More on ARTstor interoperability
Joan Beaudoin at Bryn Mawr College directed me to a piece about ARTstor that was published in the most recent edition of the VRA Newsletter. Here you will find ARTstor's plan for interoperability, as outlined by Barbara Rockenbach, Assistant Director of Library Relations at ARTstor.
This is a very discouraging statement to read.
The first problem is that a full two-thirds of ARTstor's interoperability strategy doesn't involve any interoperability. Exporting a low-resolution image isn't interoperability. Stable URL's are a good thing, but they’re not interoperability either. When a professor has to take her personal images and manually add them to an off-line ARTstor viewer, she's not reaping the benefits of interoperability. In fact, anything that doesn't mention or imply an Application Programmer's Interface (API) isn't interoperability. Applications have to speak to each other, which requires that their programmers write the code that enables transactions between them.
The second problem is that the interoperability ARTstor is planning to build doesn't actually help our primary candidates for ARTstor use. The possibility of seeing ARTstor metadata and thumbnails in the Luna Insight client is useful for occasional research activities, but is insufficient for the core activities of professors and students.
The final problem with ARTstor's approach is its self-centeredness. You may be able to search your local images from the ARTstor someday. Maybe next year they'll even offer image hosting, which dodges interoperability by offering the monolithic solution. In both cases, you can use ARTstor images, but only in ARTstor. (By the way, the pilot projects they are hosting this year are static collections. They won't be modifiable by collections managers, which is an essential level of control for visual resources departments.) Neither of these solutions is even available today.
I have been working under the assumption that the Mellon Foundation sponsored this project with neither of these expected outcomes:
In short, ARTstor is a perfect example of bureaucratic product design where the purveyors failed to understand how the end user would use their product. If they had put the needs of the teacher ahead of the needs of the collector, the ARTstor product would have been built upon an entirely different foundation—ARTstor would be unswervingly focused on getting their images into our slide shows and study sets.
An example of the right way to build a service comes to us from our friends at Apple. The genius of Apple's iTunes Music store wasn't the software interface or the integration with the iPod player. Jobs' masterstroke was that he convinced the conservative rights holders (record labels) that they needed to sign off on terms of agreement that were liberally in favor of the consumers who would pay for the music. Apple understood its customer and built a business plan around it.
If anybody reading this is from the visual resources or library communities or is an instructor who teaches with images, I encourage you to lend your support to the effort to bring about a change in ARTstor’s interoperability directions. Post comments here at Think Thunk. Better yet, put the word out in any appropriate venues you frequent.
This is a very discouraging statement to read.
The first problem is that a full two-thirds of ARTstor's interoperability strategy doesn't involve any interoperability. Exporting a low-resolution image isn't interoperability. Stable URL's are a good thing, but they’re not interoperability either. When a professor has to take her personal images and manually add them to an off-line ARTstor viewer, she's not reaping the benefits of interoperability. In fact, anything that doesn't mention or imply an Application Programmer's Interface (API) isn't interoperability. Applications have to speak to each other, which requires that their programmers write the code that enables transactions between them.
The second problem is that the interoperability ARTstor is planning to build doesn't actually help our primary candidates for ARTstor use. The possibility of seeing ARTstor metadata and thumbnails in the Luna Insight client is useful for occasional research activities, but is insufficient for the core activities of professors and students.
The final problem with ARTstor's approach is its self-centeredness. You may be able to search your local images from the ARTstor someday. Maybe next year they'll even offer image hosting, which dodges interoperability by offering the monolithic solution. In both cases, you can use ARTstor images, but only in ARTstor. (By the way, the pilot projects they are hosting this year are static collections. They won't be modifiable by collections managers, which is an essential level of control for visual resources departments.) Neither of these solutions is even available today.
I have been working under the assumption that the Mellon Foundation sponsored this project with neither of these expected outcomes:
- create a dud product that receives a lukewarm response from the higher education community; nor
- create a monopolistic software vendor that squeezes commercial and educational/not-for-profit alternatives out of the marketplace.
In short, ARTstor is a perfect example of bureaucratic product design where the purveyors failed to understand how the end user would use their product. If they had put the needs of the teacher ahead of the needs of the collector, the ARTstor product would have been built upon an entirely different foundation—ARTstor would be unswervingly focused on getting their images into our slide shows and study sets.
An example of the right way to build a service comes to us from our friends at Apple. The genius of Apple's iTunes Music store wasn't the software interface or the integration with the iPod player. Jobs' masterstroke was that he convinced the conservative rights holders (record labels) that they needed to sign off on terms of agreement that were liberally in favor of the consumers who would pay for the music. Apple understood its customer and built a business plan around it.
If anybody reading this is from the visual resources or library communities or is an instructor who teaches with images, I encourage you to lend your support to the effort to bring about a change in ARTstor’s interoperability directions. Post comments here at Think Thunk. Better yet, put the word out in any appropriate venues you frequent.
posted by EB at 5:30 PM
Site Feed
4 Comments:
Interesting blog--thanks. I have a few thoughts in response.
1. Interoperability is a serious consideration, but intellectual property concerns mean that there will probably never be utterly transparent interoperability. I'm not so sure that's a dealbreaker, though.
2. I understand the need for schools to be able to choose their own asset management tools to a certain point, but I'm not sure there's any way to do a resource like ARTstor right without "locking it down" to a certain extent, again because of intellectual property/copyright issues. I don't think that makes ARTstor into a Blackboard or Microsoft, though.
3. I don't think it's realistic or even necessary to conclude that ARTstor should "be unswervingly focused on getting their images into our slide shows and study sets" because to do otherwise is to privilege the collectors over the teachers. ARTstor is already being pretty bold, I think, in the way they've arranged for liberal campuswide access to images that have real commercial value.
