Published: March 15, 2001.
A Top Level Museum Domain
Cary Karp, Swedish Museum of Natural History, Sweden, Kenneth Hamma, J. Paul Getty Museum, USA
In November 2000 the Museum Domain Management Association (MuseDoma) was awarded the opportunity of introducing a new top level domain (TLD), museum. The Internet Corporation for Assigned Names and Numbers (ICANN) approved seven applications for new domains at that time. With the support of ICOM and the Getty Trust, MuseDoma expects to bring this domain into operation the first week of July 2000 to benefit the museum community world wide in the online environment. We explore here a short history of the Internet and its naming conventions that led to ICANN's actions, and look at the potential benefits to the museum community of a museum TLD.
Keywords: ICANN, ICOM, Getty, Karp, Hamma, TLD, top level domain, museum, domain name system, DNS, Remunere, MuseDoma, Muse Doma, Museum Domain Management Association, .museum
May I first note: the printed version of this paper was submitted two months ago on January 15. Significant changes in the since then will be reflected in the version delivered here today.
By way of introduction, I would like to start with a little history of Internet technology management -- at least as understood by a classical archaeologist such as myself. This may be far too elementary for many of you, and may be fully unknown for others coming from parts of the museum community that are able to focus on content rather than connectivity issues. I think it is essential to have some common sense of the short history of the internet and how that history has been characterized by radical change before jumping into the murky waters of expectations for the future with new TLDs, in particular with the museum TLD. Following this I will tell you what we understand to be the latest on various fronts. And this will likely differ from any previously submitted version of this paper since we are involved here in a process that brings weekly changes and has not yet come to conclusion. Finally, because this covers such a vast scope of issues and developments arising from museums as well as the Internet, I want to leave a good space of time to answer questions.
What is DNS?
There is something we use everyday called the Domain Name System (http://www.iana.org/domain-names.htm ), created in the Internet's precursor, ARPANET (the Department of Defense's Advanced Research Projects Agency Network), to handle the easily recognized names we use on the internet daily, etoys.com, for example, and to map those names to unique IP addresses, which many of you are also familiar with as a period-delimited set of four integers, 126.96.36.199, for example. This system began life pretty much as it is today but with a single person the administrator of names, Jon Postel, who eventually handed those duties off first to SRI International and then to IANA, the Internet Assigned Numbers Authority. At the outset the system supported several generic top level domains or gTLDs and a set of about 200 two-letter country-code top level domains, or ccTLDs. These designations are on the first level, the farthest to the right in a URL.
The Domain Name System is maintained exclusively as an enforceable scheme on a set of Root Servers (http://www.wia.org/pub/rootserv.html ) which have a part to play in every bit that travels across the Net. In order to ensure consistency, the primary name set is created at Root Server A, run by Network Solutions Incorporated for the National Science Foundation. This name set if periodically propagated to the other root servers.
The generic domains (http://www.iana.org/gtld/gtld.htm ) are mostly well known like .com, .net, .org. Note here the restrictions on the edu TLD: "reserved for educational institutions in the United States granting four-year degrees." These restrictions were not recently created but were part of the restricted domain at its inception.
The country codes (http://www.iana.org/cctld/cctld-whois.htm ) are based on the ISO-3166 list of two-letter country codes. These have been variously managed or completely neglected since their inception. Some, such as TV, have been heavily marketed by their owners in the last year as useful letter combinations that have nothing to do with their countries. And based on the success of that marketing effort, we are now seeing additional ccTLDs, such as LA.
In 1991-92 the National Science Foundation (NSF) undertook coordinating and funding the management of the non-military portion of the Internet. In 1992 NSF contracted with Network Solutions, Inc. (NSI) to provide many of the services (hence NSI's current ownership of RootServer A) and, just as significantly if not more so, was granted by Congressional statute the authority to provide for commercial activity on NSFNET.
