ICANN announced today that since they received over 500 applications they are triggering the use of “Digital Archery” as part of the batching of applications. The last report I saw stated that there were 839 applicants.
It had been stated that they weren’t going to use a first come, first served method and also that they didn’t want for it to be a lottery of any kind.
I had been struggling for some time to understand why what they were going to do. Now that things have played out I can see some of the reasons they didn’t want to do those things. If it were first come, first served, there wouldn’t have been a fair way to deal with same string applications that came in later. If it had been a pick from the hat, it could have been judged an illegal lottery. Both of those methods have problems.
To put it as simply as possible, this method first gives applicants a way to opt out of the digital archery method if time isn’t important to them. Then involves having the remaining applicants participate in process where they…
- select a time date in the future
- try to ping a server as close to that time as possible
The differences in time they picked and the time they pinged the server to match that initial pick (regardless of before or after) are then compared and ranked. Then any matching strings are grouped with the winners regardless of how they did in the competition so they can be judged together. There is also a locational element to make it fair for applicants from around the globe.
It seems like a game to me, but I guess business can be like a game very often can’t it?
You may review the announcement here:
This makes it clearer…
I can’t believe it! It’s really like playing darts (at least part of it)!
Geographic regions explained more here…
Here is a quote directly from the ICANN announcement:
The batching selection process determines how applications will be divided into batches and prioritized for evaluation analysis. Conceptually, the batching selection process is relatively straightforward and includes the following four steps:
- Applicants register in an online batching system to select their batching preference (i.e., earliest or any batch) and select a target date and time (e.g., Target Date: 10 May 2012 and Target Time: 12:00:00 UTC);
- Applicants re-enter the online batching system and generate a message that is sent from their computer/system to the online batching system. The online batching system records the date and time the applicant’s message is received. (e.g., Message Received Date: 10 May 2012 and Message Received Time: 12:00:01);
- The system calculates the time variance between the applicant’s Target Date/Time from step 1 and the Message Received Date/Time from step 2. This time variance is known as the applicant’s “secondary timestamp” Based on the example in steps 1 and 2 above the secondary timestamp is 1 second. The closer to zero the secondary timestamp is the more likely the application will be processed in the earliest batch, assuming the applicant has opted in to the earliest batch.
- The batching selection process then combines the applicant’s batching preference (i.e., earliest or any batch), the “secondary timestamp” (e.g., 1 second), and the geographic region to determine the batch/processing order for the specific application.