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

Posts

Posts

11-Jun-10 09:00:00
Topic: Keywords
Forum: Wishlist

I wish the Subject/Keywords (DetSubject_tab) field that occurs in the Multimedia module was in more modules.

There are many times when we need to pull sets of catalog records that can't be pulled by a simple search that could then be turned into a group. As far as I know there is a limit to the number of records in a static group, so we can't use the static group function either to pull the records later on. Even if you can have an unlimited number of irns in a static group they can become very slow to retrieve. If we could mark records with keywords then we could easily pull sets based on one or more keywords.

This has come up recently for us as we try to pull together our specimens collected from the Gulf of Mexico. Many of our records say "Gulf of Mexico" and are easy to grab but others have only a latitude longitude. Searching on the range of lat/longs leaves you with records that need to be excluded because they are terrestrial, or errors that need to be fixed. Then there are additional searches to include lots from places like Lake Ponchartrain, the Mississippi river delta, etc. This results in more than 80,000 records, well beyond the 15000 record limit that I believe you can have for a static group. If we could add a keyword "Gulf of Mexico" once we have pulled together the records it would make it possible to pull up the records in six months.

I would also like to be able to add a keyword in Collection Events to indicate which are the clean or vetted station data records versus the records that have data that hasn't been vetted.

Joanna and anyone else who can help:
After looking at the latest version of the proposed changes for the Taxonomy module I have to admit I’m a bit confused. I had hoped to be able to attend the EMu Users meeting next week so I could ask my questions in person rather than trying to explain our problems in writing but since I’m not going to make the meeting that’s not going to happen.
I had thought when I first looked at the Hierarchy tab that it was going to solve our problems but I think I jumped to the wrong conclusion. Our problem in the Invertebrate Collection at NMNH (Smithsonian) is one I’m sure is shared by other collections: we have large parts of our collection stored under outdated classifications. As a genus is shuffled from one family to the next we have not moved our collections to match the most up to date classification. We have tried to keep the genera under one classification so you don’t have to look in two or more places for specimens. In some cases we are 50 years or more out of date. Rather than having a button for “current id.” on the Identification tab of the catalog module we had it renamed to “filed as” so we know which identification, including it’s higher classification, is the one the specimen is filed under in our collection. Being out of date hasn’t been a big deal until recently. Now that our database appears on the web and with the implementation of the Darwin Core fields for loading data to the various biodiversity portals this has become a very big deal – our data is really out of date and it is embarrassing the scientists. For the most part we have not created synonyms to handle the updates even though this would be one way of solving the problem at least in part. Our old system didn’t do it very well and we haven’t had the time to do it in EMu. There is concern that if we created new taxonomy records for all species of a genus that have been moved to a new family the catalogers aren’t going to know which record to pick. If they select both so the identification grid shows both the up to date classification and the classification the specimen is stored under we can flag which one we want to be the “filed as” identification but we have no way to flag which one is the up to date one (the one we want to appear on the web). I have always assumed that the “Current Name” flag on the taxonomy tab was to indicate the current scientific name and had nothing to do with the higher classification so that won’t solve the problem. I thought the Hierarchy tab was going to allow us to have multiple higher classifications attached to a genus but on re-examination I don’t think that’s what was intended with that tab it’s just a means for auto-filling the classification tab. If we are going to have to go the route of creating new taxonomy records to show alternate classifications then we need a way to flag which is the one it’s stored under in the collection (Filed as Id.) and which is the version we want to appear when data is pulled for the various portals (Current Higher classification). Ideally, this means adding an extra radio button to the Identification grid in the Catalog module. If the Hierarchy tab is supposed to let us add alternate classifications for a genus, or higher, then we will need another box next to “Is Current” that can be used to show which one is the one used for storing the collection.
If any of you have figured out a better way to handle the problem of keeping the data up to date without having to re-label and move collections I would love to hear it. If you are at the NY meeting then please pass your ideas along to one of our department’s data managers (Karen Reed, Tyjuana Nickens and Bill Moser).

Linda

Linda Ward
Dept. Invertebrate Zoology
NMNH-SI
wardl@si.edu

Karen,
Your question is one many of us have had and now after the discussion on the list we all know how to get at our images!

I've been playing around with the options on the Multimedia tab Ben mentioned and I was getting more files showing up in the cache than were in my list for my export report. I couldn't figure out if the problem had to do with checking "all images" versus "first image" but for some records I was getting anywhere from 2 to 3 images instead of the one 250x250 resolution image I was asking for. I would purge the cache and rerun the report and again I got at least 1 extra image (always for the first record). I finally figured out that if you are in Detail view of a multimedia record and have your EMu set to automatically load multimedia images those images are getting thrown into the cache along with the ones from your report. Oops, I should have known that! So if you want to have just the images from your report in the cache make sure you are either in list view or have turned off the option to automatically load multimedia images before you purge your cache and run your report.