With the explosion of Net activity in the 90s this theretofore closely held 'research Net' suddenly became a very heavily used and widely distributed 'commercial Net' expected to accommodate much more. The distance between the two quickly became an arena for lots of complaints. The most visible because they had a clear and ever present effect on end-users were complaints about band-width. Less visible but perhaps more urgent were complaints about management and governance:
In 1996, John Postel, the author of the initial proposal for the DNS, proposed expanding the name space. His initial proposal was large and far-reaching. But with increasing commercial value in names and in some quarters concern for the scalability of the DNS, the decision to add new top-level domains was quickly reigned in. After review by many agencies and the solicitation of much public comment, the scope of the proposal was greatly reduced. It continued however with several of the original goals:
This led more or less directly to the process that the museum TLD has been involved in since last August.
As you know the application that was submitted for the museum TLD was submitted to ICANN (http://www.icann.org/), the Internet Corporation for Assigned Names and Numbers. It was submitted on October 2 and decisions on new TLDs were announced at ICANN's board meeting on November 16, 2000.
What is ICANN? Where did it come from?
Beginning in May 1997 and continuing to the November 1998 Memorandum of Understanding (http://www.icann.org/general/icann-mou-25nov98.htm ) between ICANN and the US Department of Commerce the DoC created ICANN as a non-governmental agency to take over its Internet management affairs. This is in many ways the end result of the initial adoption of Internet coordination through NSF. The change involved primarily the transfer and consolidation of authority for Internet governance in ICANN which is composed of a board of directors, an at-large membership and a string of supporting organizations. During the last three years, the Internet world has seen a significant changes in the management of the affairs of the Internet along these lines. The current re-assignment of the Canadian ccTLD from the University of British Columbia to the Canadian Internet Registration Authority, a government agency, is similar if on the opposite public-private axis. All of these developments are not without continuing challenges and difficulties, including fundamental questions about the Department of Commerce's ability to have created ICANN initially. So the process of creating ICANN continues as a current process. And it is important that this be understood since it is having some impact on the more visible process -- creating new TLDs.
At the ICANN meeting in Yokohama in July of 2000, the discussion of new TLDs got underway in earnest, being the only agenda topic aside from organizing the at-large membership.
A useful graph with detailed timeline of these governance developments beginning in 1994 is available online at the World Internetworking Alliance (http://www.wia.org/pub/dns_governance_timeline.gif ). You can see there the ICANN meetings beginning at the eleventh hour. But the changes that have taken place in the last few years -- at least one aspect of them -- is cogently expressed in two images that have been posted at the World Internetworking Alliance:
Before ICANN (http://www.wia.org/icann/before_icann.htm ): this is Jon Postel and Joyce Reynolds and was for years the extent of the assigned names business on the Internet.
And After ICANN (http://www.wia.org/icann/after_icann-gac.htm ).
Very different. And one of those comparisons that might make one yearn for the good old days. Except, like most good old days, they would be completely inappropriate and non-functional in the world as we know it.
Where did the .museum initiative start?
Cary Karp (home.pi.se/karp/) was at the ICANN meeting in Yokohama, listening in, schmoozing, learning, and advocating for the cultural heritage sector. And much to his surprise .museum was mooted by others at the Yokohama meetings as the sort of thing that the ICANN board envisioned and thought appropriate for the expansion of the name space, particularly in the arena that has come to be called special purpose TLDs. (The special purpose TLDs that were approved by ICANN include .aero and .coop as well as .museum.)
Partly out of consulting with the International Council of Museums (ICOM) for many years, partly out of strong personal commitment (www.remunere.net), Cary had been thinking about a top level domain that would serve the museum community, and more broadly the cultural heritage community since the expansion of the name space was proposed in 1996. The cumulative result of that, including listserves and lots of comment from around the world, can be found on this site: REMUNERE, Registry of Museum Network Resources. ICOM early on was recognized as the only possible partner and policy companion for such an effort, being the only world-wide NGO representing the interests of museums. The ICOM definition of museum remains a key element of policy for the .museum initiative. We may want to return to that later because of its pivotal role, but also because of the implications of .museum for the ICOM task force that oversees the 'maintenance' of the definition. In general I do not expect that the policy street from ICOM is or will be a one-way street since the growth of the importance of online resources in all sectors will certainly motivate changes in the scope and purpose of all such international representative organizations.
