Week Three Readings

Lesk
It's interesting to read an author's musings about how to retain control of the layout of a document while still allowing for flexibility of viewing options, when said musings are being displayed as a PDF. Likewise the discussion of different scanning options when reading a scanned page. The discussion of how OCR technology has progressed is encouraging, and it seems like the ease and cost-effectiveness of converting texts to sophisticated digital format with high functionality will only increase with time. Pitt subscribes to a historical newspaper collection (can't recall the name) that solves beautifully the display problems mentioned in 3.4. I didn't realize that CMU had such a large book-scanning project underway. Lesk reiterates the crucial point, however, that reading off paper is still preferable to most people than reading from a screen.

Arms
This chapter deals with a lot of the same issues I picked up in Lesk. I appreciate the in-depth discussion of Unicode, which I know is a vital standard for displaying diverse languages but I don't know much about (e.g. UTF-8 encoding). Ditto the explanation of DTDs and SGML. The section on XML is extremely helpful, as it's a concept I've had trouble grasping in the past.

Lynch
Identifier systems are a critically important aspect of digital libraries--and really any digital/networked collection--that I haven't given much thought. I think laypeople (like myself) tend to think that you have a URL or filename and that's all you need, but Lynch identifies several contexts and usages that require a different approach. Also, though I've seen them before (I believe through the Government Printing Office), I didn't realize that PURLs were an OCLC creation.