Reading the blogospheric discussions of CHORUS, I see some recurring confusions that need to be addressed. The first is the complaint that CHORUS is overly complex, even too complex to work. The confusion here is that the scientific publishing system itself has an intrinsic complexity, which any OA system must reflect, but CHORUS itself is basically pretty simple. It is a tagging and linking system.
Here is how CHORUS works, in its simplest form. During the manuscript acceptance process the author is asked if the article is based on US Federal funding. This might be a simple check box. If so then the author is asked to identify the funding agency, using a standard menu. That is all the author does that is new. This agency identification becomes the FundRef tagging metadata for the journal article.
The publisher combines this agency tag metadata with the rest of the metadata they already send to CrossRef. The latter includes the DOI that takes users to a freely-available, full text version of the article on the journal website. CHORUS collects this metadata and makes it available to the agencies. The agencies can then present it however they wish, thereby making their funding-based articles accessible.
There is nothing complex about this process. In fact it is much simpler and less burdensome than mandated repository systems where authors must identify the relevant repositories and submit their accepted manuscripts in a prescribed format. Moreover the CHORUS process reduces the confusion that arises from having multiple versions of articles in different places, by directly linking readers to the publisher’s site.
The second confusion is that publishers are somehow retaining control with CHORUS. What is missed is that the publishers will comply with embargo periods that meet the requirements set by the agencies and during this agency-specific policy development period have advocated for flexibility in setting embargoes for certain fields. Thus the government controls the OA aspect of the process. This is a major concession that the publishers are willing to make in order to retain users on their platforms rather than diverting to a mandated repository.
In its basic form CHORUS is much simpler and more efficient than a system based on mandated repositories. Of course additional features may be added, either to meet agency needs or to increase functional value. For example, preservation seems to be an issue with the agencies, although this has little to do with the basic concept of OA. The government likes to keep things forever. Publishers have, for more than a decade, addressed the matter of archiving as part of the process of delivering digital content to libraries that are known to be exacting in their requirements for permanence. Added features might make CHORUS more complex but they do not change the fact that the basic system is quite simple.
But the overarching confusion is simply that of the new idea, or in this case a new regulatory system. The OSTP memo has triggered a massive multi-agency rulemaking, to which CHORUS is simply a proposed response at this point. This sort of confusion is perfectly natural and unavoidable in major rulemakings. In my taxonomy of confusions, which was originally developed for rulemakings, the confusion of “vague or poorly defined concepts” holds a prominent place. New rules typically require new concepts, which is why major regulations typically have large sections presenting new definitions.
The design of CHORUS cannot really be specified in detail until the agencies decide in detail what they want to do, which they are far from doing at this point. The government is still in the pre-planning stages and so therefore is CHORUS. In fact the first proof of concept for CHORUS is not scheduled until August when agency OA plans are due into OSTP. As I have said before, this is a long term project, as major rulemakings must be. There is still a great deal of work to be done by everyone.
For example, it is clear that the actual mechanisms for compliance and verification are not yet included or worked out, and they are something that will require input from multiple parties. In principle, this would be something for researchers to significantly weigh in on, with agencies needing to articulate precisely what is needed. Thus a big problem is that there is so little that is clearly known at this time that it allows everyone to jump to whatever conclusion best suits their perspective. This in turn creates a classic case of people talking past one another because they are assuming different concepts. It is the confusion of speculation. What is needed at this point is patience and consideration.
19 Thoughts on "CHORUS Confusions"
American publishers, have you ever thought about a world outside the U.S.? And yes, there are even more foreign countries than Canada and Mexico.
In what way? The response now is to a US Government Mandate. I am deeply concerned that China and Canada will adopt the same policy the US decides on.
I am not a journal publisher and this is not about American publishers, rather it is about the American government’s OA rulemaking. Publishers around the world are affected. So I fail to see your point Bernhard.
Those who do not think much of CHORUS are not at all interested in making things easier for publishers. SHARE is even worse. Their workflow diagram made me carsick. I remember when the publishers were trying to work with NLM to get batched deposits into PMC from the publishers so the authors didn’t have to do it. Halfway through the work to make it happen, the NIH tried to change it to a 6 month embargo. Publishers walked away and the NIH stuck with 12 months. We know, because there is history, that authors are not going to do a good job of depositing their own papers. We also know, that publishers are not going to give up everything to make it easier for the authors. I guess we will see what the agencies decide but my money is on “neither”.
You might consider writing another post applying your taxonomy of confusions to the competing SHARE proposal. As I’ve noted elsewhere, it is vague and ambiguous about just what version of an article it is proposing this system to “share” and preserve.
Indeed Sandy, I am studying SHARE. The proposal makes clear just how burdensome and complex mandated repository systems are. It also calls for major research contract clause changes with the government getting a bunch of new license rights.