During the summer of 2000 at the Getty, after many discussions of the potential value of a museum domain, we decided that we thought it an effort well worth supporting. A quick and non-scientific poll of colleagues seemed to confirm that feeling. That was the last week of July 2000. The next week, first week of August 2000, I was in Santa Fe (http://santafeopera.org/index2.php) on vacation connected to email through a cell phone. So at 9600Kbps Cary and I were put in touch with each other by Kathleen McDonnell and quickly discovered that the common interest in a museum domain was a good thing, that is non-competitive, and that details could and would be eventually worked out.
The week after that, the second week of August, ICANN finally posted the forms and rules for applications (http://www.icann.org/tlds/tld-applications-15aug00.htm) for new TLDs with a deadline for applications only six weeks away at the beginning of October. One week later, the third week of August, we all met in Ottawa at the CIDOC meeting to strategize and make specific plans for creating, writing, reviewing and submitting the complex eight-part application due a month later. September was full of those activities in Stockholm, Paris, Los Angeles and Geneva leading up to the application submission on the second of October. Key in our thinking was maintaining this effort as one made on behalf of the museum community, and as one that could eventually be expanded to include other cultural heritage sectors, depending on what the first round of new TLD applications looked like. We did think that there might have been an application on behalf of libraries.
There has been some misunderstanding about the role of ICOM (www.icom.org), as there has been about the role of the Getty in this. More about that in a minute. But it is important to note here the critical role the ICOM plays, the eagerness with which they have embraced that role and their generosity in helping during the critical early months of the initiative. ICOM is affiliated with UNESCO but was created as an independent NGO in 1946. It currently has about 15,000 institutional and individual members world wide.
Which led to the selection of new TLDs (http://www.icann.org/announcements/icann-pr16nov00.htm ) at the ICANN meeting in Los Angeles in mid November.
New TLDs like biz currently seem -- or maybe even seemed -- to fill a clear need. But the internet is nothing if it is not change. With e-toys recently admitting that there may ever only be and ever need only be one online retailer of toys, the dot com look-alikes may seem a little less cutting edge.
Who owns the museum TLD?
As part of the application process for a museum TLD, the Getty and ICOM became founder members of a new non-profit organization (http://www.musedoma.org) incorporated in the state of Delaware and registered in the US as a 501(c)6. There are good reasons based on the quantity of United States case law for Internet governance and disputes to be incorporated in the US. We can come back to this too, if anyone wants to ask additional questions about the US incorporation. This does not mean that operations need be or is envisioned to be in the US. In fact, most activity now is in Stockholm and Paris. And so the Museum Domain Management Association -- not Getty and not ICOM -- was the sponsor of the museum TLD application, and the Museum Domain Management Association -- not Getty and not ICOM -- is the trustee of the domain.
MuseDoma, for short, will have other members, both museum and non-museums, though only museum representatives may be elected to board seats. MuseDoma has a board of six seats: two belong to ICOM and the bylaws see that in perpetuity, two belong to the Getty and the expectation is that those will be released when we are satisfied the MuseDoma has achieved stable independence, two are empty. The primary responsibility of the organization is to be the trustee of the museum domain and to set policy that affects its operation in serving the world-wide community which MuseDoma was founded to represent.
A 501(c)6 organization is a non-profit trade association, that is it exists to serve its non-profit members; and as a non-profit itself, revenue can only be used to support the goals and purposes of the association.
Other current affiliates include:
CORE (www.corenic.org) , the Internet Council of Registrars, who is proposed at the technical registry. CORE is also a not-for-profit membership organization which would like to provide a shared name registration system, and
The Swedish Museum of Natural History (http://www.nrm.se/welcome.html.se), which is currently providing in-kind support through overhead costs, staff and services.
Continuing the spirit of the Internet, all documents to date are available -- and have been available through the duration of this process on MuseDoma's Web site, (www.musedoma.org). These include the entire application to ICANN, the incorporation papers and bylaws of MuseDoma, and all supporting and related documents.
These Web pages also include a set of Frequently Asked Questions (http://www.musedoma.org/faq.html) that are added to and updated periodically.
