- Joined
- Feb 5, 2008
- Messages
- 117
- Reaction score
- 5
Due to the nature of the industry/process/niche, dropcatchers of expired domains typically hunt alone. Obviously due to the fact that we don't want other predators competing for the domains we researched and queued.
As many of us know, DNMeter was created a few years ago by expron and supposedly delivers the exact drop times of all .net/.com domains. Through further research I was able to determine how these drop estimatations are done. Much easier than I had originally thought possible, it is simply a matter of all .com/.net pendingDelete domains dropping within a 75 minute window.
On an average day you have 75000 dropping .com/.net domains. To obtain creation dates of all domains on this list takes my VPS only about 4 hours per list. My sites still run quickly while the script is running--so while I'm content with this number, I could surely speed it up significantly with proper resource allocation. With 75000 domains dropping in just 7500 seconds, you'd simply need to sort the domain list in order of creation date (ASC), then find the incremental value between each drop. The magic number here would be .06 seconds. 4500 / 75000 = 0.06. That means there would be 6 hundredths of a second between each drop if my math is correct. I think it would be safe to say the domain would almost indefinitely drop +- 10 seconds from our estimated drop time. So with this many domains (assuming they must drop in that order), we can virtually pinpoint the exact second of the drop (barring registry hang-ups, lag, and stalling).
How does this help us as dropcatchers? It doesn't necessarily if we are just using drop-catching software on our servers/desktops and running it for the entire duration of the drop window. Sure it would allow us to find out if our acquisition was successful almost immediately and move on to the next one somewhere further down the drop list. But let's think on a grander scale.
Namejet is widely revered as the auctionhouse most successful in landing back-ordered pendingDeletes when you want one bad enough to pay 60 bucks for it. They obviously have 2 enormous registrars, Network Solutions and eNom, responsible for their success in domain acquisition.
Switching gears, let's think about torrents and shared downloads. There are many many registrars out there, not affiliated with Namejet or Snapnames or Pool that we can use API to register domains with. Now what if we ran a torrent-like drop-catching program/membership program that allowed us to leverage the resiyrces of many many different registrars to obtain our domains.
After receiving a request from the server, all drop-catchers would then launch constantly ping <em>various</em> registries using your API registration URL (which would be undisclosed and unhackable on a perfect internet) for exactly 15 seconds, increasing the probability of you landing that domain by exponential odds. Obviously a lot of registrars have stipulations regarding whitelisting and blacklisting of IP addresses, so a work-around would have to be performed for this... though I certainly do not see it as a deal breaker.
In my many years of studying drop-catching and attempting to program the fastest drop-catchers out there, I've never come close to competing with the big three (NJ, Snap, Pool). However IF these aftermarket companies are all playing by the same rules we are and when a domain drops it truly is fair game for all, then there is hope. Hope in the form of domainers being a little less secretive and having a little more "open-source" mentality to share their knowledge and findings with the rest of the community.
As many of us know, DNMeter was created a few years ago by expron and supposedly delivers the exact drop times of all .net/.com domains. Through further research I was able to determine how these drop estimatations are done. Much easier than I had originally thought possible, it is simply a matter of all .com/.net pendingDelete domains dropping within a 75 minute window.
On an average day you have 75000 dropping .com/.net domains. To obtain creation dates of all domains on this list takes my VPS only about 4 hours per list. My sites still run quickly while the script is running--so while I'm content with this number, I could surely speed it up significantly with proper resource allocation. With 75000 domains dropping in just 7500 seconds, you'd simply need to sort the domain list in order of creation date (ASC), then find the incremental value between each drop. The magic number here would be .06 seconds. 4500 / 75000 = 0.06. That means there would be 6 hundredths of a second between each drop if my math is correct. I think it would be safe to say the domain would almost indefinitely drop +- 10 seconds from our estimated drop time. So with this many domains (assuming they must drop in that order), we can virtually pinpoint the exact second of the drop (barring registry hang-ups, lag, and stalling).
How does this help us as dropcatchers? It doesn't necessarily if we are just using drop-catching software on our servers/desktops and running it for the entire duration of the drop window. Sure it would allow us to find out if our acquisition was successful almost immediately and move on to the next one somewhere further down the drop list. But let's think on a grander scale.
Namejet is widely revered as the auctionhouse most successful in landing back-ordered pendingDeletes when you want one bad enough to pay 60 bucks for it. They obviously have 2 enormous registrars, Network Solutions and eNom, responsible for their success in domain acquisition.
Switching gears, let's think about torrents and shared downloads. There are many many registrars out there, not affiliated with Namejet or Snapnames or Pool that we can use API to register domains with. Now what if we ran a torrent-like drop-catching program/membership program that allowed us to leverage the resiyrces of many many different registrars to obtain our domains.
After receiving a request from the server, all drop-catchers would then launch constantly ping <em>various</em> registries using your API registration URL (which would be undisclosed and unhackable on a perfect internet) for exactly 15 seconds, increasing the probability of you landing that domain by exponential odds. Obviously a lot of registrars have stipulations regarding whitelisting and blacklisting of IP addresses, so a work-around would have to be performed for this... though I certainly do not see it as a deal breaker.
In my many years of studying drop-catching and attempting to program the fastest drop-catchers out there, I've never come close to competing with the big three (NJ, Snap, Pool). However IF these aftermarket companies are all playing by the same rules we are and when a domain drops it truly is fair game for all, then there is hope. Hope in the form of domainers being a little less secretive and having a little more "open-source" mentality to share their knowledge and findings with the rest of the community.
