Internet Performance Delivered right to your inbox

Wither WHOIS!: A New Look At An Old System

No, that title is not a typo. The WHOIS service and the underlying protocol are a relic of another Internet age and need to be replaced.

At the recent ICANN 43 conference in Costa Rica, WHOIS was on just about every meeting agenda because of two reasons. First, the Security and Stability Advisory Committee put out SAC 051 which called for a replacement WHOIS protocol and at ICANN 43, there was a panel discussion on such a replacement.  The second reason was the draft report from the WHOIS Policy Review Team.

This is hardly the first time there has been hand-wringing about WHOIS, especially at ICANN. So what’s all the noise about now?

What is WHOIS?

To understand why we have WHOIS at all, a little history is needed.  In the ancient pre-history of the Internet was a network called the ARPANET.  It was an experimental network and as you might imagine, an important part of running an experimental network is being able to get in touch with the people participating in the experiment when something goes wrong.

Initially, the contact information was maintained at the Network Information Center, and over time, it migrated online. It appeared in the NICNAME/WHOIS service and the protocol was published in RFC 812 in 1982.  To give an idea of how long ago that is in Internet terms, the ARPANET didn’t officially transition to TCP/IP and DNS didn’t exist until 1983.

Because WHOIS was really intended to be a service devoted to finding people’s contact information when one needed to reach them, it was also a service designed to be consumed by humans.  This made for a very simple protocol with free-form text in replies. In the 1990s — when our contemporary domain name management system came to be with ICANN, registrars, registries, and billions of people online — WHOIS came along for the ride.

People started using the term “WHOIS” to mean the protocol, but also the service (which is sometimes delivered as, for example, a web page), and even the data that you can get out of the service.

The registration data for domain names can be useful.  Different parts of the data are useful to different people, but WHOIS cannot make those partial distinctions.  Also, WHOIS is anonymous, so not only does everyone get the same data, but the WHOIS service doesn’t even know who asked for the data.  Because of that, many people who value their privacy simply lie when they enter registration data.  That way, their phone numbers or street addresses can’t be looked up by just anyone on the Internet.

A different environment, a different tool

The Internet has evolved considerably since WHOIS was specified and we have different problems than we did in those days.  On a network (like the ARPANET) where it was at least theoretically possible to get a list of every person on it, things like spam were not a problem.  Today, we need to be able to tell whether several domains are controlled by the same person in order to combat mail abuse.

And while it might be perfectly appropriate for law enforcement to be able to get your street address under the right circumstance, it isn’t clear that your address needs to be published for more than two billion people to see just so that you can have a domain name. Solving these sorts of problems will be impossible if the Internet community doesn’t settle on a new data access protocol without the limitations of WHOIS.

This is why Dyn was pleased to express our support for the SAC 051 recommendations and a plan to implement them. Dyn Labs is working on prototype versions of WHOIS protocol replacements so that once a new protocol standard is ready to go, we can move quickly to replace the old, less useful service with a new one.

Work is just getting started and at this week’s IETF meeting, we hope to take another step on this path. We hope others in the ICANN and IETF communities will also work on making this much-needed improvement to the registration landscape.


Share Now

Whois: Andrew Sullivan

As Tech Evangelist at Oracle+Dyn, Andrew Sullivan concentrates on the architecture of Internet systems, working on Internet standards, protocols and policy. In his more than 15 years in the business, he has focused on DNS, Internet standards and internationalization, systems architecture and databases, but not always that order. Follow us and like us to get involved in the conversation.