Comments on: Unexpected Progress

By: Investigations of a Dog » TEI Update

Investigations of a Dog » TEI Update — Wed, 02 Jan 2008 11:29:22 +0000

[...] first attempt at transforming the text to HTML using the TEI style sheets wasn’t very promising. I thought [...]

By: Gavin Robinson

Gavin Robinson — Thu, 19 Jul 2007 17:45:03 +0000

Thanks for your encouragement. There are so many exciting possibilities that one of the major problems is deciding where to start and finding time to do it.

By: mercuriuspoliticus

mercuriuspoliticus — Wed, 18 Jul 2007 19:06:19 +0000

This post and its predecessors were absolutely fascinating. It’s great to be able to read both about the theoretical and practical issues of transferring sources online in this way. As someone who is just finding his way into learning HTML and figuring out other possibilities the internet has to offer, it was really inspiring to learn what you’re up to. I’ll definitely be following the project with interest.

By: Brave new world « Mercurius Politicus

Brave new world « Mercurius Politicus — Wed, 18 Jul 2007 18:50:51 +0000

[...] post was partly inspired by Investigations of a Dog, where some really interesting thoughts on the internet and history have been posted. Thank you [...]

By: Gavin Robinson

Gavin Robinson — Thu, 12 Jul 2007 12:24:26 +0000

Thanks, I didn’t know about that but it looks really interesting. I’m hoping that when I’ve got this book finished I can work on ways of scaling things up, involving more people, and digitizing more books. I’m also wondering whether I could get a grant from anywhere to help with the costs.

By: Ross Mahoney

Ross Mahoney — Thu, 12 Jul 2007 11:35:02 +0000

Looks good. You might want to contact John Bourne at the Centre for First World War Studies as he is running an eBooks project. You can see the works he is looking at bringing back at http://www.firstworldwar.bham.ac.uk/ebooks/index.htm

By: Gavin Robinson

Gavin Robinson — Thu, 12 Jul 2007 09:10:36 +0000

I am intending to include page numbers and links from the original index but so far I’ve found it’s more difficult than I thought. I’ve kept pb elements in the XML source to mark the original pages, and giving them number attributes should be easy, but I haven’t worked out how to do it automatically. Oxygen doesn’t seem to have a tool for doing it. I thought I might be able to write a beanshell macro in jEdit but couldn’t get it to work as the replace function seems to screw up if you call it inside a loop. And that was before I realised that I haven’t used pb elements at the start of chapters. Now I have to decide whether to put pb elements at the start of chapters, or give the chapter div elements an id corresponding to the page number. I also need to include page numbers to make it easier to link to page images. I’m going to start learning Python soon anyway because I need it for other things, so that might offer a way of doing it. I also wonder if there’s any way of doing it with XSLT but I don’t know enough about that yet.

I just used a default TEI XSLT stylesheet which came bundled with Oxygen. I haven’t looked at it, but from this experience I’ll probably need to change a few things in it (and even more when I get to linking names together and linking to page images). Oxygen also allows you to set some parameters before you do a transform so I need to do some more experiments with those.

By: Ben Brumfield

Ben Brumfield — Wed, 11 Jul 2007 21:28:34 +0000

Another suggestion — could you add page break anchors to the document, then hotlink the pagenumbers in the index? I found myself wanting to click the numbers to find out where people were mentioned. Your eventual markup-based indexing may supplant the index within the document, but it seems like stripping out the page numbers loses information.

By: Ben Brumfield

Ben Brumfield — Wed, 11 Jul 2007 21:25:03 +0000

One other line-break suggestion — would you consider replacing the LB tags with a newline when you’re switching to word-wrap mode? The HTML should render the same, but anyone who does a “view source” on the document will find a much more readable text.

By: Ben Brumfield

Ben Brumfield — Wed, 11 Jul 2007 21:22:19 +0000

Congratulations! The resulting document looks great -- I found myself engaged by Private Gross and the Stokes mortar. I wonder if you might post the XSLT you used? Other people converting TEI to HTML might either be able to learn from it or suggest ways to incorporate some of your size reduction code into it. Regarding your line-break issue, I've run into something similar. In my intermediate XML format, I use TEI's lb tags to represent line breaks in the original document. At render time, I have the choice of converting those lb tags to either br tags or nothing, depending on user setting or view. (For per-page rendering I preserve line breaks, but for bloggy-style multi-page rendering I'm wrap lines. So this intermediate code

A clear day & colde Ben & Josie and Hellen & Virginia come to the House Half passed two oclock this morning Is rendered (link goes to dev site and may be problematic) as A clear day & colde Ben & Josie and Hellen & Virginia come to the House Half passed two oclock this morning