Linda Ward
Smithsonian - NMNH

22-Nov-07 11:00:00
Category: Using EMu

Lee-Anne,

Thanks for your response. I was hoping that another organization had already started to tackle this issue. The more consistent we are in how we store this data the easier it will be to share the data between organizations.

If possible I would love to have screen shots of your Tissue and DNA tabs. If these have the basic fields the scientists want to track it would be a lot cheaper for us to have them added to our version of EMu, even with some tweaking, rather than having them created from scratch. There's no point in re-inventing the wheel.

At the risk of showing my ignorance what is the WISE system?

Thanks again,

Linda

21-Nov-07 11:00:00
Category: Using EMu

Are any of you using EMu to manage tissue collections? If you are do you use Whole/Parts records with Parent/Child relationships or do you link the records in some other way?

In the past we have treated tissues as just another preparation type and recorded the information in our preparation grid. We have fields in that grid that include preparation type which in many cases is really a storage medium (i.e. ethanol, isopropyl, dry, slide, block, etc.), we can record a count as needed (i.e. 134, for a set of 134 serial section slides), we can record comments, who prepared it, when it was prepared. Soon we will also be able to link to a specific location for each row in the preparation grid. There is no way to record bar codes for each preparation type or other unique numbers other than embedding them in the comments field. Now that we have scientists who are actively doing genetic studies the tracking of the tissues has become much more complex and our current method is not sufficient. With programs like the Bar Code of Life we will see an increasing need for tracking tissue collections and sharing information between EMu and other databases. The tissues need unique numbers so they can be tracked independently of the specimen they were removed from and in some cases we could end up taking a sub-sample of the tissue and that sub-sample will also need tracking. Do we assign the same catalog number to all of the Parts records as was assigned to the Whole or do we give each it's own Catalog number? If they all have the same Catalog number then do we just use the Other Number grid to assign unique numbers to each part. If you are using Whole/Parts and Parent/Child relationships to link these records you could end up with the case where a Parts record, which is a Child of a Whole record, could become a Parent record of another Parts record. I'm not sure how well that will work. Setting up parent/child relationships doesn't allow you to easily view the siblings in the same way setting up related objects and their relationships (ie. host/parasite, etc.) because the parent/child link is not building a grid.

Then there is the whole issue of data that is being stored in outside databases including those used to track the processing of the samples. Do you try and create a place in EMu to store this information, or do you simply link to the database via Multimedia? The scientists are concerned about having to do data entry multiple times rather than having a way to do data entry in one place and have it shared as needed with the other data sets or moved from the other datasets into EMu. Just as we are seeing a link being made in one of the upcoming EMu clients between EMu and GenBank not to mention the various biodiversity portals linking to EMu with the Darwin Core fields we need a way to share data between EMu and BOLDS as well as with other museum collections.

Any suggestions on how you have tackled these problems would be very helpful. If you haven't done anything with tissue collections yet but you anticipate a need in the near future it would be useful to know that too.

Linda

Linda Ward (wardl@si.edu)
Dept. of Invertebrate Zoology, Data Manager
National Museum of Natural History
Smithsonian Institution

As I understand it the summary lines and derived names can be customized for each unit within an organization, you don't have to use the defaults that come with the product. I assume there is a cost involved in having it changed and having the data refreshed so the changes show. AT NMNH we came up with museum wide standards for the Parties module since it was the set of data we were most likely going to share once our catalogs were joined together. This standard format has yet to be implemented so currently each department is likely to have different rules for building the summary line and the derived names. In Invertebrate Zoology the summary line in the Parties module builds differently depending on the value in the Party Type field. So even though our organization records have a useless derived name created automatically it doesn't appear in the summary line. I would strongly recommend adding the organization to the summary line of any person type records. We initially had the organization in the summary line and during one of our upgrades I asked to have it removed. The reason I wanted it removed had to do with using the summary line value in some reports and not wanting to have to create a formula to strip out the organization. As soon as that change was made and KE ran our data through the new rules I realized I had made a mistake. Without the organization in the summary line it makes selecting similar parties records more difficult. I just had to add one new catalog record to realize the problem because we have Parties records for two people named "Linda Ann Ward" myself and someone who has donated mollusks to our collection. Oops! If people don't take the time to look at the organization tab in a person record they could select the wrong one. This isn't too critical for us right now since we aren't using EMu for loans, accessions, etc. and less than 4% of our person records have an organization but this will need to be fixed in the future. I keep hoping that the Museum wide standards will be implemented so my department doesn't have to pay for the change.

