![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
hrjWhen I first started posting entries for the Lesbian Historic Motif Project, each one had a set of keywords relevant to the content of the publication. Mostly I was considering "big picture themes" such as cross-dressing, sex between women, same-sex marriage, and so forth. At some point I dropped the keywords, not because I didn't think they were useful, but because I wanted to do something more systematic. That meant I'd need to go back through all the entries and re-code things, which was a bit like doing a tune-up on a moving car. And in the early stages, there were few enough publications that one could read through the titles and get a sense of which ones were worth browsing further. But now the New! Improved! version of the Alpennia website is in the structural coding phase and we're discussing exactly what sorts of things should be covered.
Indexing is an art form in itself, and past experience has taught me that you never really know what you want to index until you've processed a substantial chunk of text/data. So I'm now at a stage where I have a sense of what I, myself, might want to see in keyword word cloud. But perhaps my readers have additional ideas.
There are three major types of information that jump out at me for index/search purposes. The first is the sort of "major theme" topics that I tend to use when considering publications for inclusion, as mentioned above. These are the categories that answer the question, "What makes the subject of this publication 'lesbian-like'?" This would probably be a fairly small, closed category of topics, but I'm not sure that I've identified everything I would want to include or the level of specificity I'd want. (For example, same-sex marriage is different from romantic couplehood is different from female co-habitation without romantic bonds, but how specific do I want to be?)
The second type of information relates to specific individuals, characters, and events that recur across the data. Specific individuals who can either be categorized as lesbian or whose lives had lesbian-like characteristics. Specific literary figures whose stories involve lesbian-like themes. Specific court cases involving questions of sexuality or gender identity. Specific authors who treated lesbian-like themes in their work. So, for example, if a reader wanted to identify all the publications that discuss the case of Katharina Hetzeldorfer, or all the discussions of the medieval romance of Yde and Olive, or all the analyses of the poetry of Katherine Philips, that's something I'd like to make possible. (Of course, there's the question of whether I'm indexing only specific details mentioned in my posts, or whether I'm indexing all the individuals mentioned in the original text, many of which may have been cited only in passing.) This would necessarily be an open category.
The third type of generally useful information would be regarding the time and place covered by the publication. This can be a very fuzzy category at times. A publication focusing primarily on England may have brief passing mentions of material from other cultures. What level of presence is worth indexing? How specific should the references be? If an article discusses material from Islamic Spain, does it get counted as Spanish, as "Islamic cultures", as both, as something else? Even time periods aren't as straightforward as one might think. If a classical Roman author is writing about Sappho, should it be filed under his own century or under hers? For both time and place it will be useful to establish a clear structure to how the indexing will be handled, but it will need to be at least somewhat open-ended. (And I'm pretty sure that my Eurocentric bias will emerge in a tendency to use broader cultural categories the farther I get from that center.)
And, of course, every indexing system needs a "miscellaneous" category. There are themes that repeat regularly that I don't necessarily want to identify as "organizing principles". Things like references to the use of dildos in sex between women, or the topic of love poetry between women, and so forth. To some extent, I'll identify these tags in retrospect when I notice that a certain topic has been showing up regularly.
So what do you think, as users of the Project? What types of index and keyword information would you find most useful when treating the Project as a research reference?
Indexing is an art form in itself, and past experience has taught me that you never really know what you want to index until you've processed a substantial chunk of text/data. So I'm now at a stage where I have a sense of what I, myself, might want to see in keyword word cloud. But perhaps my readers have additional ideas.
There are three major types of information that jump out at me for index/search purposes. The first is the sort of "major theme" topics that I tend to use when considering publications for inclusion, as mentioned above. These are the categories that answer the question, "What makes the subject of this publication 'lesbian-like'?" This would probably be a fairly small, closed category of topics, but I'm not sure that I've identified everything I would want to include or the level of specificity I'd want. (For example, same-sex marriage is different from romantic couplehood is different from female co-habitation without romantic bonds, but how specific do I want to be?)
The second type of information relates to specific individuals, characters, and events that recur across the data. Specific individuals who can either be categorized as lesbian or whose lives had lesbian-like characteristics. Specific literary figures whose stories involve lesbian-like themes. Specific court cases involving questions of sexuality or gender identity. Specific authors who treated lesbian-like themes in their work. So, for example, if a reader wanted to identify all the publications that discuss the case of Katharina Hetzeldorfer, or all the discussions of the medieval romance of Yde and Olive, or all the analyses of the poetry of Katherine Philips, that's something I'd like to make possible. (Of course, there's the question of whether I'm indexing only specific details mentioned in my posts, or whether I'm indexing all the individuals mentioned in the original text, many of which may have been cited only in passing.) This would necessarily be an open category.
The third type of generally useful information would be regarding the time and place covered by the publication. This can be a very fuzzy category at times. A publication focusing primarily on England may have brief passing mentions of material from other cultures. What level of presence is worth indexing? How specific should the references be? If an article discusses material from Islamic Spain, does it get counted as Spanish, as "Islamic cultures", as both, as something else? Even time periods aren't as straightforward as one might think. If a classical Roman author is writing about Sappho, should it be filed under his own century or under hers? For both time and place it will be useful to establish a clear structure to how the indexing will be handled, but it will need to be at least somewhat open-ended. (And I'm pretty sure that my Eurocentric bias will emerge in a tendency to use broader cultural categories the farther I get from that center.)
And, of course, every indexing system needs a "miscellaneous" category. There are themes that repeat regularly that I don't necessarily want to identify as "organizing principles". Things like references to the use of dildos in sex between women, or the topic of love poetry between women, and so forth. To some extent, I'll identify these tags in retrospect when I notice that a certain topic has been showing up regularly.
So what do you think, as users of the Project? What types of index and keyword information would you find most useful when treating the Project as a research reference?
