Joshua wrote:

Jon Noring wrote:

...

And finally, since DP is the major player, should we move this design discussion to the relevant forum at DP? Or keep it here on gutvol-d?

If you want anything to happen any time soon, keep it here. DP will *not* be moving on this issue any time soon (not without an influx of new developers that decide to push forward on this issue, anyway). As JulietS said, there are plans to get there eventually, but there is a fair bit between where we are now and where that step comes into play.

O.k., so long as there's enough knowledgeable DP folk here, I'll not move discussion to one of the DP forums. A major issue is to assure the scan sets are findable and linkable to DP projects and to PG texts. This requires proper design of the metadata (such as identifer mapping), as well as careful organization of the scans (file and directory structure issues.) There is also the workflow issue: the repository system has to mesh pretty well with the DP system. It is probable that whatever will work for DP will not cause problems for PG text digitization projects done independent of DP. Thus, it is critical that some DP people contribute to the architectural design of the scan repository. If none of the important DP folk have any time at present to do so, then the repository design should be put on hold until the time the right DP folk can contribute. The worst thing that could be done now is to bull ahead with some sort of scan repository without sufficient feedback from the tech-folk at DP. What may then result is just a bigger mess that serves no one. (To reiterate, I see the metadata and identifier issue, and the directory structure, as the two main issues of the scan repository. There are others, but these two are pretty critical for the proper working and usefulness of the repository. On another note, we should discuss the general purposes of the scan repository(ies) -- this also guides the design process.) Jon