I agree that having all the initials appear in the brief name would be more useful than just having the first initial. The derived name is very useful in building a variety of summary lines but I dislike how they are used in various fields for searching. For example, if you add a new record in Taxonomy and type in a full name in the either of the author grids it will come back and tell you there is no match because it is using the Taxonomic derived name for comparison and it consists of just the last name. While you want just the last name when building the author string it would be a lot easier if you could search from the grid using the full name. You are more likely to hit the one record you want. I would also like to see a rule set up so collaboration records are retrieved only when you are searching from a single value Parties field where you would attach a collaboration record, such as the IdeIdentifiedBy_tab field in the catalog module and not when you are searching from one of the grids like Participants or Authors.

Linda
NMNH, Invertebrate Zoology

John,
Speaking for the folks in Invertebrate Zoology at NMNH we would like to see the author and date as part of the scientific name. It simplifies reporting. We haven't created any records for just the genus without a species instead we use the traditional place holder "sp." (Cancer sp., Polydora sp. etc.) but I can see the virtue of recording the authorship and publication date of the genus if you were going to use it more as a research tool or for creating some reports. I do object to the recent change in the scientific name that includes family name (or any other higher classification level for that matter) when something as been identified only to a level above genus. If family name, or any of the other fields from the Higher classification, stays as part of the scientific name then something needs to be done to the script that builds our taxonomy summary line so we can avoid redundant data such as "Spionidae : Spionidae : : Polychaeta : Annelida" or "Gastropoda : : : Gastropoda : Mollusca". In both of these examples the first element in the summary is the scientific name. This also creates redundancy in our label reports where we have the scientific name on one line and a string of data from higher classification on another line. The formula to create that string is ugly enough without having to check to see if family, or some other bit from Higher Classification, appears in the scientific name.

Linda

29-Apr-06 09:00:00
Category: Using EMu
Forum: Searching

Are you tired of typing in long lists of search terms in EMu? Do you wish you could just cut and paste a list of search terms into a field and run your search? Well guess what you can. And no you don't need to play with the Texql/Sql statement to do it either. It uses the usual Windows copy/paste commands. Your list of search terms can be in a document or spreadsheet, they just need to be on separate lines. I've attached a document with two examples one of which includes some screen shots. This works in both EMu clients I have access to 3-0-06-004 and 3-1-00. I don't know how long this feature has been available but Ducky just discovered it today and passed it along to me and she asked if I could share it with everyone else. Our thanks to the folks at KE for this feature it will save us a lot of time.

Linda (NMNH)
Ducky (NMAI)

Attachment: Search.doc

25-Apr-06 09:00:00
Category: Using EMu

Warren,
Thanks so much. That's exactly the sort of report and documentation I was looking for. Figuring this one out should keep me occupied for quite a while.

Linda

21-Apr-06 09:00:00
Category: Using EMu

Gerard and Forbes,
Thanks for the documentation and loading it to the web site. I was actually able to recreate the report in the example. I was also able to create one of mine own from the Catalog module and share it with others, though it only contained one field and had no fancy formatting. I'm what you might call a "cookbook" programmer: give me examples and the logic behind a report and I can modify it for my own needs but I'm terrible at figuring things out from scratch. If anyone has some more complex reports/views that they are willing to share I think many of us would find them useful. Even the default page view report would be helpful. A screen shot of the results and/or a list of the fields used would make it simple enough for others to recreate the EMu end for testing.
Thanks,
Linda (NMNH)

18-Apr-06 09:00:00
Category: Using EMu

Nick,
Is there an update on when the documentation for the XSLT report type will be available?

Linda
NMNH (IZ data manager)

Will,
You can attach pdf files to a Multimedia record but rather than the first page of the document showing as the thumnail what you end up with is a default Adobe icon. I've added videos and other multimedia objects to a multimedia record and it just adds the type to the Mime format picklist.
We have a 300 page pdf attached to a number of records on our EMu web interface. If you go to http://www.mnh.si.edu/rc/db/colldb.html
and select the invertebrate zoology collection and search on catalog number 142035 - the Eunice record has the attached pdf. Don't attempt to open it from a dial-up connection you'll be there for ever.

Linda
IZ
Smithsonian Institution

  • Index
  • » Users
  • » wardl
  • » Profile

Board Info

Board Stats
 
Total Topics:
601
Total Polls:
0
Total Posts:
1362
User Info
 
Total Users:
827
Newest User:
Marcus Swann (Axiell Melbourne)
Members Online:
2
Guests Online:
152