4. I don't think the iTunes analogy is a good one. Those songs cost 99 cents apiece. Multiplied by 350,000 (eventually 500,000), that's a lot of money for a comparable ARTstor collection. And of course you can't make a song purchased from iTunes freely available campus-wide. Then, too, there's a lot more intellectual/professional overhead and instructional value built in to what ARTstor is doing than what iTunes is doing. Also, iTunes sells music that is lower-than-CD resolution. ARTstor doesn't decrease the resolution unless you export the image. Finally, iTunes is a store, a retail outlet. ARTstor is a teaching/learning resource that's building a collection.
I don't mean to be too harsh in this comment, because you raise some valid concerns. And I appreciate your taking the time to blog on this issue.
Thanks for your comments, anonymous...I've written updates to this old piece, as new info has come along, so I encourage you to go to the top level of this blog if you want to read the most recent news. But here are some reactions to your remarks
"Interoperability is a serious consideration, but intellectual property concerns mean that there will probably never be utterly transparent interoperability."
I can agree with this to a point...doing anything we wanted with the images was never in the cards. That said, the whole V.R. community has been disappointed--and surprised--by the limitations ARTstor has presented by virtue of the restrictive terms it has negotiated with various IP interests.
"I understand the need for schools to be able to choose their own asset management tools to a certain point, but I'm not sure there's any way to do a resource like ARTstor right without "locking it down" to a certain extent, again because of intellectual property/copyright issues. I don't think that makes ARTstor into a Blackboard or Microsoft, though."
Of course it doesn't make them a Microsoft. (We get along fine with the Blackboard folks, so I'm not going to lump them into that crowd.) But I think it might have been able, for instance, to set some terms for non-commercial partners who agreed to the same terms and restrictions to have been able to build some search-and-retrieve interoperability into the ARTstor database. Maybe some of these museums and collections would have balked, but some might have gone along...and once a few go along, more would probably have followed.
"I don't think it's realistic or even necessary to conclude that ARTstor should "be unswervingly focused on getting their images into our slide shows and study sets."
The rationale for ARTstor that was conveyed to us at the outset was to reduce or eliminate the redundant effort of IHE's all reproducing the same images. Which begs the question...for what purpose were we scanning and photographing all those images. Most Visual Resources offices serve images that are used by faculty to create slideshows for classes and for student study sets. Sure, there are other uses, but especially in Art History, that's the dominant mode. To focus on the size of the collection without accurately calculating how the collection can lead to a terrible mismatch between demand and supply.
"I don't think the iTunes analogy is a good one."
I'm sorry to say this, but I think this criticism misses the point by being too literal...I didn't say that ARTstor should have endeavored to BE iTunes. I was using it as an example of how an entity (Apple) created
(a) a customer-focused service that
(b) revolutionized what was possible by convincing I.P. holders (the recording industry) to liberalize their usage policies for digital music far beyond what they'd ever allowed previously (purchase-by-song; reasonable price; almost unlimited CD-burning; while
(c) accomodating the interests of the rights holders through appropriate compensation and digital rights management restrictions.
The particulars are specific to the business model and the goods in question. The general theme is that Apple figured out the needs & requirements of the customer, then convinced the rights holders to meet those needs with Apple in the middle.
(Also, fwiw, you say that the analogy doesn't hold because the audio quality is below-CD quality. The AAC files are actually quite good, and far better in compartive quality to originals than most of the artstor images are. The largest collection of ARTstor images is scanned at pretty low quality. But like I said, don't get bogged down in particulars. The analogy wasn't intended to be stretched that far.)
Thanks for the response, Eric. Three quick thoughts to carry on the conversation a bit:
1. I think we agree, mostly, on the interoperability issues, especially with this latest news from ARTstor.
2. My institution, the University of Mary Washington in Fredericksburg, VA, doesn't have a digital image database, so we're starting with a cleaner slate with regard to integration than many larger institutions are. I think I understand the concerns of those larger institutions better now, for which my thanks to you and to others on VRA-L.
3. I'm sticking to my post on the iTunes analogy. Intellectual and cultural resources in higher education raise a whole host of issues that aren't addressed by what Jobs achieved with some (not all) music content providers. That's the main reason why I think the analogy doesn't work. (We can argue about AAC another time.) The larger issue, perhaps, is that our market-driven society hasn't yet figured out the relationship of education (and other such cultural processes) to that market. I don't envy ARTstor the tightrope it's trying to walk, but I do applaud the ambition and think the early results are impressive.
Gardner Campbell
University of Mary Washington
(who at this point is reluctant to register with Blogger, for reasons he can't quite articulate :-))
Gardner - the interests in the current market exchange are pretty much all not-for-profits, but the reason for all the consternation is still all about money. I don't think this is a case of "market-driven society hasn't yet figured out the relationship of education (and other such cultural processes) to that market." The market isn't functioning if actors on all sides of the exchange can't see opportunity and value from participating in trade. I stand by my guns on this--ARTstor would be in a better place today and tomorrow if it had understood the marketplace better before it jumped in. We would have all been better off, I feel, if it had been run somewhat more like a business.
My iTunes example is valid! Valid, I say! :-)
FWIW, I agree that ARTstor is noble in its way for getting in the middle of this dysfunctional relationship between institutions, but I also think that a few poor compromises in the wrong direction could undermine their best effort. I am encouraged by the direction they're heading now, but as you can tell from the VRA-L list, getting to this point has caused quite a lot of exhaustion and frustration for the members of the V.R. community. (I may have written my criticisms on this blog, but the comments on the list reveal that I'm hardly the extreme point in disappointment with ARTstor.
Post a Comment
<< Home