With completion of a new top plan for the Northeast Insulae Project, the “Plans” menu now has a full complement of plans (orthographic projections,DEM, walls, loci, and top plans in three formats). These plan will be used in the upcoming Final Report and offer the most consistent and complete overview of the work on the project.
The work is tedious, but a final report requires a new top plan. Working from an orthographic projection and other photographs, I have begun to construct that plan. [I do have to take breaks every hour or so to rest my eyes and wrist.] The layering possible with Adobe Illustrator makes for useful end products. Above, you see the NEC and some of the surrounding structures. More to come.
More updates …
There is a collection of new site plans posted to the “Plans” link on the menu.
Also, the core “Finds” galleries are uploading today and tomorrow (22-23 April 2021). They include ceramics, items, lamps, frescoes, metals, and architectural pieces.
The end of Adobe Flash corresponded with the beginning of my sabbatical. As a result, old galleries could not longer be used. In addition, in preparing the Final Report I wanted to update all galleries to HTML5 format, add new galleries, and sort through some 35,000 photos from the field.
I have completed the first step of this work by posting a new gallery of field photos for each day in the field between 2002 and 2019. For this project, I am using Adobe LightRoom. I have been able to correct white balance problems (especially from 2009 and 2010). I have removed less valuable photos. The photos are now in collections that I can more easily edit and annotate (and upload to the web site).
Next on my agenda is a paper for a regional conference on the herbs used in the diakonicon. Then, I return to the catalog of photos to build galleries of small finds — ceramics, metals, architectural elements, etc.
Over the last several weeks, I have been working to clean up the first complete lidar model of the site. The result is the following:
I solved a number of minor alignment problems and have rather comprehensive coverage. The model involves 146 scans with 384 links among the scans. There are close to 630 million points in the cloud.
The bundle is quite accurate, with an error of only 0.014 m.