A taster get 5 days to taste the domain at no charge. they simply release the domain before the 5 days are up. Of course there is no problem in them tasting it again after that. There are literally millions of these domains in this state of flux. Of course, if there is no traffic, then generally they would drop the domain. But that doesnt prevent another taster picking up the domain and trying their luck with it.

People aren't too bright if they aren't doing the query directly at the registry level and using some 3rd party whois or registrar.

Internic is ICANN, and is responsible for .com and the query goes directly to Verisign.
Not that I trust verisign but they are a multi-million dollar corp that can be sued for bad behavior so I tend to lean that way instead of the registrar level:

[internic.com...]

[edited by: amznVibe at 7:02 am (utc) on Nov. 10, 2006]

Verisign also released sitefinder, so as far as trust goes... well I would not include Verisign as one of the best companies on earth. But they are the source of .COM so their is no way around not using them in some way. Internic.net is ICANN, but if you want to go to the source, go to verisign-grs.com and do your query their.

As far as being liable, all companies can be sued. Any company over 10 people in size has a lot to lose in a legal battle. I hear what you are saying, a one man shop is more likely to do something dishonest. A random no name whois site may be a one man script plugged into a taster's system.

What I suspect is really going one, Registrars and Large whois sites can be trusted. But the worst thing you can do is test if a domain exists by typing it into the web browser. This is less secure then using a whois site. By using the DNS of your ISP to resolve a domain that doesn't exist will cause the ISP to log NXD data. Earthlink I know for sure sells that data to domain tasters, I would suspect AOL and other ISPs do the same.

Coming late to this thread... It has been mentioned as a sidenote a few times in this thread, the last time by Lisa, but I don't think everybody got the message yet. And I think there is a huge confusion with regards to:

1. Typing your domain-name-of-choice into a "Check availability" search form on a website and
   
2. Launching your Windows/Mac/Linux Whois-application or using a command-line whois and querying a root nameserver

My gut feeling is that option #1 will be logged for sure, and most likely also mined - the smaller/cheaper/obscure the website the larger the chances. And my gut feeling for option #2 is that it is possible to eavesdrop it, but that it simply does not happen because it's simply not manageable.

So, if you go to

www.ReallyCheapDomainNames.biz

and check for

MustHaveThisDomainReallyHard.com

, then chances are good it will be taken if you return a few hours later. If you do the same via a whois query from your Linux server, it most likely will still be there.

On a sidenote: if you need to simply check if a domain as already live, a simple "ping" works wonders.

...Only one of mine has been registered by someone since I turned in my last list, which may not be statistically significant out of the 200 I checked for the test.

Actually that is more than sufficient. According to statistical experts and mathematicians a sample size of 30 is valid.

Your test strongly confirms my view that it's not an issue.

I love how a concrete example of a domain name being scooped up is being spun as proof that this doesn't happen, or at least is "not an issue", whatever that means.

In fact, including previous posts, the poster you are quoting reported a total of six domains being registered after doing a WHOIS. It seems to me that people are rushing to the conclusion of their choice before a reasonable amount of evidence has come in.

a total of six domains being registered after doing a WHOIS

Really a whois? From a command shell via the OS's whois command? Or rather a query from a webform from some site? Those are two completely different issues!

Yeah you already said that. The thrust is this thread is whether it's safe to do a whois via a registrar's own website interface, but if it turns out that a command-line query is inherently safer then I can imagine a lot of people switching to that.

Anyway I believe this person used a registrar's website, but read the thread for confirmation.

This topic has been discussed here several times in the past. To my mind theres just to much anecdotal evidence, over a long period of time, to say it doesnt happen.

I imagine that this could happen depending on the registrar. I'm like others in that I'd expect the better/longer established outfits are less likely to be doing this.

Note to Webwork: whois queries are actually different from queries to see if a domain is available or not. Whois queries are to find out the owner registration info on a domain. While that can tell you if there is an owner and hence if the domain may be available, plain availability queries are actually a different beast.

I know a site that queries whois info from all the registrars, depending on the TLD, and returns that info in a normalized format... I prefer using a service like that, since I believe it obscures from the registrars who is performing the request.

For a registrar to monitor queries and then respond to that data in some way is just far too easy and obvious. That some registrars would go the next step and use the data for direct profit is a given. No sane person can claim it's not happening.

How widespread it is is definitely a matter of debate but I don't see how there can be any argument that it happens/has happened at some registrars.

Paraphrasing the thread description I must say: It is totally unreasonable to doubt it.

It happens, period. I don't need proof. I believe on this happening since 3 years ago in minor scale, because it happened to me.

I thought the object of this thread was about investigating on HOW this was made, because there are many ways of doing this. But the excessively puritan rules on WebmasterWorld make this kind of test prohibited.

Hmm

I used to work for a Registry and out CTO and the Senior Sponsors for the Sponsoring organisation - where Absolutly sure that this happened.

