There is a myth about information technology (IT) and in particular about those companies involved in building technology. It is a belief and assumption that such companies are always fast, nimble, agile, and can turn on a dime. Those not involved with technology are amazed when something new is introduced and seemingly works without any issue; we are struck by the marvelous engineering. In part, there is that sense of wonder and magic of, “How did they do that?” Of course, the hidden process of development may actually have been quite difficult and time-consuming. As outsiders, we don’t see the sweat and tears that go into making the technology function. Our perception is that it simply happens without error or flaw, and is introduced without difficulty into the mainstream. Organizations don’t broadcast their internal research or development work or telegraph their failures on the road to success.
The Internet turned 50 years old last week. The first computer-to-computer communications was transmitted — though the first attempt failed — in California on October 29, 1969. For all the internet’s ubiquity and considering the connectivity in nearly every manner of device these days, it is important to remind ourselves that this transformation didn’t happen overnight. It literally took decades. What seems obvious and apparent today, back in the 1970s was hardly commonplace. The first publisher I worked for only had a single connection point from the company to an outside internet, via a dial-up connection, which was controlled, for the most part, by the head of the IT department. In the years I was there, connectivity expanded, but it wasn’t an overnight switch.
This came up last month as we were discussing the adoption progress of the ORCID system in comparison with Crossref and the potential of the ROR organizational ID system. Crossref had a built-in advantage, which required a handful of organizations to commit to using digital object identifiers (DOIs) in their references to get the system up and running. With that advantage, the Crossref system was able to move from being a start-up to assigning more than 200,000 DOIs per month within its first year. It still took Crossref almost two years to reach more than 75 members and in the ensuing 20 years, it has grown to more than 14,000.
ORCID has experienced a different path toward success, in large part because it has been driven by individual adoption. While there are more than 7 million ORCID users today, its adoption curve has been far less explosive. It is in wide use now and continuing to grow in implementation. A lot of people were quick to focus on the possibilities and potential.
The new Research Organization Registry (ROR) for organizational IDs is just getting off the ground. It faces a similar adoption problem as ORCID, perhaps with the added difficulty of not having the excitement of an individual user-base driving excitement around the standard. Whether it proves to be successful or not really hinges on getting people to use it. And this won’t be obvious for some time. Even something as simple as using the same identifier for a thing takes time. The community needs to understand the value and believe that it will improve their efficiency or address some unmet need. Systems must be built up around it, and implementers have to incorporate it into their tools or processes. Each of these elements takes time. Sometimes this takes years. Unfortunately, because we presume technology can turn on a dime, we have unrealistic expectations.
This is true of NISO’s work as well. Last week, I was on a call listening to a tech team at a large publisher, as we talked about the first implementation of the SeamlessAccess.org technology, based off the RA21 Recommendation published earlier this year. There are challenges and issues to work through, which is exactly why the process is moving through a Beta phase with a limited number of partners. While public buzz about the project has been minimal since the summer, a lot of work behind the scenes is ongoing as we move towards launch of the service. Authentication systems are complicated and involve a variety of interconnected data stores and services, all of which need to be carefully vetted, due to the nature of authentication and the security issues surrounding it. There are legal questions as well as policy questions regarding data that must be worked through as well as issues of privacy and consent to be vetted. Fortunately, we are near to having a sufficient number of these challenges addressed to permit the launch with the first beta partners in the coming weeks.
Also, the FASTEN project is taking a significant step forward this month. Long in the development cycle, this recommended practice project seeks to modernize library-vendor technical interoperability using RESTful web service application programming interfaces (APIs) and standard mobile application intent calls. That group has released its draft for public comment of the recommendation. Please take a look at their work, and provide your feedback before the end of November.
Key to all of these advances has been commitment and dedication to seeing technology projects through to a fully-functional, mainstream conclusion. This was true of the Internet, of the on-going adoption of ORCID, and it continues to be so with the FASTEN working group. We need to understand that these are each works in progress, but — as with each and every new technology — rarely as magical or as swift in their development as we assume.