There is as well (http://listserv.musedoma.org/archives/musedoma-questions.html) an on-going discussion of topics. In short there is a lot of information here. There is the latest information here. And here again, let me say we are involved in a process for which, given ICANN's own history, even the parameters may change weekly.
We hope and expect that MuseDoma will be a financially viable and financially independent corporation within two years of accepting registrations.
In the emails that have arrived and in the discussions that have ensued after the ICANN meeting in November, three questions have come up with some frequency.
When can I register in the museum domain?
ICANN suggested last November that the new TLDs would have their contracts with ICANN in place by the end of 2000 -- or that at least the first of these would be in place. It is important to recall here the recent history of the creation of ICANN and the governance changes that are taking place. This is among the things that has considerable slowed the process of contract negotiation. Assuming that this will be resolved and not continue to delay the process of rolling out the new TLDs, MuseDoma is aiming for initial registrations the first week of July 2001 in conjunction with the triennial ICOM meeting in Barcelona. One need not be at the meeting to register, obviously.
The motivation for this conjunction with the ICOM meeting is important however. Given current facility to market to museums, it is relatively easy to reach EU and North American museums. Exceptionally strong and active national associations exist in the US, Canada and the UK. It is more difficult to effectively reach museums elsewhere; in some parts of the world exceedingly more difficult. And in the long-term value we see for the domain, it is exactly those museums -- the difficult to reach, the barely visible -- which we want most to reach, for which we want most to create a welcoming community online and for which we want to help provide visibility.
Can I register mymuseum.museum? What are the naming conventions proposed?
The primary criterion for registration is being a museum by the ICOM definition (http://www.icom.org/statutes.html ), as well as the many professionals and professional membership associations that serve the museum community. Membership in ICOM or any other professional or regional organization is not a criterion for registration.
"A museum is a non-profit making, permanent institution in the service of society and of its development, and open to the public which acquires, conserves, researches, communicates and exhibits, for purposes of study, education and enjoyment, material evidence of people and their environment.
(i) natural, archaeological and ethnographic monuments and sites and historical monuments and sites of a museum nature that acquire, conserve and communicate material evidence of people and their environment;
(ii) institutions holding collections of and displaying live specimens of plants and animals, such as botanical and zoological gardens, aquaria and vivaria;
(iii) science centres and planetaria;
(iv) conservation institutes and exhibition galleries permanently maintained by libraries and archive centres;
(v) nature reserves;
(vi) international or national or regional or local museum organizations, ministries or departments or public agencies responsible for museums as per the definition given under this article;
(vii) non-profit institutions or organizations undertaking research, education, training, documentation and other activities relating to museums and museology;
(viii) such other institutions as the Executive Council, after seeking the advice of the Advisory Committee, considers as having some or all of the characteristics of a museum, or as supporting museums and professional museum workers through museological research, education or training."
Virtual museums (http://www.diarioelpais.com/muva2/) like this one run by Alicia Haber in Montevideo are not specifically included in the current ICOM definition -- and must be considered viable registrants in the museum domain. This is a discussion that ICOM is having now, and one example of the ways in which the domain will likely impact this arena of policy.
We would like to permit institutions to register on the second level -- next to the left from the first level -- but see no viable scheme for this at the moment. I expect museums will be able to request second level names at the outset, but we will register individual institution names on the third level with all second-level designations being generic or regional. We can see that .org on the second level would be useful, also some designation for virtual museums; and registrants may propose additional second level names which if adopted would be available for all registrants.
We are also aware of the proposals for multilingual character sets in the Domain Name System and expect that the museum domain will be among the early adopters of non-Latin character sets to work in parallel with each other and with the Latin, museum.
We have not set aside nor yet see the need to set aside any reserved strings. But, in order to make sense of the domain, registered names for museums will have to have their origin in corresponding organizational names and locations.
What is the value of a museum TLD?
This is a question that each of us will answer for ourselves. It reminds me a little of a moment years ago. A small and arcane group I belonged to -- PhD candidates in Art History at Princeton -- would get together now and then for a pot-luck supper. A student from Cuba volunteered every time to bring menudo, but had the grace to ask first if the rest of the group liked tripe. As surveys go, that was an easy question. The surprising part was that opinions varied.
