CrossRef

This week's blog post is about Amy Brand's coverage of the product CrossRef. I had seen a presentation on CrossRef in LIS 644, Digital Tools, Trends and Debates, but I understood little of it and nearly none of the presentation had stuck with me, which is why I decided to cover it for today's blog post. One thing I noticed right off the bat is that I believe that Amy Brand, the author of this article, is an employee of CrossRef. Brand reports that CrossRef is considered a non-profit, but in the next breath describes it as a cooperative venture between 12 scholarly publishers. CrossRef was started by a loan from the publishers. Although it may be run as a non-profit, it likely has the interests of the publishers in mind.

The system was set up to be a cross-publisher citation linking system and also a DOI registration agency, originally mainly for STEM research. Brand also mentions the term "deep web" resource discovery, which, at its most basic, means searching and accessing websites that require a login through the World Wide Web that cannot be found through normal search engines. Authors read articles online, and then can click on a citation, which takes the user to a linked article. CrossRef is necessary, due to the fact that URLs change often. DOIs are more stable than URLs.

Factoids about DOI

• NISO Standard
• Alphanumeric
• Published in place of a URL so that it doesn't break a link
• DOI does not need to be descriptive or transparent
• Made of a prefix and suffix
• Updating is very easy, DOI doesn't change
• Prefix: assigned to content owner by DOI registration agency, all start with 10, then 4 digits
• Once a DOI is assigned to an article or book, it does not change, even if the item changes ownership, DOI doesn't reliably name the publisher, since this could change.
• 2-3 million DOI records being made per year, most to backfile
• At the time of publication of this article, 3 million DOI clicks a month

How CrossRef Works

Publishers use batch process to update metadata records. Then CrossRef stores XML metadata about the article or work; this can be called up with DOI. Simply put, they provide a database of DOIs and metadata that allow articles to be looked up. They make linking possible at any level, such as the book, chapter, or article, without having to worry about URLs being broken (109).
One thing about Brand's article that I found very helpful is that she explains that OpenURL and DOI are not competing technologies: this wasn't clear to me after class, so it was good to read this and get clarification. Brand describes how they work together, and that they are streamlined when they do work together.

Much of the article sounds like an advertisement for CrossRef and DOI technology. For instance, Brand reports that the "usefulness of the system is directly proportional to the robustness of coverage (106)," which, to me, smacks of self-promotion. She also mentions that a benefit of DOI is to create opportunities for article or chapter-based sales; this is the real reason that the scholarly publishers were interested in creating CrossRef, I'm guessing. One shortcoming is discussed in the article: CrossRef and DOI do not take user's institutional access rights in to accounts. This means it is not like our Find It button, rather, you may be prompted for a login. This is later discussed as a future upgrade to the system, near the end. Another upgrade that was thrown out near the end was forward linking. This is where a user could tell what also linked to that article, not just what linked from that article. Cool!