1 / 21

Bottom Up Repositories: Why UCT-CS has a Repository

Bottom Up Repositories: Why UCT-CS has a Repository. Hussein Suleman hussein@cs.uct.ac.za University of Cape Town Department of Computer Science Advanced Information Management Laboratory High Performance Computing Laboratory July 2007. EPrints Software Author self-submission

mabrey
Télécharger la présentation

Bottom Up Repositories: Why UCT-CS has a Repository

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Bottom Up Repositories:Why UCT-CS has a Repository Hussein Suleman hussein@cs.uct.ac.za University of Cape Town Department of Computer Science Advanced Information Management Laboratory High Performance Computing Laboratory July 2007

  2. EPrints Software Author self-submission Checking of submissions Archive-everything! UCT-CS-specific metadata and classification systems Hierarchical browsing Simple and fielded searching OAI-PMH compliance The UCT-CS Repository: Revision

  3. Why we started a repository project • It was faster than simply waiting! • CS departments internationally archive technical reports (NCSTRL). • Research websites don’t last long (enough). • UCT doesn’t have an ETD project yet. • We need to improve ACCESS to our work. • We need to preserve our research output. • Bureaucracy (UCT, NRF, DoE, etc.) requires tracking publications. • We (think we) know what we are doing.

  4. Submissions to Archive over Time

  5. Analysis of Access • 187 items • 478520 log file entries • 798 unique user agent types • 65% from crawlers and bots • 34 from metadata harvesters • Data extraction from log file: • Unique IP addresses (visitors) • Source of access • Which resources are accessed • Access patterns of popular/old resources • Effect of submission sequence • Distribution of accesses • Analysis of partial responses

  6. Disclaimer • Web log analysis is at best an approximation! • Assumptions: • download = access • if useragent != bot, then real user • no caching • every user has a single IP address • …

  7. Unique Local IP Addresses

  8. Unique Non-local IP Addresses

  9. Unique Crawler IP Addresses

  10. Source of Access

  11. Which Resources are Accessed

  12. Access Patterns – Most Popular Resources

  13. Access Patterns – Oldest Resources

  14. Effect of Submission Sequence

  15. Distribution of Accesses

  16. Probability of Success on First Download

  17. Percentage of Partial Responses

  18. Summary of Results • The archive is regularly and aggressively indexed by search services. • The vast majority of access to resources comes from outside UCT. • Most users find their way to resources through Google and other search engines. • Most resources are used in most months – by both crawlers/harvesters and end users. • Access appears to be dictated by content, and not age of resources. • While many downloads are successful immediately, larger files result in more partial downloads.

  19. Conclusions • The results support our initial reasons for the repository! • A document repository is a simple and effective way to increase visibility to the research of a department or institution. • But it is imperative that documents are archived as soon as possible, to maximise visibility!

  20. Links • Open Archives Initiative • http://www.openarchives.org/ • Budapest Open Access Initiative • http://www.soros.org/openaccess/ • BioMed Central • http://www.biomedcentral.com/ • UCT CS Research Archive • http://pubs.cs.uct.ac.za • EPrints • http://www.eprints.org/ • DSpace • http://www.dspace.org/

  21. That’s all Folks! direct all questions and comments to: hussein@cs.uct.ac.za

More Related