If it is going it's a waste of time. People realize that the domain name may be taken these days and they also realize that they may have to pick others or be creative.

Very few people will pay $100 or more for a domain name that somebody is parked on.

All somebody does that buys domain names and then sits on them is waste their money.

Very few people will pay $100 or more for a domain name that somebody is parked on.

I just paid 800€ for this very case.

It's amazing how so many of you are totally convinced this is happening in spite of basically no evidence of it being true.

Webwork had challenged you to do a Whois on lists of names and report back how many were reg'd but few have done that after all this time, why? I think the answer is obvious! One of you did report and said only 1 out of 200 was taken.

Most experienced domain people know this is not an issue and as a domainer with 10-yrs experience in the trenches can also say it is not. With that said, it's always possible a 2nd rate registrar or rogue employee may see a Whois inquiry on a name they like and grab it but that would no doubt be a rare event if it happens at all.

So you are not doubting that it is happening at all, but rather the scale on which it happens?

It's amazing how so many of you are totally convinced this is happening in spite of basically no evidence of it being true.

I just reread the whole thread. 20 different people saying it has happened to them personally = evidence.

"no doubt be a rare event, if it happens at all."

P.S. Looks like I need help from others or I will be ganged-up on and flamed real soon! Where are you Webwork and gpmgroup?

Webwork had challenged you to do a Whois on lists of names and report back how many were reg'd but few have done that after all this time, why? I think the answer is obvious!

yep ..with the great respect I hold for webwork and the regular members of this fora ..I have better things to do with my time than to futilely test something which as I have stated I had done to me by networksol nearly a decade ago ..

I have since used Ev1 ( an open srs reseller )..been an open srs reseller myself and used godaddy ..never seen it happen since with either of those two ..the latter I would recommend highly ..

spending time testing things for the benefit of others who are not known to me isn't gonna get on my list of urgent to do's anytime in this incarnation ..

@trader..

you've been given indicators whom to avoid ..

you dont want to beleive ..thats fine by me ..

but don't call my experience and word into doubt ..based upon your personal criteria and experience ..

accepting that YMMV is the key to getting on and staying amigos ..

Webwork had challenged you to do a Whois on lists of names and report back how many were reg'd but few have done that after all this time, why? I think the answer is obvious!

Yes, the answer is obvious--though not for the reason you're implying. It's because we don't know the criteria according to which the decision is made to take one name and not another.

If this happens at all, and if not every single queried name gets registered, then one of two things must be happening:

1. lists are assembled--we don't know the mechanism--and then human-reviewed, or
   
2. lists are assembled and then programmatically evaluated
   

Either way, some domains meet some (unknown) criterion or criteria. In the case of our client, the most obvious suspect in this regard was an existing .org version (which they already own) of the .com they queried--and yes, we are quite sure the .com has not previously been registered.

Just as a mental stimulant to the naysayers present, imagine how useful an sql query such as "SELECT * FROM domains WHERE name LIKE %searchresult%;" could be if you had access to lists of both existing domains and incoming searches...

-b

...but don't call my experience and word into doubt ..based upon your personal criteria and experience .. accepting that YMMV is the key to getting on and staying amigos...

I never doubted that so why say I did? I knew I would get flamed soon! What is YMMV anyway?

Where is Webwork and gpmgroup as I can't hold down the fort by myself much longer?

1. These are hardly flames.

2. I already quoted your dismissive statement that the various preceding personal accounts amounted to "no evidence". I don't really know what kind of evidence would satisfy you, since you've constructed an inpenetrable worldview in which monitoring queries is assumed not to happen, and if any domains DO get scooped up it can be explained away as random chance.

I never doubted that so why say I did?

you did say here ..

It's amazing how so many of you are totally convinced this is happening in spite of basically no evidence of it being true.

Inspite of the personal example I gave early in thread ..and inspite of the personal examples ( not speculations ) of others who have been in this business for a long time ..

here you presupposed that you knew why so many of us had not done the test ..

but few have done that after all this time, why? I think the answer is obvious!

I presume they also have better things to do with their time than to show you and others via elaborate and to they and myself , non productive tests ..

YMMV is ..your mileage may vary..meaning that in this case it isn't because you personally have no experience of this happening ..that it didn't happen to myself and other posters in this thread ..

your post was dismissive to say the least of our input ..however I didn't flame you :))..just pointed out how your post comes across ..

when I actually flame people ( which is rare ) ..I am not known for my politesse in the particular post ..;-)

however gently admonishing folks has been known to get thebear to send smilies..:)

edited ..typos

[edited by: Leosghost at 6:20 pm (utc) on Nov. 13, 2006]

I think the criteria is mostly based on keyowrd searches.

May be overture/google/msn etc combined with the PPC revenue being generated by these engines per click.

these scripts obviously cant be manual. The criteria after getting the list of queries would be to mix and match the domain names queries with a predefined set of keyowrds and per click revenue the industry generates.
then you run the script and whichever shows up on the green list, register them.

