Stephen]

the lottery

Code changes only help so much. You've still got so much competition around the same second for the given domain. A good script will allow you to compete but not really give you a consistent edge in the lottery which is drop catching.

As you say it is lottery but the point of this thread says somebody seems to have cracked the formula, so just trying to put another spin on what it may be!

I used to see the code as 2 parts when I wrote my own system, when is the domain available to register and secondly registering it, approaching it that way, I shaved off quite a bit off time of the latter registering code.

But when you try to register the domain, presumably Nominet does not lock the domain until it has processed the register order, so if your 'please register order' is quicker to process, even if it is not first, you will get the domain.

Did you not find by including 'no name servers' in the domain register code that was quicker?

 

[Retired_Member42]

I don't include nameservers no. I've done a lot of things to get a few extra milliseconds out of my script which has had a positive effect but you're still only throwing your hat in the ring as you've got lots of other people going after these premiums at the same time. Shaving a few miliseconds off your reg time is ineffective when 100 other drop catchers are trying to reg the domain at the same time. It's not a case of fastest registration code wins. There is an angle at play here, no script is that fast to catch so consistently at all times of the day. It doesn't matter how good your programming is. It's either something to do with knowing the drop time or drop time range or it's multiple tags. It's not "good code".

 

[DaveP]

I don't include nameservers no. I've done a lot of things to get a few extra milliseconds out of my script which has had a positive effect but you're still only throwing your hat in the ring as you've got lots of other people going after these premiums at the same time. Shaving a few miliseconds off your reg time is ineffective when 100 other drop catchers are trying to reg the domain at the same time. It's not a case of fastest registration code wins. There is an angle at play here, no script is that fast to catch so consistently at all times of the day. It doesn't matter how good your programming is. It's either something to do with knowing the drop time or drop time range or it's multiple tags. It's not "good code".

Nail. On. Head.

 

[RobM]

*edited - apparently suggesting that nominet's claimed randomness is perhaps not random is the same as accusing people of 'wrongdoing'. Let nominet get on with whatever they want eh? *

 

[Paullas]

The only people who can check logs and spot obvious discrepancies are nominet and, newsflash, they don't give a shit.

Nail. On. Head

 

[Retired_Member42]

I'm using this thread as a way to gain insight be it theories, speculation or otherwise. I don't see it as a personal attack on Denys nor is it a way to applaud him. I think people just want to figure out what he's doing and get in on the action. It's not about shopping him to Nominet or even Nominet finding out what's going on and stopping him. As I've said before, if he's figured something out, fair play to him.

 

[Murray]

I don't think he is doing anything improper.

If you had this trick up your sleeve that was dodgy surely you would save it for the really good drops so it just looks like you're lucky, rather than catching everything and being so obvious about it.

Saying the above, the fact he is catching as much as he can whilst he can might suggest it's an exploit that could close at any time? or anyone could find it so he is making the most of it whilst he can.

 

[RobM]

*edited - apparently suggesting that nominet's claimed randomness is perhaps not random is the same as accusing people of 'wrongdoing'. Let nominet get on with whatever they want eh? *

 

[WealdDomains]

(Edited as it didn't add anything to the discussion)