Forgot password? | Forgot username? | Register
  • Index
  • » Users
  • » rpatters
  • » Profile

Posts

Posts

Hi...I just want to add a general comment on summary data changes and bilingual institutions. Any additions to summary data are reflected twice for bilingual institution. Once on the English side of the delimiter and for us, once on the French side. In some cases, our summary data (particularly for parties) is already quite long. I would suggest that any default changes only be implemented for bilingual institutions if they request it. Perhaps one set of defaults for unilingual institutions and another set for bilingual institutions.

Of course, one way around this is to have the summary data rules configurable in the registry for each institution. I am hoping that this is one of the things that will be in a future release.

Robert

We do not have herbarium records but we have debated this issue many times in our institution in regards to all types of multimedia documents. We have found through trial and error that the name that is most important to list as creator in the multimedia module is the creator of the multimedia item. We have had many instances where the multimedia item has been changed through improper conversion or editing and it was important to know who had created that multimedia derivative. The creator of the multimedia derivative is not always the person who inserts or modifies the multimedia record. We catalogue the original multimedia item (e.g. photograph, slide) in our catalogue and the photogapher/author is listed in the catalogue while the multimedia module is reserved for the creator of the multimedia derivative. All multimedia fields are reserved for the description of the multimedia item not the original document. Our multimedia documentalists take great pride in their attention to detail and "conversion" accuracy when working in the multimedia module and they certainly feel that they are primarily responsible for the intellectual content of the digital derivative. Other contributors are added to the other contributor
field.

In your example, we would list the person who scanned the document as the creator unless the person who attached the scanned document has manipulated it in some fashion. Then both names would probably be listed in the creator field. The person who assembled the sheet could be listed as a contributor in the multimedia module or an author in the catalogue.

We try at all times to keep a tight separation as to whether we are discussing a dublin core record for an artifact, or a dublin core record for a photograph, or a dublin core record for an article or a dublin core record for a multimedia derivative of any of the above.

15-Oct-05 09:00:00
Category: EMu Administration
Forum: User Support

>>I understand some organisations have set up an internal helpdesk to >>support their EMu users. I'd be interested on any information in >>particular:
>>- number of EMu users in their organisation
We have approx. 300 users in our organization
>>- number/type of queries handled by the helpdesk/day
As the dba, I handle all queries regarding the use of the system. We have an outside contractor (Conputer Associates) who handle the hardware but I act as a liason between them, KE and our users when their is a problem. Many of the trouble calls I get are simple windows problems since many of our users have enough trouble simply opening ke and their email and do not really know the difference between a windows issue and a ke issue. A number of my queries are realted to data standards and registration/cataloging issues so they are not directly related to ke but since those functions are entered in ke, the calls come to me.

It is not unusual for me to handle 10 or so calls a day and that number is even higher when training is involved or has just been given. I and the assistant dba's are responsible for all training for users of the system.

>>- number of staff on helpdesk
I am the dba and there are two assistants dba's that back me up when I am away. We are also located in two separate locations so one of the assistant dba's is the primary contact at the other location.
>>- structure of support in the institution, i.e. do users call the helpdesk with any EMu-related queries, or are they encouraged to "ask a friend" first?
We encourage the users to call us first since we are trying to get all users involvefd in the system and ask questions. We encourage them to call with any questions and or issues so we get a first hand knowledge of what is happening. Nothing is worse then hearing users grumble about something which they have not called the help desk about and which will take a few seconds to fix.
We do have a separate trouble call line for non-ke issues that is handled by Computer Associates and they forward many calls to me and I forward many back to them.
>>- is the helpdesk specifically for EMu or does your IT helpdesk also handle EMu calls?
We have a separate help desk for non-ke related issues.
>>- how do you pay for the heldpesk?
There is no charge for the ke help, it is listed as duties which we are to perform in our job descriptions (my job description is quite long).
- do you have any particular support documentation, i.e. Andy's "How To" sheets?
We have an faq which we have posted but we find that most users tend to call us first even if we guide them to the answer in the faq. They are used to calling the CA help desk for all non-ke issues so they automatically call the ke help desk no matter how small the issues. I still get 5 or 6 calls a week from people entering the wrong password.
Many users are at beginners level in windows so just describing a problem is difficult. I use CA's remote access software to access their desktop and view the problem.
>>- do you log calls using a particular software?
We do not log the calls. I simply keep a count. I used to log calls in CA's help desk tool, but that was taking too much time and no one seemed interested in the numbers or results. If I do have to start tracking them again, It will be in the CA help desk tool.