I expect opinions will vary on the value of the museum domain. But I also see that it is not one of the easy questions. If you read through the comments made on the museum TLD on the MuseDoma Web site (http://listserv.musedoma.org/archives/musedoma-questions.html) or on the ICANN Web site (http://forum.icann.org/tldapps/) , where you can read also about the other new TLDs, the extent to which we are involved in something new and the extent to which that elicits predictable emotions quickly becomes clear. At the same time I readily admit that in proposing the domain we were acting with a certain set of assumptions and with a certain view of museums world wide that not everyone will necessarily share.
Where is the value to museum audiences?
I think it is in the potential for greater visibility of museums online, thus facilitating the communications and educational missions of museums world wide. Visibility will be enhanced greatly by a common naming scheme in a single domain, rather than as is the case now, being scattered across several of the top level domains. This is not to say that multiple alternate names will not continue to exist for many museums; I expect that will happen for lots of good reasons. This consolidation may seem like a small thing for the National Gallery in London or the Ameican Museum of Natural History, but these institutions are not typical of the museum community in terms of Web activity. Visibility will also be facilitated through enhanced WHOIS and DNS services which MuseDoma can take beyond the services traditionally supported. Beginning with additional information gathered in the registration process we expect to be able to greatly increase the precision and recall of searches for museums online.
Because the museum domain is restricted this visibility will translate also to recognizability and the general realm of authenticity. There are, it seems to me, two issues here that revolve around the ability to be sure you are where you think you are with a Web address.
One has to do with the difficulty to being sure there is a one-to-one match in every Web browser's (I mean the person) expectation between the URL and the institution it purports to represent. Some of you may have never encountered the situation where name appropriation is done specifically to divert Web traffic because of the value of that name in the offline world. But I can assure you it happens often on a large scale to large institutions and with less frequency and perhaps with a more local intention to smaller public institutions. I could provide lots of examples based solely on Getty where the insistent provider of a .xxx site -- at the time of the opening of the museum in 1997 -- sought out version of a jpaulgetty address which we had not already registered.
The other has to do with the use of trademarked names, or names close to trademarked names, that are shared. Again there are lots of Getty examples since we share that part of our name with many individuals, many smaller companies, and two large companies: one of which deals in oil with which there is seldom serious confusion, and one of which deals in image licensing with which there is probably more frequent confusion in Web browsers' minds (people, again) than we appreciate.
It is clear in my mind that the more we can do to create visibility and recognizability in the online environment for museums the better.
Where is the value to the museum community and its individual members?
Some of the above goes directly also to this question. But this leads also to the value-added services that the museum domain hopes to be able to deliver. First on my list here is the mere existence of the museum domain. We expected that this would begin to increase the sense of community in the online environment for museums. Response so far strongly suggests that this is the case, and that by creating a locus for that community we may be able to target a set of concerns about being online that neither money nor software are able to. We are anxious to see how this will develop and are hopeful that it will contribute in this intangible way to help bridge the so-called digital divide that exists for thousands of museums and exponentially larger numbers of online visitors. Those among you who have dealt with similar visitor or audience profile issues will already have an appreciation for this. Among the most frequently recurring of reasons for not visiting a museum from the traditionally non-museum-visiting segments of population in Los Angeles is that they do not perceive museums as places they may go. They expect not to be welcome. They expect not to know how to behave. They expect to feel foreign. I believe we have an opportunity to mitigate that online for institutions as well as their audiences.
Each museum, too will find its own value in its own time in the museum TLD. But MuseDoma, as the trustee of the domain, will act proactively and will be able to initiate value-added services for its registrants. We have had only preliminary discussion of what these might be, but you might imagine the field is both broad and well-directed since MuseDoma will operate as a non-profit in the services of its constituents. We could:
and so on. While it is true that this envisions something of a progressive redistribution to benefit the less Web-enfranchised museums, it also begins to imagine a model in which this domain can think locally with museums and their associations around the world, and then act globally in ways that benefit the entire community.