Hi Linda,

Thanks for your questions.

Here are a few resources that will help your process and a couple of
suggestions of how you might get started. I am sure that others will
chime in as well with suggestions.

1. I would recommend that you start with a data mapping from your
   existing fields to the fields in CollectionSpace.  There is a sample
   mapping template that we've prepared and it is available here:
   http://wiki.collectionspace.org/display/deploy/New+Implementer+Jump+Start

You might also look at the sample implementation plan for guidance.

If you'd like to see examples of other museum mappings, have a look at
these two pages:
http://wiki.collectionspace.org/display/deploy/SMK
and
http://wiki.collectionspace.org/display/deploy/MMI+Data+Analysis+and+Mapping

2. Yes, to installing an empty instance of CollectionSpace.
   It will be helpful to do this for a number of reasons including so
   that you and your team of end users will have a chance to check your
   mapping against what you see on the screen. I find it is very helpful
   to see where data are being mapped to rather than just a list of
   fields on a page.

3. I wouldn't be too concerned about knowing about SPECTRUM.  We use
   it as a starting point for our data schema.  The most important
   documentation for the software engineer installing CollectionSpace is
   available here:
   http://wiki.collectionspace.org/display/DOC/CollectionSpace+Release+Documentation

This documentation relates to the current public release v2.0.
With each new release, new documentation is added, so, you'll notice
that there are pages for 2.1 and we're currently working on releasing
2.3. I would urge you to start with v2.0

4. You may want to look around at the deployments section of the wiki
   to see what others are doing and how they are developing their plans
   and documentation.  There is a lot of good material there.

5. There is a page about using Talend Open Studio for data migration
   that might also be helpful.
   http://wiki.collectionspace.org/display/deploy/Data+Migration+using+Talend+Open+Studio+-+DRAFT

I hope that helps!  Keep the questions coming!

Angela

On Mar 9, 2012, at 8:10 AM, Linda Hocking wrote:

Hi Linda,

I'd only add a couple points to Angela's excellent suggestions, and I'll ask a question or two as well.

It is really important to figure out what parts of CollectionSpace you want to use and which ones need data migrated into them.  Will you be importing in information about loan transactions for example, or deaccessions?

When you install CollectionSpace, you basically get an empty system. It might come with some values in drop downs already initialized, but otherwise, you've got an empty database to start with.  At UC Berkeley, we've found that we usually start by importing some of the key authorities -- persons, organizations, storage locations (if those are all important to you).  That's important because when you import other data, you need more than the person's name in order to put a person in a certain field (such as the authorizer of a loan transaction).  You need a form of the name called the refname.  On Cataloging, you can get away with very little of this refname business.  It all depends on which fields on the Cataloging page need information from your existing system.

So here are my questions: How many objects do you have in your system now, and which procedures (transactions) will also require data migration (loans in, loans out, acquisitions, intake, object exit, media)?

Thanks,
Chris

Chris Hoffman, Ph.D.
Manager of Informatics Services
IST-Research & Content Technologies, UC Berkeley
chris.hoffman@berkeley.edu
510-642-9643

On Mar 9, 2012, at 6:10 AM, Linda Hocking wrote:

Linda,

I don't know if you already know this or if someone else has explained
it, but I believe the way most collections migrate data from exiting
systems is by using the cspace import service to load data into cspace.
In order to do that they create XML documents from their existing data
following cspace XML schemas. ETL tools like Talend and Pentaho Kettle
can help you do this for data in a database or data in text files (like
csv files created from excel files), and maybe directly from Excel files.

Susan

On 03/09/2012 06:10 AM, Linda Hocking wrote:

Many thanks to Angela, Chris and Susan for responding to my questions.
To answer some of yours, there are 6620 records in our existing
database. There is a loans module where some data regarding loans
exists. It doesn't generate the paperwork but records that an item was
out. There is an exhibits module that tracks what items were on display
and for what dates. There are three authority areas- people,
publications, and thesaurus. The location seems to be part of the item
record and not an authority file. We're hoping to use features
CollectionSpace provides that we have either not had in our current
software (or underused because they are unwieldy.) I think the most
important thing we need to get out of one and into the other are the
records for the objects and the associated images. The export does
create csv files, so we will need to figure out how to convert that to
XML. We have one other quirk- the archives and the object collections
basically come from the same people and they document the same people.
Our authorities duplicate each other. Is there a way to merge the two,
or point the name authorities from collectionSpace to the files stored
in Archon? Again, many thanks for your help!

Linda

On 03/09/2012 06:10 AM, Linda Hocking wrote:

only

excel.

programmer

with

the

http://lists.collectionspace.org/mailman/listinfo/talk_lists.collections
pace.org