Crowdsourcing Ur III
Once again I'm happy to see that people are interested in helping with the Ur transcriptions and that the option for presentation of the scans is overwhelmingly to give the document in its entirety and let you figure it out.
I have now created a site using Omeka and Scripto that allows for basic transcription. I'm currently in the process of testing the site and coming up with best practices / unification of effort help-files that will make sure we are all transcribing in the same way. I have a very small number of testers with me, making sure I can issue logins and passwords, that the site works on PC and Mac, and that it remains stable most of the time. We've managed to break the site several times so it's still in development stage. I have uploaded around 100 documents (early letters, 1919-1921, and some of Max Mallowan's notes) and begun transcriptions myself to see what problems people are likely to encounter. That feeds into my transcription guidelines, suggesting how to mark up strikethroughs, underlines, and marginalia. This all has to go through Mediawiki with Scripto, so html markup doesn't work. That's a bit of a pain, and it might mean that I should use another program, but setup on the Omeka site wasn't too bad overall, so we'll test this one and might change later onto FromThePage, which I like quite a bit, but have had more trouble setting up.
I agree that some sort of measure for completeness of transcription on individual files is a great idea, but at the moment I'm not sure how to program that. Each document shows up as a scan, so the computer doesn't know how many words are on that scan. Thus, it can't calculate the number of words left to transcribe and give a measure of completeness. In this iteration of the site it's even hard to find out what has been transcribed and what hasn't. I'm trying to find a way to indicate in Scripto whether a scan has any text transcribed or not, but so far I haven't been able to do that. It means that as a user it will be problematic since you would have to login to Scripto and click on document scans until you found one that had no text, or only partial text, associated with it. These scans also don't have much in the way of metadata. They don't really have titles since they are notes and the closest I have is the page number in the volume notebook they are stored in at the British Museum (which is the filename of the scan). This organization is not always correct, however, and we may find that one page joins with another farther down the line.
So, basically I'm asking that you all bare with me while I get the site working. It will likely go live still having some troubles and I will ask all of you to be my next level of testers. We can discuss problems with the Mediawiki pages both on individual transcriptions and on the site as a whole. I have not populated the Mediawiki site with many files as yet and may not have much time to fill in some help pages right away, but Mediawiki runs behind Scripto which populates pages as the transcriptions are created. The connection works overall and I think we can begin transcribing soon.
The only other issue is that I'm scheduled to go on vacation next week (diving in Bermuda!). That means that the site should go live in the second week of September, somewhere between the 10th and the 14th. I want to give everyone who participates credit for helping, so I have disabled anonymous functions. That means I will need to send you a login and password. I have many emails of interested transcribers already, but if I don't have your email address and you know you want to participate, you can send it to me at any time (whafford AT sas.upenn.edu) just be sure to include Ur Crowdsource in the subject line so I know where to store your request. I do not use your emails for other purposes than logins to the Scripto transcription software.
I will be the primary/only editor at first, so I hope I don't get overwhelmed. I'll try to answer questions on the site after it goes live, but do be patient with me.
Many thanks to all!