Good Luck!!

Robert Patterson
Canadian Museum of Civilization

01-Jul-05 09:00:00
Category: Using EMu

Hi Nick..I have tried this a few ways but I will try to simplify the problem I am having. I have made several cross module reports and I have not had any problems linking the .csv files until now. I simply want to link up the appropriate comment with the object that is listed beside it in the list. When I look at the resulting .csv files, I don't see what the linking field is between the object and the associated note. The link via eeventskey only associates the notes and the objects separately to the appropriate event. The only other field I have to link are the ObjectAttachedObjectsNotes_key and the ObjectsAttachedObjectsRef_key but when I compare the numbers in those keys, they dont line up with the data in the fields. When I link them together by key, I get the information repeated each time for the number of objects in the sets.

The records I am trying to report on are on multiple event records and i have no problem associating the notes to the event and the objects to the event. The problem is linking the right note to the object which it is beside. Perhaps I am missing something in the linking between the three tables. The set I am looking at right now has 7 events and I can see where the 7 events keys are in all three .csv files. There are 1913 ObjectsAttachedObjectsRef_key and only 1788 ObjectAttachedObjectsNotes_key. If i compare ObjectAttachedObjectsNotes_key 1788 with ObjectsAttachedObjectsRef_key 1788, the object listed does not correspond with the note listed.

Hopefully I am just missing something in this process.

Thanks,
Robert

22-Jun-05 09:00:00
Category: Using EMu

Hi ...
Has anyone figured out an efficient way to report on the objects and associated notes fields from the objects tab in the events module. How do you get the right associated notes to show up with the right object. Since those fields are not linked, it is difficult to get them associated together in a report. We are using 2.1.008. Is this handled differently in later version? Has anyone figured out an easy way to report on this?

Robert Patterson
Canadian Museum of Civilization

12-Mar-04 11:00:00
Category: Archive
Forum: Archive

Hi....We have a number of different public access sites up through KE right
now.

They are set-up in a number of different ways. Each of these interfaces was
created by KE and some of them were then turned over to our web teams to
complete the design work.

1) Public Access Artifact Database - This is an older version (4 years) that
uses information from our Artifact Catalogue. This Interface searches a
database that is specifically created for public access use only. Each
night this emu database is repopulated from scratch with information and
multimedia links associated from our internal catalogue. We have about
200,000 records in this database.

2) PHP Nursing Public Access Artifact Database - This is a newer web
interface that only searches a sub-section of the database from #1. It is
based on the newer EMu PHP features such as the contact sheet. It is in a
test version now and we hope to launch it in the next month or so.

3) Sites Database - This is another database created for the web that is
updated a few times a week. This Database brings together information from
the Sites, Collection Events and Bibliography Modules and presents them in a
single record grouped by the site record. This interface was created for
remote external clients that deposit their data in our system. All queries
here go against a database separate from our internal system that was
created for this use only. Since this is sensitive information, all access
to this interface is by password only.

4) Multimedia Database - This is an online php interface that searches our
multimedia module only. It is used by external designers for specific
projects which require access to certain images from our multimedia module.
The queries in this interface are run live against our internal multimedia
module. The "live" aspect of this is something which the web designers
required. They need immediate access to things which are entered into the
system. Even a one day lag created a problem.

5) Genealogy Site - this is a KE database built specifically for the web and
there are no equivalent modules in the Emu client. This project involved a
separate area of our institution and I am not as familiar with its set-up.
It has been set up as a pay for use site.

6) WarClipping Database - this is a database built on a sub-set of our
catalogue containing about 150,000 records of newspaper clippings from WWII
along with associated scanned pdf files. This project also involved a
separate area of our institution so I am not familiar with all the details
on its operation.

We are also in the process of updating the interface listed in #1 to the new
standard php interface.


