|
Germanic Lexicon Project
Message Board |
|
| [ Main Message Index ] [ Previous | Next ] | [ Reply ] |
Author: Sean Crist (Swarthmore College)
Email: kurisuto at unagi dot cis dot upenn dot edu
Date: 2004-11-18 13:06:18
Subject: A tale of two projects
Everybody,
I'd like to thank Bekie for her gracious post earlier today. I'd also like to thank Peter Tunstall for acting as referee.
Before I get to my main points, I need to apologize for something. In an earlier post, I made some comments about the scanning credits listed on Bekie's web site. It appears that someone in Bekie's team did in fact re-scan the entire main volume of Bosworth/Toller, so her site made no misattribution of credit. I had thought that Bekie's project was dead back in late 2003, but apparently this period of seeming inactivity was actually when the re-scanning was going on.
So far, so good. Now, on to the main stuff.
There are two points where it looks like Bekie and I are going to have to respectfully agree to disagree.
1. There are differences in the way our two projects encode the data. Bekie sees this as a major difference between the projects. I view the conversion of one encoding to another as a trivial process.
2. We differ as to whether it is a waste of effort for two teams to correct the same text. Bekie correctly points out that if there are two versions of the text, you can automatically compare the two to find errors. I agree that you can, and I naturally agree with the goal of greater accuracy. I simply believe that there are much less labor-intensive ways of achieving a comparable level of accuracy.
What matters to me is that there be an online edition of Bosworth/Toller which I can freely manipulate and redistribute without any intellectual property encumbrance. It doesn't matter to me how this comes about. Even though the text is out of copyright, I would not have been altogether comfortable copying the corrected data from Bekie's site if this were contrary to her wishes. However, Bekie said in her post that she has no problem with this. Excellent.
So it looks like the reasonable thing for me to do is to write a little script to convert Bekie's data to the form we're using here, substitute it for our existing data for the pages in question, mark those pages as completed, and of course credit Bekie's team properly for those pages.
I had proposed (in an off-the-board email, via Peter Tunstall) that the two teams divide up what's left of the main volume to avoid duplication of effort. Bekie does not wish to do this, for reasons she discussed in her post. I don't agree with her decision, but it's her decision to make, so I will simply do my best to rationally adapt my own strategy to these circumstances.
So what should you do if you're volunteering? If you want to avoid duplicating effort, then I recommend that you work on the supplement volume of Bosworth/Toller, or on Cleasby/Vigfusson. Neither Bekie nor anybody else is working on those volumes. It's OK if you still want to do pages in the main volume of Bosworth/Toller, but Bekie's team will probably end up duplicating your work later on, so I'd put a lower priority on that volume.
I think it would make sense to do the supplement and Cleasby/Vigfusson first, and then re-evaluate the situation. Perhaps Bekie's team will be done with the main volume at that point.
--Sean