CHeriCoS

Open Library recently announced the availability of a RESTful API to access Open Library book metadata in RDF/XML format, which combines elements of the Dublin Core metadata terms with some elements from bibliontology and the registered RDA schemas. This service is in an early stage. The given example returns a nice RDF/XML, but without namespace declarations, whereas this book of Edgar Allan Poe returns a RDF/XML with declarations. Open Library offers several identifiers for their entries like Dewey, LC, ISBN 10, ISBN 13, LCCN, IA & OCLC, so this service could be very helpful when making mashups of data from museums, archives and libraries.

The term "memory institutions" is used throughout many papers in the field of cultural heritage. But till now i couldn't find a reliable source of it. Although you can find this term mainly in publications from around the millenium nobody seems to provide any clear definition or source. It's interesting that till now there is even still no wikipedia article about "memory institution". In this blog post about memory institutions the author quotes an article by Lorcan Dempsey 1999. He claims that

"Archives, libraries and museums are memory institutions: they organise the European cultural and intellectual record. Their collections contain the memory of peoples, communities, institutions and individuals, the scientific and cultural heritage, and the products throughout time of our imagination, craft and learning.

He is quoted many times, but he doesn't provide any source of her definition. On the WWW you will sooner or later find a definition by Birger Hjørland, who says that

Memory institutions [...] are for example libraries, museums, archives, cultural heritage institutions like monuments and places, botanical gardens, zoological gardens and all kinds of "collecting institutions".

His definition is based on a paper his compatriot Roland Hjerppe published in the proceedings of the 3. ISKO-Conference in Copenhagen, Denmark 1994 under the title "A Framework for the Description of Generalized Documents". This source is also quoted several times. But this could hardly be the first definition of "memory institutions". I haven't got the time to dig deeper into this, but a quick google book search reveals much older sources. For example you can find this term in an article in "Opposition", a journal of the Institute for Architecture and Urban Studies, in Issues 13-17 of 1978. The oldest use of this term I could find is in this report in 1972: Libraries and information technology: a national system challenge: a report to the Council on Library Resources. I guess some research will reveal even older usage. Meanwhile any additional hints are welcome.

As recently announced, the URL-shortener tr.im, is dead. So all shortened URLs will be gone shortly too. This is a good example why you should not rely on such service when you have something to write/link/post that should be available (and reliable) in the long term. Persistent Identifiers (PID) in Cultural Heritage like DOI, URN, PURL or other do have something in common with those URL shorteners. They are man-in-the-middle-services: id -> resolver -> resource. A resolver in the middle is always necessary to decode a given PID and point to a resource. The institutions and companies behind them guarantee that there will be services to do the resolving forever, but what if a company shuts down or an institution changes? What if in ten years OCLC decides to get rid of PURLs? Will there be someone to take over? URL shorteners do not claim that the shortened URLs will be redirected forever. But for example bit.ly says that they are "in the process of developing an archive system to make sure that links are available beyond our systems" (http://bit.ly/pages/faq/). So I think the "normal user" assumes his links will work the next few years and will be very bothered if they are not someday. But if you don't mind, you can check out these alternatives:

• TinyURL - The first(?) URL shortener is still online.
• Bit.ly - The default URL shortener of twitter.
• Redir.ec - A new service that claims to be very fast.
• u.nu - The shortest urls. Period.

Need more? Check out mashable's post about 90+ URL Shortening Services. Update 08-12-2009: tr.im is resurrected! Because of the overwhelming feedback they got after announcing the end they restored tr.im and re-opened its website. Sounds like a big marketing joke to me ...

Hand of Google It's not a new thing that sometimes the hand of the digitizer can be seen on scans available on Google Books. The Register and many other newssites and blogs have already talked about it. But now Google's hand has been spotted in Munich. The Bayerische Staatsbibliothek (Bavarian State Library) has a widely discussed agreement with Google Books since a few years about scanning millions of books from that library. The institutional database bears very nice scans of the hand of google in action. Take a look at "Bechstein, Ludwig: Deutsches Museum für Geschichte, Literatur, Kunst und Alterthumsforschung (1842)". Flip the pages, and you will see why Google is called Kraken ... so many hands ;-).

