HELP! I need advice on which approach to take with a text encoding project...

Digital Humanities Questions & Answers » Applications, Tools, Formats

HELP! I need advice on which approach to take with a text encoding project...

(7 posts) (5 voices)

Asked 5 years ago by Raskolnikov
Latest answer from Wayne Graham

Tags:

Raskolnikov
Member
Hi there,

I desperately need advice at the moment, and really hope some people here can provide some suggestions.

I have a fairly extensive data set in the forms of letters (from the 16th century) which have been encoded in XML.

My job is to create something which aids content analysis of the letters.

- This could be a GUI written in Java, which (via SAX or DOM) allows the user to extract certain features from the letters, compare them side by side rather than have to laboriously flick between letters on a computer system.

- Or it could be an elaborate XML project, where XSLT, HTML hyperlinks, and possibly some type of query language such as XQuery or more likely XPath would be used.

- Or it could be XML documents displayed in IE and using a JavaScript window/console to select, extract, manipulate, compare different letters.

Does anyone have an opinion or a suggestion on which of the three would be the best option to take?

Thank you very much in advance
Tweet this question
Posted 5 years ago Permalink
Dorothea Salo
Member

Which is easiest for you to write and maintain? Which works best across browsers? Which allows you to implement the features that users have asked you for? Which allows for a comprehensive API, so that third parties can write tools against the material as well?

Offhand, I'd vote thumbs-down on the Java GUI (hides too much), but I have no particular opinion about the other two.

Posted 5 years ago Permalink
Raskolnikov
Member

Thank you for your post.

Replying to @Dorothea Salo's post:

Posted 5 years ago Permalink
Raskolnikov
Member

I am not sure if the XML option would provide the level of interactivity I need, perhaps JavaScript would be the best option.

Replying to @Dorothea Salo's post:

Posted 5 years ago Permalink
Bethany Nowviskie
Admin

Dorothea's questions are good ones. I'd add these:

Could you say more about the scholarly or intellectual aims of the project? What kind of content analysis will end users want to do? What aspects of the letters will they be most interested in?

Are there other projects that do similar things with XML/TEI? (That is, are you sure you're not re-inventing the wheel?)

Posted 5 years ago Permalink
Stéfan Sinclair
Admin

I'd suggest you see what mileage you can get from existing frameworks and then decide how to build any custom functionality, if need be. Allowing comparison of letters, for instance, may be as simple as putting two instances of an interface in a window with iframes or something similar. Given that it's XML, you might consider trying XTF (http://xtf.cdlib.org/), which should allow you to define fields to index.

Posted 5 years ago Permalink
Wayne Graham
Member

One quick question, is this for you to do analysis, or for a more general audience?

Depending on the needs of that group should really drive the approach. If the analysis is for your own research, it may not be worth the effort to develop an interface; if its for a more general group, you should spend some time defining what exactly your target audience needs. There are a lot of ways to proceed, and I would encourage you to try some off-the-shelf software (like XTF that Stefan suggests, or try out something like this project https://github.com/mwmitchell/raven) and see how far that gets you down the road before developing any custom solution. The less time you spend with dealing with software internals, the more time you can dedicate toward the real part of your project.

Posted 5 years ago Permalink

RSS feed for this topic

Reply

You must log in to post.