As Ian and I work on the Glenister collection we have continued the work on the Daily Iowan newspaper archive that I began this summer. We have been processing files following the workflow that I developed, creating metadata and uploading editions into CONTENTdm. Currently we have 160 editions uploaded and have several years ready to be uploaded.
Processing:
The Daily Iowan has been published since 1868, beginning monthly, then weekly and eventually being published daily. Each of these issues range in size from 4 to 16 pages. Because of this there are a ton of files to be edited. Ian and I have been using Photoshop to crop each page by hand. This makes every page look really good but is very time consuming and we need to weigh the value of having every page look perfect versus being able to have the whole collection available to people. I think we should concentrate on making the collection available and in a functional state. To remedy this Ian and I have begun to look at alternative methods (batch processing within Photoshop or using other software) to crop the images. I have been running tests in Photoshop using a action and batch processing. So far this seems to be working pretty well, out of a batch of 20 files I ran only 1 was not what I considered passable. I am now experimenting with larger amounts of files.
Metadata:
We have not run into any major metadata issues. We have been following the template that I created over the summer and filling out the unique fields in an Excel spreadsheet. We then save the spreadsheet as a tab-delimited text file.
Uploading:
We have run into some trouble uploading our files into CONTENTdm. Everytime we uploaded an edition we recevied an error message. Eventually we figured out that when we saved the spreadsheet as a tab-delimited text file extra space were being added. These extra spaces caused CONTENTdm to look for information that was not there and to return the error message. We went through and corrected the text files and also recreated the metadata template to solve this problem.
While uploading some of the issues we forgot to have CONTENTdm set to add banding to the images. To correct this we pulled the files back into the acquisition station and then replaced the jpeg images. This added the required banding to the files and then we uploaded and approved them.
Next week:
During the next week Ian and I plan on figuring out the best way to process files and begin to discuss the collection interface.
