Minutes of the Registry Information Database Exchange Formats and P (RIDE) BOF
Reported by: Dhaval Shah
The meeting started with David Kessens describing the motivation for RIDE and a presentation that covered the proposed charter including the schedule for the goals and milestones.
Rick Wesson followed with a brief presentation on the internet draft that describes the various RIDE classes. Some people felt a bit uneasy about the fact that the classes were represented by using a C++ representation during the presentation. It was clarified that the draft itself didn't use this representation. A question was asked if the representation in the draft could be more clear on required, optional, and/or single valued fields. It was explained that this was planned for the next version of the draft.
David explained the relation of RIDE to COREdb, since it was felt that there were some common issues that need to be worked out. We had a lively discussion and decided that we should strictly focus on the data, dataformats, and protocols to be used for exchanging data between IP, domain and routing registries. It was felt that the RIDE standard could and should be used by the COREdb if we took this approach. The RIDE standard would automatically cover COREdb since it is just a particular instance of a domain registry.
The next agenda topic was NIC handles and how to deal with them. David did a strawman proposal to define a NIC handle as the second level domain of the .NET gTLD with certain information stored in an associated TXT record to be able to find the associated information. Another idea was to let IANA assign the suffixes or maintain a 'registries.int' domain that would store the possible suffixes. The discussion that followed made clear that we were in fact trying to solve references and NIC handles at the same time, while we should, in fact, take a step back and look at references in general and NIC handles are just a specific case. We decided to address references in the charter and be more explicit in our boundary conditions like that we are looking for the best solution that can work with the current registry technology since we assume it to be fixed and to a large extent unchangeable.
Then, Masaya Nakayama held a short presentation on the whois servers that are currently in use by 17 Internet registries. He found that four registries are using the InterNIC formats, seven the RIPE format, and six their own private format. Also, seven servers had support for local languages, which made it clear that we cannot easily ignore multiple language support.
In order to start the work on the dataformats and protocols, we had a short discussion on how to choose or design a protocol and format to exchange the actual data. It was felt that we should choose an existing technology, if possible.
Finally, it was proposed that David would become the chair of the working group. David asked if Dhaval Shah could become a co-chair in order to share the workload a bit. The working group agreed to this. mib)
3. Globally Unique NIC Handles