Heather Piwowar
RT @ORCID_Org: Please help us move ORCID forward by filling out this survey (by Oct 28): http://survey.euro.confirmit.com/wix...
Done. - Joe
Done... agree with D that it could have been better designed. - Bill Hooker
Agreed. Instead of "What are the main services, if any, that you would expect the ORCID initiative to provide?" I want it to ask me what are the main services I WANT it to provide. What I expect depends on how optimistic I'm feeling at the moment. Maybe it asks that later? I got frustrated and ran away, will probably revisit. - Heather Piwowar
I did the survey, but got increasingly disenchanted towards the end to see the mindset: sustainable=subscription. A closed garden ORCID system will never work. This is one system that needs an open API. - Chris Rusbridge
Also, they are trying to build a much larger system doing bibliographic & other stuff as well. Leave that to Zotero, Mendeley et al; stick to the core. Make it ubiquitous & essential; sustainability will follow. - Chris Rusbridge
Thanks to those that filled out the survey. The ORCID survey is just one way to get feedback, and you can always argue about how to phrase the questions. I will try to listen (and respond) to all comments given here and on Twitter, and you can also contact me by email. - Martin Fenner
Good to know, Martin! - Heather Piwowar
Given the pace of progress here, it certainly seems like just sticking to the basics, like name disambiguation and ID registration, would make sense. That's also the essential stuff - the features that only ORCID can provide. If you really want profiles, why not pull the info in from other places it already exists? Unless compliance is enforced by funding bodies, I think that you're not going to get people to give up much info, and even then only the bare minimum. - Mr. Gunn
ORCID is an initiative with many different stakeholders, including publishers and commercial companies, but also universities, public funders and non-profit organizations. It should be obvious that they have very different views on what ORCiD should do and how this is paid for. The survey should help in these decisions, so make your thoughts known. - Martin Fenner
Heather, I'm one of the voices behind the @orcid_org Twitter account ;). - Martin Fenner
to a certain extent you need _some_ profile information to disambiguate the authors if the automated system isn't enough... but when they start talking about tracking grants... bleh. leave that to another system - Christina Pikas
The way Twitter grew has some useful parallels to how ORCID can grow. I think the main thing to point out is that Twitter stayed very simple and focused on it's essential core features at the beginning. Twitter could easily have built desktop clients and offered monitoring services and follower management and all those sorts of things, and they could probably have built better versions than all the third party sites & services that grew up around twitter to provide these offerings. However, Twitter couldn't have built as many services as eventually got built, and working on advanced features would have distracted from making the core functions as polished as possible. Of course,Twitter has been notorious for downtime, but those were all scaling issues - the downtime wouldn't have been news were it not so easy to take it for granted. By not moving aggressively to fill in the feature gaps in their initial product, they motivated third-parties to come in and offer value-added services and thereby got far more publicity and promotion than if they had tried to build those extra features themselves. So I think the lesson for ORCID is similar - keep it simple and essential and become a part of the infrastructure. Focus on doing what only ORCID can, and see what grows up around you. As far as sustainability goes, well, charging end users to access profile data about themselves should rather obviously be a non-starter. Let the aforementioned third parties offer services that the ORCID registry facilitates, and perhaps the third-parties would be interested in chipping in. These groups could include funding bodies registering people on behalf of their grantees or publishers registering IDs on behalf of those they're publishing. It's only indirectly that end users will find this valuable, via the services that it enables. - Mr. Gunn
Authors themselves have an obvious incentive to help... tie in with PubMed or something similar, that nearly everyone uses, and you could get a lot of crowdsourced labour that way. - Bill Hooker
William, what you say makes a lot of sense. But I see at least three arguments against it: a) it is very difficult to convince people (especially the people that give the money, e.g. funders) that "less is more", b) a more comprehensive ORCID profile might compete with commercial offerings from companies, including Mendeley, so what is best for ORCID should be distinguished from what is best for someone else's business and c) do we have an open and comprehensive source of bibliographic information? The PLoS article level metrics is a good example. Both CrossRef and Scopus are commercial offerings, and they routinely give very different numbers for articles citing a particular PLoS paper. - Martin Fenner
The "author profile" idea doesn't make much sense to me, it will just make user segmentation worse and increase the confusion about the actual goal of the project. ORCID should work as a seamless, invisible resolver, not as a hoster of researcher-centred (and/or researcher-contributed) content. Ideally, it should be invisible to the end user and only visible to institutions and services managing researcher identities or to services managing researcher metadata. - ReaderMeter
I don't think "less is more" is how I'd frame it. The value of a service is not measured by how many features it has, but by how many people use it. In the case of a well-designed service that does a few core things well, you're looking at a "more is more" situation if you measure what matters - the usage. Let's face it, design is hard. Making something that everyone wants to use is very hard. ORCID has a unique opportunity here to provide something no one else can provide and achieve ubiquity in the face of no competition, so I guess I just don't understand why those things aren't being pursued with a single minded focus instead of worrying about things like tracking grants and such that will require extensive user experience design skills & won't make ORCID indispensable like the author disambiguation stuff will. - Mr. Gunn