Green list: lets say you look for a skill gaming domain as thats the big thing.
Skill gaming pays well, let it be Xbox or PSP or just plain online websites offering skill games.
you see its avialable and dont register it immediately.
In a few hours your query is passed on by one of the shaddy registrars to the domain tester (those questionalble companies whose name keeps pooping up in every discussion we have on this topic),.

Now they run the script.The predetermined scripts will read through the domain name flag the important key words skill and gaming ..check on the overture search count ...and lets say (hypothetically) if its above 1000 , register the domain for testing.
If its not move on to some other domain.

thats purely someone doing it for domain testing purposes. But I beleive it is possible if even one of the 660 (if i got the number right) credited registrars is willing to sell the information.

Dose not sound like to far fetched to me.
Or may be its time to stop working for the day.

BTW I agree with the guys who say its happening :)

Nick

The problem with these kinds of threads is they are statistically very unfair on the bigger registrars and therefore give a very misleading impression to newbie's that they are more likely to be spied on at the larger registrars.

GoDaddy has ~ 13,500,000 domains
Net Sol ~ 6,700,000
Enom ~ 5,500,000
Tucows ~ 5,300,000
Melbourne IT ~ 4,100,000

Each of these registrars will have thousands of more clients than the smaller registrars therefore statistically they are more likely to have a customer who at some time for whatever reason who has been beaten to the draw so to speak.

Given in any day up to 3,000,000 domains are tasted and most people who speak English know 10 - 20,000 and only use a fraction of those words (the rest solely being used for recognition or recall) it is statistically very likely that some names will registered in between checking the WHOIS and subsequently rechecking (to buy). Obviously the longer the period between the initial check and the subsequent check and the more generic the domain (most of the tasted names are for PPC/Traffic) the more likely statistically this is to happen.

Having said all that; any good business person will keep their cards close to their chest. Even discussing the next Google in a bar is a bad idea as somebody, maybe even one of the bar staff might overhear and decide to act on it.

Is the risk statistically significant?

My guess is in the main it's highly unlikely to happen and the bigger the Registrar the safer you are. And to a certain point Webwork's test has borne this out though it would have been very entertaining if a registrar had been caught out :)

That's not to say it doesn't ever happen and the more people involved in any chain the more likely there is to be leak somewhere along that chain. In the end it's all a matter of odds and to minimise the risks the closer you are to the source and the fewer people both off line and on line you share your future Google with the less likely you are to be pipped to the post.

My results:

The first bunch I tested for this exercise - only one was registered, and it was registered by the same "Maltuzi Holdings" that had gotten three I was looking at earlier.

I also tested a group of sports related domains for the exercise - probably 150 to 200 domains, and around 20 to 25 have been registered (none were registered when I first checked). Every day I check, it fluctuates - I think around 12 or so are permanently registered, and others are cycled through the tasters. Of those registered, around 90% of them are being registered by this same Maltuzi Group. Again, I say around 90%, because every day I check, the number fluctuates.

Friday, I checked 5 "Iron Bowl" (the Auburn - Alabama football game) related domains, and this same company had registered 4 of those 5 by Saturday. Truthfully, I don't see how they could spy on my "domain searches" as I did them from a registrar and not a third party "suggested domains" site - I think that whoever or whatever is registering the domains at Maltuzi is on the same wavelength as I am - we are tracking the same sort of domains. So, when I saw the one I was going to use was available, I grabbed it instantly.

On another note, one of the original domains that I had insisted had been "idea hijacked" became available again! Since I first checked and lost it near the end of October, it has been registered by 3 different "domain tasters". I wasn't fast enough to get it myself, so even though the domain expiration date wasn't until next year, I put a Godaddy backorder in on it, and got it yesterday when the "taster" let it go!

I found that that particular domain had been registered by 35 different "owners" in the past 4 years. Here is some info on it:

Registrar History: 3 registrars.
* NS History: 35 changes. Using 17 unique name servers in 4 years.
* IP History: 23 changes. Using 10 unique IP addresses in 2 years.

I imagine that this is unusual for a domain. Interesting stuff! has anyone else seen this type of activity on one domain?

[edited by: Webwork at 3:42 pm (utc) on Dec. 17, 2006]
[edit reason] Charter [webmasterworld.com] [/edit]

Webwork has been in LasVegas all week.

I'm baaaaaack! ;-P

Trader, I think you make a fairly good point that the "challenge" has not been met, at least not in a big way. If the practice of "sniffing" was as widespread as the rumors I imagine we'd have quite a few "me too-s" popping up.

Where are they?

So: Rumor > Fact by a long shot. Some evidence, but mostly anectdotal.

Rumor > Fact by a long shot.

Maybe. This thread has amply demonstrated that it can and does happen. But whether there are rumors of it happening on a far greater scale than in actuality, I don't know. I haven't heard those.

Since I first checked and lost it near the end of October, it has been registered by 3 different "domain tasters".

How do you check previous owners in the first place - apart from archive.org?