Wesley Alan Wright: Notes, Lessons, and Projects

Met with Paul to discuss next phase of my involvement:

Programming Issues for Wes (5/18/06)

Preprocessing Images

Goal is to have thumbnails of the right size ready to serve up and for these thumbnails to be created automatically when new images are uploaded.

Needs to be done en masse for existing images and one by one for new images using scripts.  Image size requirements are :  for small thumbnails,  80 pixels high with resulting width no greater than 160 pixels.  For most images this will be no problem.  For a few panoramas, there will need to be automatic cropping to a total width of 160 pixels.  For large thumbnails, images should be 360 pixels wide; height is not restricted.

Since we are cleaning images prior to upload in photoshop, my guess is that the highest quality php down sampling routine will work fine.

Zoomify

Goal is to have zoomified images ready to go for delivery

Needs to be done en masse for existing images and one by one for new images using script.

Not sure where we ended up on this.  Ben Wolfe isn’t answering emails.  Hoping that this is something you and Justin can help with.

Quick Search

Ben suggests that you implement the full text searching of all database fields in the quick search box on the header in landscape new.  He will concentrate on the “refine search” part of the site.

Refine Upload and Editing Pages

Goal is to have pages that work with new display system and allow for input and editing of new fields including: getting field names reconciled between management pages and Ben Fenster’s database and display, linking URLs to images, linking PDFs to images, adding sound files to images (need to chose format), linking images to other images, contextual dating information (date and rationale), adding a theme field in the database and allowing us to add themes, making source a restricted vocabulary on the input page and allowing us to add sources to the list, making sure we can attach all meta data individually to each image in the series (000, 001, 002).  Limit title input to specified number of characters.  Keep title the same for all images in a LS series. Change n.d. to no date. Embed keyword URL and output into upload page.  Add field for data checker name in page processing out of temp (add to database).  Add field and entry for 911 address.
( HYPERLINK "http://www.uvm.edu/perkins/landscape/keyword_selector.htm" http://www.uvm.edu/perkins/landscape/keyword_selector.htm). Remove current pull down and option to add keywords

Spell checking of existing records as well as new records upon entry is key.  Ben mentioned GNU Aspell.

Bulk Upload

Goal here is to be able to go from our metadata excel worksheet to temp directly by uploading a folder of image files and a text file with the associated metadata.  Will require some type of error checking to make sure that images cited in the worksheet and those in the file match names as well as checking that all fields have values

Database issues

Figure out way to calculate and enter in the database the size of the original image both in MB and in pixel dimension.

Add columns to data table to deal with contextual dating and rational for such dating.  The idea here is that we have a student working on dating images using clues in the image.  She will come up with a date range most times (we could use the mid point for sorting) and rarely she will have an exact date.  She will want to be able to add a comment that explains why she has assigned the particular date range (text field).  Emily will find a way to display these data.

Stuff

Comments come in with / marks before all quotes….
How can we spell check entire existing database?

Google findability for all images and metadata

What do we do?  Can you create a means for this to happen?