We have gone through a whole number of discussions internally as to the best
way for each of these to operate. That is why they each have been done
differently. Our overall preference was to use an external database that
handled the public access queries separate from the internal client. We
have over 250 internal accounts in the system and we do over 100,000 edits a
month so we were worried about queries going against records that were in
the midst of being edited or created. Having a one day lag allows us some
flexibility. We were also worried about the security aspect and by having
the public using a separate database, we had several more options on how we
could set this up.

We also have had an issue here in that all our records posted on the web in
some of these databases have to be fully bilingual. All the data and all
the interface screens had to be translated and equivalent(English and
French).

We have also found that query times against these databases specifically
created for the web are faster due to their more simplistic nature. We have
also been able to reformat the data through the export in a way that is more
user friendly for display on the web.

The one consistent problem that has arisen is the timing of all these
updates. We have found that the timing of backups, database refreshes,
batch processes etc....have to be carefully co-ordinated to allow everything
to smoothly complete on a nightly basis. We have dedicated 5 licenses to
our web use.

We have had other individual issues and "challenges" with each of these
databases or interfaces due to their unique use and set-ups. If anyone
requires more specific information on any of these set-ups, please feel free
to email me and I will answer or I will forward it to the appropriate person
involved with that interface.

Robert Patterson
Canadian Museum of Civilization

20-Sep-02 09:00:00
Category: Archive
Forum: Archive

Hi all....

We have some questions, issues and comments regarding the use of the thesaurus module. We have been developing our own bilingual thesaurus for Object Names and we wonder how other people are handling it especially if you are dealing with bilingual data . We are not experts on thesaural functionality but we have referred to literature on the subject during all of our tests. We are entering text as well as images for each of our thesaural records.

One problem we are having is entering multiple records for the same term. Since a term may have multiple meanings and uses, we have been creating a different record for each use. Here is a very simple example:

Object Name: bottle;:;bouteille
Object Type: cleaner;:;à nettoyant
Object Sub Type: suede;:;à suède
Objct DIsplay Name: cleaner bottle;:;bouteille à nettoyant

Object Name: bottle;:;bouteille
Object Type: anti-freeze;:;à antigel
Object Sub Type: gas line;:;conduit d'essence
Object Display Name: anti-freeze bottle;:;bouteille à antigel

Our original concept was to have the Thesaurus contain Object Name Terms with one record for each use of the Term. We would therefore end up with multiple bottle records since a term can only exist once in the thesaurus, we had to add some modifier to make it unique . For Example:

Term: bottle;:;bouteille (1) or bottle (cleaner);:;bouteille (à nettoyant)

Scope Notes: to be used for bottles which contained suede cleaner

Term: bottle;:;bouteille (2) or bottle (anti-freeze) ;:;bouteille (à antigel)

Scope Notes: to be used for bottles which contained gas line antifreeze

The first question is why create multiple records for the term bottle;:;bouteille. We wish to add different scope notes and images for each use of the term explaining further information about the individual term (time frame used, geographic area used, image of a typical example etc....) all this information relates to the term itself, rather than the actual artifact it is related to. This means we need multiple records.

The scopes notes and the broader term would help indicate which bottle record referred to which type. This created problems in that we did not like the Display Name of the term from the record. Having to enter these modfiers after the term to make it unique creates an entry that is not as pleasing to the eye especially in reports. It is also a duplication of information contained in other fields in many cases. We tried different types of delimiters to make each usage of the term unique, but we were unhappy with each of the results. This especially becomes a problem when you are validating a field from the catalogue against the thesaurus. We currently do not have any fields attached to the thesaurus so the method of validation is only something that has been explained to us. It is one thing to have a term with descriptors (qualifiers) that you are are using strictly as a lookup list in an offline or unlinked format. It is quite another when the term is copied over and validated against a field. The display "look" of that term in the field then becomes important also.

In a bilingual thesaurus this only becomes a problem where the English and the French term are the same for both usages of the term in both languages. When the French side is unique, it makes the whole term unique whether or not the English is the same.