Of course, I'm speaking just for myself here. If there's been any meeting at Mendeley where a policy regarding ORCID has been discussed, I've not been part (and I imagine Martin would have been invited). I just want to see this badly needed service exist. - Mr. Gunn
Oh wow... they even ask if they should go pay-wall here... NOOO!!!! That caused the whole mess we are in right now... Open Data, from the start to the end... - Egon Willighagen
@ReaderMeter -- how could you achieve disambiguation without some profile information? Name alone is never going to be enough, in fact I'd almost say the richer the profile the better. Even if everyone from this moment forward gets an ORCID and uses it religiously, there is still the existing literature to catalog... - Bill Hooker
@Bill... but name + ORCID should be enough, not? I see no need for further profile info... certainly not grant or course info - Egon Willighagen
I'm with @ReaderMeter... it would just be more profile information that gets outdated too quickly... did you see that post today about scientists spending hours and hours on promoting their work, but not spending a day a year on updating their main web presence? - Egon Willighagen
Instead... universities should link their databases to ORCID... just put the ORCID field in uni contact info databases, bits of RDF (s/RDF/JSON,REST,Foo/g if you like), and we achieved world piece... well, maybe not... simple, OPEN data, and lets mashups do the rest... will also considerably keep down the operational cost of ORCID... Oh, can't you just feel all those big icons of big players smell the money that can be involved in this?? Don't you see those logos eagerly looking at you, to ask you even more money for information that 'we' gave them in the first place?? ORCID solves a problem the publishers made, not the scientist... - Egon Willighagen
Please all raise your voice, fill out that survey and make them hear that us scientists don't want yet another pay-model service... - Egon Willighagen
@Egon, I wasn't thinking so much of grant or course info, though people could link their own web pages/FOAF/whatever just by including their ORCID. But suppose you find one of my papers in PubMed and it's just so bad that you want to see what else this blithering idiot has written... now you want to click on my name and have all my papers come up, but not those of the other "B. Hooker"s or "CW Hooker"s out there, right? How is that going to happen if somehow (and this will have to be an automated somehow) my ORCID hasn't been attached to the right papers and not the wrong ones? Without some profile info (geography, co-authors, subject matter) the auto-ORCIDifier isn't going to be able to tell me from the other Hookers... Or are we admitting that back-catalog is such a hard problem that the solution will only ever be approximate, and concentrating on future uses? I could go for that -- and for future uses, sure, name + ORCID should work. - Bill Hooker
The disambiguation of papers and other scholarly works already out there (as mentioned by Bill) is the reason ORCID wants to collect bibliographic information. At the beginning of this author identifier discussion I was very much in favor of a minimal system, but bibliographic information is often really messy and just adding an ORCID field to as many databases as possible might not be enough for proper deduplication of researchers and scholarly works. - Martin Fenner
The current list of ORCID participants is really short on funding agencies (the Wellcome Trust being a notable exception). We would probably see very different business models being discussed if a few more of the larger funders would commit to ORCID. - Martin Fenner
I'm starting to come around to the minimal idea -- maybe that should be put in place first, so that future publications will be covered and the back catalog problem won't continue to grow while we debate? Then we might also be in a better position to ask specific questions about what else is needed to deal with back catalog. - Bill Hooker
The ORCID initiative assumes that some authors will claim their own scholarly works, but that publishers, institutions and funding organizations will do so as well. There are several reasons to use this approach, including authors that are no longer active, authors that rather have their institutions do this work for them, and external verification of self-claimed works. - Martin Fenner
@Bill Hooker "how could you achieve disambiguation without some profile information" – I mean you just need an accurate way (institutionally controlled maybe?) to check someone's identity. The profile idea seems to fall prey to the same problem of inferring someone's identity based on someone's contributed metadata (affiliation(s) etc.), but I am not sure that's the right approach. @Marius Kempe's suggestion reminds me of the idea of GPG keys, Web of Trust etc. an example of an excellent infrastructure to represent someone's identity and its authority in a distributed way without any centralised effort. - ReaderMeter
While I very much a appreciate the ideas and good intentions behind ORCID, the clarity of what 'it is' was lacking in the survey questions. For example will it be a service ? or will it be a specification of formats and behaviors that allows for an eco-system of services to grow around it ? Resolution systems, profile information, linking with publications etc. Also this kind of discussion on a public ORCID mailing-list/wiki would be preferable than friendfeed, blogs, and twitter... - Greg Tyrelle
Greg, ORCID will be a service that at least provides a unique researcher identifier. The other services are still in discussion, hence the survey. Why would this kind of discussion be preferable on a public ORCID mailing list? - Martin Fenner