For those who have not seen the SHARE proposal it is here:
I am sure the battle has just begun and we are a long way for the final outcome. What does come to mind and might concern users is dealing with the publishers access control systems. All the major hosting services have an access control system to keep out the non-subscribers? How does the publishers hosting site allow an unregistered person into a journal issue to access the article covered by federal funding but yet blocks access to the other articles in the same issue. Also how does the publisher turn off the capture of usage data on the site when an outside user comes into the site to use the OA article. Most users do not want their information captured. If they are using PMC there is no active tracking of who is using what, but everything is tracked on the publishers site. My final thought is subjecting all of the users to the myrid of hosting services. Having to go out and learn the ins and out of all those hosting services is a royal pain. It seems that CHORUS is a quick and dirty solution to attempt to grab control but it might just be lacking a few features. What I do know is that currently libraries spend a great amount of time and frustration just keeping access to the titles that they have rights too. There is a patchwork of the good, the bad and the ugly hosting systems that are hard to navigate, unfriendly, and often provide poor help desk support. Libraries and their users need a seamless service that does not require additional work or support.
Access control for individual articles is a pretty basic, standard technology for any journal platform these days. It’s how journals can make articles freely available via Gold OA, how they’re made freely available for promotional reasons, etc. So that’s not really an issue.
Are there particular types of data being captured that you’re referring to here? Should research journals be subject to different rules than any other website? And PMC certainly does collect some information about usage of articles, that’s how they’re able to report the traffic that they see.
Only if that were true. In my consulting business I cannot tell you how much time I spend sorting out access control issues. Every day there is something new. Just listen to any resource librarians daily activities and you will soon see how much time and frustration is spent on access issues. Almost every change in hosting platform generates months of work and often the library still lacks access to content that has been licensed and paid for. In my line of work I am often knee deep into these systems and there is some ugliness that has not gone away. When a publisher has access control issues at the title or date range, it is of little comfort that someone pronounces that access control for individual articles is a pretty basic technology.
Well, there’s always the difference between what the technology can do and real world performance, but it is a lot easier to make an article free across the board than to work through complex access statuses for individual articles and journals for individual customers. A basic yes/no switch is simpler to implement than a yes for some/no for others switch.
That doesn’t mean that some papers don’t fall through the cracks. That’s why a system like CHORUS is so important, as it can (at least in theory) automate things and ensure better compliance.
You are right there. Audit the help desk log or sit with the help desk staff for a day (I often do both) and you can see first hand the difference between what should work and what doesn’t work. Developers often build systems to do what they think the user will like and often miss the mark. I believe about 95% of user access problems are related to the access control system. I was dealing with an ACS problem as recently as last week. Some hosts are better than others. I have my list.
Just an AMEN to Dan’s posts. We are still in the midst (and no it isn’t the publisher’s “fault” in this case) of access issues for our Springer content from their move last winter to a new platform. Springer has been very helpful, but other links in the system aren’t being as responsive. Which is why more than publishers need to be at the head of the table:: Indexers (NFAIS please) perhaps Google and Bing, Aggregators, Librarians (probably ARL), OASPA, DOAJ, NISO just to name a few If the Chorus proposal is actually going to have support beyond the blue list of publishers. I think if it’s seen as purely a publisher proposal it doesn’t have legs. And I don’t mean these other players should be “consulted” I’ve been on that end of the table, and it doesn’t work very well..(Crossref, Academic Press, etc.) They need to be integral to the proposal for it to work. I’m not against the concept, but if its actually going to succeed, it needs more depth of buy in from other key participants in the scholarly communications chain.
Given that ARL has just made a counter proposal this sort of global collaboration seems unlikely. I think keeping it simple is the key to success. The government has a simple need and the publishers are proposing to meet it. Piling on players and features is the road to ruin for the government. Remember it is budget cutting time.
I have bagged another confusion! (Sorry for the enthusiasm but we collectors are like that.) Duke’s Kevin Smith says that CHORUS is a “dark archive” but of course CHORUS is no archive at all, dark or bright, just a tagging and linking system. His point seems to be that CHORUS does not provide much discovery but that is not CHORUS’s job. The discovery systems will provide the discovery, should they choose to do so, using CHORUS metadata. He even suggests that publishers do not want discovery, which is truly strange given that the purpose of CHORUS is to help deliver eyeballs to the publisher websites.
Mind you there is a so-called dark archive issue but it is on the federal side not the publisher’s. Some feds want to collect every accessible article and put it into the vault just in case the publisher goes away some day, or quits CHORUS, or CHORUS folds, or whatever. I am not sure the articles are that valuable but it is not a publisher issue at this time.
In any case if others see possible CHORUS confusions feel free to post them here. Help build my collection.