We also had problems where the term exists in more than one heirarchy. You have this problem with a term like bow. This can relate to a musical instrument, a weapon, hunting equipment etc....Another example is a playpen used as a laundry hamper. We originally organized the terms according to their functional use rather than its form. By entering them this way, it was possible for a single term to end up in more than one heirarchy in the thesaurus based on its fuctional use. This created issues in the tree structure when terms on the same level or on a related level only existed in one tree of the functional heirarchy. When we linked them to a new level heirarchy in the thesaurus, they brought across all their relationships. To make them unique, you therefore have to add the functional category into the term and once again we have the display "look" problem.

We have ended up deciding to use the Object Display Name as the future link to the Thesaural Record. We will also be adding our Object Sub Type to the Display Name. We realize that this means we end up with a lot of Compound Descriptors. We have also reorganized the structure by Form rather than Function. Although this does not meet some of the formatting features of a thesarual record (eg. parenthetical qualifiers) , it does satisfy the need for both a validating authority and a presentable display format. This has solved almost all of our problems, but we are interested in what other people who are developing their own thesaurus are doing and how are these issues being handled.

Robert Patterson
Canadian Museum of Civilization

30-Jul-02 09:00:00
Category: Archive
Forum: Archive

The whole concept of parent child records is something we have been
struggling with in regards to the events module. Many of our exhibits are
divided up into components with different staff, objects and loans
associated with each component. There is some information which they all
have in common but a lot of information that is different. We currently
record each of these components in different event records and have a common
naming convention with hyphenated extensions which brings them all together
in the title field.

Canada Hall - Permanent Exhibitions: Zone 2.9 - Early Settlements Exhibition
Canada Hall - Permanent Exhibitions: Zone 2.11 - British Military Exhibition
Canada Hall - Permanent Exhibitions: Zone 2.11 - British Military Programs -
Uniforms


A way to link all the individual exhibit components and programs to the same
master event record would be extremely helpful.
The only way to produce adequate zone driven reports currently is to use
multiple grouping reports in crystal and as all crystal users know, once you
add 3 or 4 external modules and have to do grouping reports, it can be quite
finicky.

Robert Patterson
Canadian Museum of Civilization

23-Jul-02 09:00:00
Category: Archive
Forum: Archive

The Exhibit Reference Number is also something which we have thought about
using.

The Exhibition Catalogue Number and other numbers which are specific to the
Object are stored in our catalogue along with an associated field for the
type of number. Our goal is to keep any object specific information in
catalogue. We have already found it difficult enough to run proper reports
on the Objects tab and its associated Notes in the Events module. We would
continue to use the Catalogue to record any object specific numbers unless
someone has come up with an easy way to associate the rows across those
columns (Objects associated with Event, Notes)on an object level for
reporting.


Robert Patterson
Canadian Museum of Civilization

03-May-01 09:00:00
Category: Archive
Forum: Archive

We have had to deal with the same problem. The solution we have
come up with is to add a field in our Object Catalogue called Party Remarks.
We have moved most of our "Party" fields to one tab. We have a field at the
bottom of the tab name Party Remarks. In this field we explain any
attributions we wish to place on a Party, such as an artist listed in one of
the fields above. We link the Party Field to their proper Party Record and
the attribution is explained in the Party Remarks Field in the catalogue.
We only place generic information which applies to all Objects in the Party
Record itself while information that relates directly to an individual
Object is entered in the Catalogue.

We also thought about using different roles to show this attributed
data. We also toyed with the idea of adding an "attributed to" check box
beside certain fields in the catalogue. The problem with those ideas is
that it allowed people to add an attribution without having to enter the
reason anywhere in the database. We have many historical records in our
system with attributed data and no explanation as to why the attribution in
question. The only reason we could think of to place the attribution into
its own field was if it was important to be able to query on which pieces
were attributed to which artist or manufacturer. It is not very often that
we are required to search specifically for attributed records and it is more
important for our needs to display the attribution in a report rather than
query precisely on it.

An example entry for use would be:

Artist: John Smith

Party Remarks(in the Catalogue Record): This piece is unsigned but
attributed to John Smith based upon an original bill of sale signed by the
artist and dated at the time the artist was working in Brownfield area.

It is not a perfect solution but it is one that we have found meets
our needs and it forces a person who wants to enter an attribution to
explain why.


Robert Patterson
Database Administrator and Artifact Documentalist
LADS
Canadian Museum of Civilization

  • Index
  • » Users
  • » rpatters
  • » Profile