Minutes of March 17, 2004 Meeting
Present: Ann Montanaro, Jeffery Triggs, Ron Jantz, Yang Yu, Nick Gonzaga,
Anne Butman, Sam McDonald, Judy Gardner, Dave Hoover, Mike Giarlo,
Excused: Karen Wenk, Lynn Mullins, Bob Nahoury, Linda Langschied, Tom Frusciano
Guests: Kalaivani Ananthan and Ruth Bogan
Meeting opened with Ann M stating that Kalaivani and Ruth were invited to give their respective reports. She also expressed the need to involve others on the committee that had not been involved in any of the subgroups that have formed to carry out DAWG work.
The 1st set of proposals were all rejected as either being too high priced, or not meeting the specifications that were required. We reworked the RFP (streamlined it, spelled out what FEDORA was included included the upper price) and submitted it back to the same vendors. We got responses from 4 of them (DELL, HP, SUN, VIACOM) the only ones that did not respond were EMC, and CDI/ADIC. There will be vendor meetings scheduled for Friday 3/19 and Monday 3/22 to review the proposals with each vendor. We hope to make a decision by next week.
Question asked - How long to delivery/setup ?
Answer - It depends on the amount of consultation time required by the vendor to assess our needs and develop the installation plan.
Once design templates (workforms) are ready we may be able to involve some members that currently lack subgroup assignments (Yang, Judy Sam) to test the system and give us feedback from an end user perspective.
Kalaivani reported that the WMS subgroup met on 2/27/04 and reviewed the requirements, identified main tasks and were ready to start recoding the system. They setup up their own deadlines for certain tasks that would allow them to know if they can meet their project target deadline. John K is working on the MODS piece, Jefferey T. is working on the pipeline (need to run test and know what software will be required), Kalaivani is working on organization structure and what will be collected (Ruth Bogan), Vince is working on the look of the interface. The group will meet again on 3/26/04 and will focus on the pipeline functions.
The need for documentation was emphasized and to that end John K. set up a documentation website where the working papers of the subgroup can be stored. These are only intended for the subgroup to keep track of decisions made and work progress, they are not meant as final documentation.
Question - What does the grant provide for software ? Jeffery is looking into using DejaVu SDK that would be very useful but will cost a couple of thousand. Workstation editing of files needs to be avoided at all costs. The next version of DejaVu will release a java applet (so that people will not need to download the plugin) and a watermarking feature.
Answer - There was only money in the grant for scanner machines and scanner based software for producing the images. There was no money budgeted for central server software. A pitch will need to be made if more money for software is required.
Vince has a prototype ready for simple objects. The WMS can produce an XML file that can be hand edited to produce MODS from the Dublin Core elements, the TIFF file can be run through a manual process to create the derivatives. Ron J. proposed using this prototype for a beta test collection at the SCC, to allow for the project target date to be met while the WMS subgroup continues to develop the system to automate all of the above manual steps. Discussion ensued as to what target we were meeting if the steps have not been fully automated.
Questions to be answered :
Do we want to use the above manual process to meet the project target deadline ? Do we want to release the manual process to end users ?
A question was brought forward as to whether we should scrap the Mysql database in favor of using an XML structure for the data. The main benefit seen was that we would have one common code base for doing editing since Fedora already has editing capabilities that work on XML files. While the common code idea sounds good, the cmtte felt that the Mysql db should remain - citing flexibility of having a db structure, not wanting WMS dependent on Fedora development, and that Fedora editing works in the context of the whole system they may not be able to be easily adjusted to a set of XML files located somewhere else. The issue was referred to the WMS subgroup for more discussion and a final decision. This is a critical issue !!
The idea of consistency of metadata being added into the WMS and after ingestion allowing it to be searched properly was raised. This mainly came up in the context of date searching. Do we only use yyyy/mm/dd or are other forms acceptable. How is the end user going to know the format allowed for searching by date ? A picklist of everything, instructions on the screen, are they going to be able to search date ranges, what about searching for I want all 19th century images ?
A question was raised as to what the DLR prototype contains. It is a learning tool for the developers it is not the final design for end users. Everyone on the cmtte is welcome to look learn and comment. Maybe a Bulletin Board feature can be added for people to submit comments. It was decided that we would look at the DLR in the lecture hall at the end of the meeting.
Metadata encompases everything Descriptive, Administrative, Techinical, etc .. Desciptive metadata will be given by the providers. Ruth wrote up her own list of operating assumptions. She is assuming a stable base of MODS, change in forms and templates to guide the user in their task. NJDH is building the forms, which fields are repeatable, and the controlled vocabulary lists. People in the field are not catalogers, wants help built into the forms to alleviate the need for extensive training/documentation. Forms developed will need to work for both NJDH and Rutgers - build in flexibilty.
Growing concern about Technical metadata. Rhonda starts with the NISO standard for images and tries to pare it back to one set of technical metadata for each mime-type. Do we want to store this data in Digi Prov and Tech MD sections ? Needs to be resoved !!
Question Can Ruth use her OA to have multiple forms with one underlying MODS data structure ?
Ruth will delever Metadata requirements to the WMS subgroup soon.
Login and look at Targets and Tasks that you have been assigned.
Use it to assign yourself subtasks, that will help you to chart any internal progress you have made on your Target. At the least add percentage completed on your Target.
The more information included the easier it will be for others to see the overall progess and work that is being done. You will be able to see where someone else is at espcially if your work is dependent upon them completeing a task. Unfortunately you need to know your dependencies on your own the software does not display them for you.
We need to start looking at the real public interface that will be delivered. DAWG will produce a General DLR interface with search and retrieval capabilites for all collections. People can customize their own search interface, they may need to know how to parse XML if they want a different record format than the one offered in the public DLR.
Good idea to form a Rutgers interface design group so we don't get to NJDH focused.
A Bulletin Board will be added for comments.
Submitted by: Dave Hoover