JQuery has a lot of very nice plugins, so was jQuery Tools really to be missed? There is already a UI extension, but jQuery Tools provides the most common UI effects used on (Web 2.0) websites in one single plugin. It features:

• Tabs
• Tooltips
• Overlays
• Scrollables
• Flash embedding

All effects can be combined, for example to create an artist's portfolio. Well, all these effects can also be archived by using an already existing plugin from the plugin repository, but here you got it as one single package of just 5.8 Kb! And the effects are damn good looking. The guys who developed the flowplayer too made a fantastic job. jQuery Tools is compatible with all common browsers and is published under GPL2 and MIT Licence. Via: tagdocs.de

Okay, I'm truly late with this. The 50th Birthday of the Hard Drive was 3 Years ago. As Wikipedia says, the Hard Drive was invented by an IBM Team in 1956. Recently i stumbled upon a nice Video on YouTube made by Hitachi to cherish this event. It's called "Life Without The Hard Drive" and features besides some nice music interesting and funny visualisations of how retrieval would be without hard drives.

Instead of the World Wide Web, there would be a World Wide Warehouse, where all the information of the world is stored. You'd have to type your search with an old-fashioned typewriter, and the browser would be a guy on a bicycle trying to retrieve an answer.

Well, see what happens ...

Maybe I will use this in one of my next talks. This video is one of a series of videos made by hitachi to promote hard drives and the underlying technology. Be sure not to miss "The Dawn Of The Tera Era", "Licence To Read" and the fantastic "Get Perpendicular".

I just stumbled across another strange Internet Explorer bug when it comes to input fields and forms. One of the most common requirements when coding a cultural heritage website with a database backend is to implement a so-called "google search". One input field, hit run and go. In PHP i'm used to check if the name of the hit button and it's value is set inside the POST-array. So if there is a form like this one ... <form id="searchform" action="index.php"> <input type="text" size="40" name="term" value=""/> <input type="submit" name="submit" value="Search"/> </form> ... I expect that after hitting return I have this $_POST array available in PHP: Array ( [term] => mySearchTerm [submit] => Search ) Afterwards you can do something like this in PHP to test if a and which form has been submitted: if (isset($_POST['submit']) { ... do something } This is especially useful if there are more than one form on the initial page. This will work with all common browsers, but sadly this is not true when using the Internet Explorer! You will get this array if you are using the mouse to click on the submit button. But when you hit return to submit the form you will get this array: Array ( [term] => mySearchTerm ) How cuckoo is that? What did they think in Redmond by implementing this behaviour or did they think at all? With the help of almighty Google I finally found a work-around which is described in this blog post by Richard Szalay. He deals with ASP.NET, but the problem and the solution are the same for other environments, because it's actual a problem on the HTML level. The described behaviour only occurs in IE when there is only one input field in a form and you submit this form by hitting the return key. If there are 2 or more input fields, IE acts the same way as other browsers. So the workaround is to add a second hidden input field to the form for IE only and disable it. A typical form would then look like this: <form id="searchform" action="index.php"> <input type="text" size="40" name="term" value=""/> <input type="submit" name="submit" value="Search"/> <!--[if IE]><input type="text" style="display: none;" disabled="disabled" size="1" /><!--> <input type="submit" name="submit" value="Search"/> </form> Using this workaround, IE will always submit the send button too, wether the return key is used or not. The if-statement is a conditional comment to assure that this tag is added in IE only. The attributes are added to reliably hide this input in css and non-css browsers.

Xenocode, a leader in application virtualization and delivery technology from Seattle, created an interesting showcase for his application virtualisation technology called the Xenocode Browser Sandbox. They took instances of the most common browsers today and wrapped each of them up in a virtual sandbox. The browser and the sandbox are packed inside a Windows execute-file which can be run without installing. Best of all, it is free to download! At Xenocode one can download packed version of Microsoft's Internet Explorer. Version 6 is available as well as the brand-new beta of Version 8. Mozilla's Firefox, Opera, Apple's Safari and even Google's Chrome are available too. Simply double-click the exe-file and the selected browser starts without installing anything, and it looks like a normal desktop application. All browsers can be run simultaneously , which makes it very useful for testing webpage's browser compatibility. So if you are into web development (and a Windows User) you should definitely check it out.