All posts by eia

vLib Resource Types revised

Resource types have been revised.

The former resource types Factual Database, ePrint Server, and Subject Gateway are now generally typed as Miscellaneous.

Old links should redirect if defined for interfaces /start or /resource_type (more info on service interfaces).

A complete list of eBook-related resources is now available at http://vlib.mpg.de/start?type=Book%20Collection.

Hence the option eBooks has disappeared from the Special Content dropdown at http://vlib.mpg.de/start?srv=locate. Former deep links, for example http://vlib.mpg.de/start?special=ebooks, should continue to be supported if set up using the above service interfaces.

MPG/SFX scheduled downtime (Sunday 14 February, early morning)

In order to prepare an upcoming migration, the GWDG scheduled a few hours downtime for the MPG/SFX server on Sunday between 01:00 a.m. and 06:00 a.m.

Please be aware that MPG/SFX services will be unavailable during that time, while the upcoming system is synchronized with the MPG/SFX server in production.

On Wednesday, February 17, after thorough testing, the new system will be switched to production during our regular maintenance hours, which will involve another short downtime.

!! PLEASE NOTE: Switch to production was re-scheduled to take place on Thursday morning, February 18, between 8 and 9 a.m. Apologies for the short notice. !!

We apologize for any inconvenience.

vLib upgrade planned for Wed 16-09-09

We are planning to upgrade our portal software to MetaLib version 4.3.

Unfortunately, a downtime of several hours will be inevitable, consequently vLib is likely to be unavailable between 08:00 and 12:00 a.m. that day.

We would like to apologize in advance for any inconvenience.

UPDATE (16 September, 12:19): Installation process delayed. Services continue to be unavailable.

UPDATE (16 September, 15:45): Services available again.

Extended metasearches enabled in vLib

To enable extended metasearches, and thus to emphasize the big asset of metasearching (which is covering the widest possible range of information sources in the course of carrying out a single search) the number of resources searchable simultaneously via vLib has been increased significantly.

This enables creating overall search interfaces such as, for example, a search across all MPG Library Catalogs available at

http://vlib.mpg.de/start?module=metasearch&stype=locate&type=MPG%20Library%20Catalog

which may be useful to locate a needle in a haystack.

Of course, users will be deciding whether the usability of huge metasearch interfaces is acceptable. Hence we are curious to have some feedback.

information resource feeds – is this the data we want?

vLib information resources are available in RSS 2.0 format, see http://vlib.mpg.de/vlib-rss-feed.html

What’s the benefit of this? A news stream to hook interest? A huge list of entry points to be filtered and maintained in a user’s own environment?

As items basically consist of a title, a description and a link, crucial information about the vLib resource represented by an RSS 2.0 item may fail to be conveyed.

This may be a mere mapping issue as discussed at https://devtools.mpdl.mpg.de/projects/vlib/wiki/RSS (i.e. map more fields to RSS 2.0 item description) – where we are not taking into account additional atom elements yet (to contain, for example, an html-formatted description), or atom format to contain more detailed data about a resource.

Or we might even consider pointing subscribers to our own interface to a resource rather than to the resource’s original web interface.

The vLib resource interface URL is actually present within items as guid, and, in fact, it is meant to be a permalink – however, certain feed readers appear to prefer a permalink in guid to the URL in item link, that’s why guid isPermaLink is presently set to false.

Integration of electronic resources

On the one hand, there’s a rich abundance of electronic resources, on the other hand, it requires some effort to maintain and organize them.

An integrated platform may be a solution, but at least to some degree we’ll always be dealing with multiple interfaces and decentralized data.

From a user’s point of view, it may be desirable to integrate the various interfaces like links to web platforms, search engines, lookup tools, etc., with his or her own browser environment. Maybe this is actually a good place to combine everything, because this is where other useful tools and interfaces provided by third parties are to be found, see

e-resources in browser environment

If a browser search engine does not deliver results as desired, access to a more specific search interface is probably the best solution. By using vLib resource feeds, it is possible to organize these links in a feed reader or as bookmarks. Above I’m using Wizz RSS, a Firefox add-on feed reader.

Combining resources tends to blur the precision of a search, but on the other hand covers a wider range of content. Hence it may be useful to organize resources (with search interfaces) in sets, such as

vLib resource set

or use a deep link to point users to the interface which may best serve their purpose, examples: a list of resources, vs. a specific search triggered right away.

Documentation about deep linking may be found here.

ISI Web of Knowledge cross search released

In addition to Web of Science (including Conference Proceedings), major ISI WoK databases (Biosis Previews, Current Contents, Zoological Record) are now cross-searchable on vLib – example (MPG on-campus only) – which may be considered to be used as an alternative to the WoK All Databases search.

The metadata quality of the All Databases data source does not seem to match the quality of data as retrieved from single databases, which involves all sorts of disadvantages, e.g. problems with full text linking (wrong metadata is sent to the link resolver which is thus unable to provide a correct full text link).

Please note that cross search in general may or may not deliver results as desired. Please see vLib search tips for more information.

vLib, exporting records with IE 7.0

might give you a hard time. In fact, it looks as if many problems of this type have been reported to the provider of the digital library software the vLib portal is based on.

Internet Explorer 7.0, by default, does not automatically prompt users for download of file resources without their consent.

Unfortunately, vLib tends to send the download request "en passant", and is unable to repeat it (or present a page that contains a link to the file resource) after users have confirmed the download.

A massive but efficient response is turning on automatic download prompt – see http://vlib.mpg.de/vlib-browser-notes.html

Still, some versions of IE 7.0 may claim to be unable to find a file resource on the server – which is not true, as apache logfiles clearly indicate that the resource has been delivered. However, IE 7.0 is likely to be unable to cache it properly, and hence does not "find" it.

No apache settings that influence caching were apt to solve the problem. An update of IE 7.0 is probably a good idea in this case. Comments welcome.

A fix has been announced for the next minor release of the MetaLib software, so that file resources may be downloaded, even though a download has been tackled by IE 7.0’s security mechanisms.

vLib resource feeds

Resources referenced under http://vlib.mpg.de are available as feeds in RSS 2.0 format.

Resource feeds may be created by applying the usual filters to the complete list of resources as they appear in vLib.

Feeds are either created on the fly, for example, by requesting all Full Text Databases, or they can be addressed as static feeds which are updated daily, for example resources recently added or complete list of resources (A-Z).

Local views exist for every feed.

Please see http://vlib.mpg.de/vlib-rss-feed.html for detailed information.