LC's Adoption of Silverlight -- Good Deal for Microsoft, Bad Deal for the Rest of Us

Earlier this year, Microsoft announced that it was giving $3 million in "funding, software, technological expertise, training and support services" to the Library of Congress to build on-site and online exhibits of LC historical collections. Others have commented on this. From a Jester's point of view, I've got problems with this on two fronts: Microsoft using LC in a cheap marketing ploy and LC's use of a new technology that impedes access for no good technical reason.

Library of Congress as shill for Microsoft

Interestingly, neither the Microsoft press release nor the Library of Congress press release mention the dollar figure. The first mention of it appears to be from Government Computer News in an issue dated 21-Jan-2008. That article says, "Microsoft will provide an initial grant of technology, services and funding worth more than $3 million" -- I can't find mention of how much of it is in the form of cash and how much of it is in the form of in-kind licenses and/or equipment.

This excerpt from the undated page "Microsoft signs cooperative agreement with U.S. Library of Congress" is the beginnning of my uneasiness. Emphasis and links added:

Scale: The Library of Congress receives upwards of 2 million visitors per year. Also, LOC.GOV is one of the top sites for search engines for international and U.S. historical searches and receives millions of hits and unique users per month. Children and teachers across the country will learn by using this site, the materials created here will be approved curriculum in all 50 states.

Scope of influence: This initiative will influence library technology worldwide. Libraries large and small from around the country and the world look to the Library of Congress for technical guidance and are certain to take note of what tools are being used in the NVE ["New Visitor Experience"] and on MyLoC.gov [not yet operational]. Because of its scale and breadth, it will have influence not only in the U.S., but also for scalable web sites in general. The Library of Congress will engage top educators to create educational content which will meet strict guidelines mandated by state departments of education for inclusion in public schools.

Partners: Schematic: User Experience. Portal Solutions: infrastructure implementation. [What the heck does this mean?]

The Technology: For the on-site visitor, Kiosks will be built using Windows Vista and Windows Presentation Foundation (WPF) using touch-screen technology for an immersive experience. Visitors will be given a “passport” (optionally mapped to a Windows Live ID), which can be used to digitally keep track of what exhibits each visitor has seen.

The publicly available web site will offer the ability to visit the library in a virtual environment which will complement the physical visitor’s experience. Because the Library of Congress is so vast, tourists will be able to “complete” their visit on line using the web site, which will have kept track of their visit when they were present at the library using the “passport” technology.

Clearly Microsoft is using this $3 million "gift" (again, we're not sure how much of it is real cash and how much is in-kind software licenses and suck) to push the adoption of Silverlight. I've got a problem with a public institution like the Library of Congress being used to push a commercial advantage. If Microsoft paid YouTube $3M to make videos available in Silverlight rather than Flash, it would be a different story. That Microsoft convinced HardRock (not a cultural heritage institution) to put up content in Silverlight doesn't bother me. If Silverlight had been adopted by the Library of Congress based on its own merits, it would be another story entirely. (Based on the information released to date, it doesn't appear that this was the case.) That leads into the next issue I have with the LC/Microsoft deal.

Silverlight-exclusive Content Impedes Access

Libraries (and cultural heritage institutions in general) exist to provide wide access to content to the public. That Silverlight is used internally to the New Visitor Experience kiosk environment is not a big deal. As soon as it leaks out into the open web, though, it is. (There are already Silverlight versions of Library of Congress online exhibitions.) For some, it is because Silverlight isn't supported for our computers. (For instance, when I visit the previously mentioned Hard Rock Memorabilia site, I'm told "The Silverlight Plugin does not work on pre-Intel Macs. Sorry.") For others, it is because the act of installing a plug-in is a barrier.

Silverlight does not have a wide scope of adoption; it is not installed already on the vast majority of machines on the net. Many access the internet in places that don't allow for plug-ins to be installed. (We hear about this at OhioLINK with regards to the educational videos available through RealMedia at OhioLINK -- and RealMedia has been out for a decade! We're actively investigating a shift to Flash-based players, by the way.) Content that is available exclusively in Silverlight is effectively not available to those that cannot -- for technical or know-how reasons -- install the plug-in.

A justification could be offered if Silverlight represented a big-enough shift in capability to justify the added effort to install the plug-in. I don't see any evidence that it is. For instance, I'm not convinced that Deep Zoom in the upcoming version of Silverlight is really all that interesting. It would appear to use JPEG tiles to get information from the server to the applet in the browser -- the same fundamental technique used by the Zoomify Flash applet and AJAX techniques like Google Maps. The Silverlight framework seems to give a clean, one-step way to implement the creation of the tiles on the server, but that can be replicated in other ways. (See for instance, my own efforts to create a shim between JPEG2000 and Zoomify.) Besides, from a content accessibility perspective, why would we make the programmer's life easier if it makes the viewer's life harder.

Final Thoughts

The LC/Microsoft would appear to be a good deal for Microsoft: for a token sum of money, probably primarily in the form of in-kind software licenses and support, they get a big platform for the exposure of Silverlight. For the Library of Congress: the deal is okay for now, but when the gift ends, to what extent will the money for software licenses and support have to be diverted from other operating budget items. For the users of the New Visitor Experience: probably a wash -- visitors get a slick experience that could be replicated in any number of technologies and techniques. For users on the open web: a bad deal because their cultural heritage content has been put behind significant, if not insurmountable, barriers.

The text was modified to update a link from http://www.gcn.com/print/27_2/45710-1.html to http://web.archive.org/web/20080924075059/http://www.gcn.com:80/print/27_2/45710-1.html on October 21st, 2013.