Technical Matters.

As far as possible, I have tried to make the digitised version of the book follow the format of the original.

The text portions of the book were input using optical character recognition software, followed by human "sub-editing". In general the OCR worked very well, but there were certain combinations of letters which it found ambivalent. It is very likely that I have missed some of these, and so if you find any text which seems suspect, I should like to know. This applies in particular to personal names, as I should like the content of this book to be of use to genealogists searching the web.

You will notice that the images are of low quality (in the technical sense), and this is particularly noticeable in the text in those pictures. This low quality is selected specifically to cut down on the space required on the web-server, and time required to download. For search-engine purposes I have created "keyword" sections in these image pages, which contain any references in the photographic text.

I would have liked to link the page numbers in the index to the pages of the book, but this would have taken an immense amount of time, given the technology used. Instead I have opted to have a page which contains links for each page so that the reader can get to a specific page once it is found in the index. A link to this "links" page is provided at the bottom of each index page.

I have learned a great deal while digitising this book, and while I am quite satisfied with the result, I would not use the same methods again. This is a very static, HTML based solution, where in future I would certainly use a mixture of XML and CSS to facilitate automatic generation of page links etc., etc.

Jim Lawton .

back.