Comments on: Loading/updating to Copac: how easy do you find it? http://copac.ac.uk/development-blog/2009/01/loadingupdating-to-copac-how-easy-do-you-find-it/ What's happening behind the scenes at Copac Fri, 27 Jan 2012 19:08:08 +0000 hourly 1 http://wordpress.org/?v=3.3 By: Ashley http://copac.ac.uk/development-blog/2009/01/loadingupdating-to-copac-how-easy-do-you-find-it/comment-page-1/#comment-129 Ashley Wed, 04 Mar 2009 12:36:51 +0000 http://copac.ac.uk/development-blog/?p=156#comment-129 A thousand libraries is a frightening number of libraries to combine. I think harvesting through a mechanism such as OAI-PMH would be the only sensible way of collecting all the data. You would want the OAI-PMH servers to support the reporting of deleted records, so you don't have to re-harvest the whole data set to work out the deletions -- though most servers seem find retaining delete information difficult. If you're going to de-duplicate a union database of 1000 libraries, then you really need an efficient algorithm. It is the de-duplication that really slows us down in getting the records in the database. We're hoping a move to a relational database and a change in the way we add holdings statements to bibliographic records will help us get records in more efficiently. The problem with updates is that you often don't know which piece of information in the record has changed and so you have to update the bibliographic record and the holdings. We're hoping to move to a system where we can updat the holdings immediately and put the update of the bibliographic record in a queue to be updated at our leisure later on when the system is less busy. A thousand libraries is a frightening number of libraries to combine. I think harvesting through a mechanism such as OAI-PMH would be the only sensible way of collecting all the data. You would want the OAI-PMH servers to support the reporting of deleted records, so you don’t have to re-harvest the whole data set to work out the deletions — though most servers seem find retaining delete information difficult.

If you’re going to de-duplicate a union database of 1000 libraries, then you really need an efficient algorithm. It is the de-duplication that really slows us down in getting the records in the database. We’re hoping a move to a relational database and a change in the way we add holdings statements to bibliographic records will help us get records in more efficiently. The problem with updates is that you often don’t know which piece of information in the record has changed and so you have to update the bibliographic record and the holdings. We’re hoping to move to a system where we can updat the holdings immediately and put the update of the bibliographic record in a queue to be updated at our leisure later on when the system is less busy.

]]> By: Domingo http://copac.ac.uk/development-blog/2009/01/loadingupdating-to-copac-how-easy-do-you-find-it/comment-page-1/#comment-126 Domingo Mon, 02 Mar 2009 16:43:28 +0000 http://copac.ac.uk/development-blog/?p=156#comment-126 Really this is a very interesting blog. We are involved in a project to create a union catalog from a thousand libraries here in Spain, and we are very interested in "What’s happening behind the scenes" at a project so model as COPAC. So I wanted to ask if you think that OAI-PMH could be a solution to harvest records from library catalogs OAI-compliant as data provider. We would also be very grateful if you could provide any information about hardware,software and operational requeriments in COPAC, maybe public reports o something like that. Really this is a very interesting blog. We are involved in a project to create a union catalog from a thousand libraries here in Spain, and we are very interested in “What’s happening behind the scenes” at a project so model as COPAC. So I wanted to ask if you think that OAI-PMH could be a solution to harvest records from library catalogs OAI-compliant as data provider.
We would also be very grateful if you could provide any information about hardware,software and operational requeriments in COPAC, maybe public reports o something like that.

]]>