Page 1 of 2

AT Robertson Grammar

Posted: May 21st, 2011, 11:14 am
by Louis L Sorenson
A.T. Robertson's 'A Grammar of the New Testament in Light of Historical Research (1921), is one of the best resources for the advanced Greek student - often when no other person comments on a construction, Robertson talks about it. His book is in the public domain and has been digitized into an MS Word document and PDF by Ted Hildebrandt at Gordon College. You can find his versions at http://faculty.gordon.edu/hu/bi/Ted_Hil ... WebBib.htm. He has granted B-Greek the right to host it's online version.

Louis Sorenson has converted the Greek and Hebrew fonts to UNICODE, and created a temporary web page for the html version at http://www.letsreadgreek.org/resources/ ... mmar/html/. The file is currently one massive htlm document (also available in an MS Word document at http://www.letsreadgreek.org/resources/ ... ongrammar/. But that version is full of MS junk formatting code. It would be nice to get a clean HMTL version.

There are a number of things yet to do on the HTML version, which could be done by members of B-Greek
Yet to Do:
  • Chunk the grammar into manageable size, whether page or chapter
  • Proof read the grammar for errors.
  • Divide indexes into separate columns. Currently the flow is as a single column for the whole page.
  • Put footnotes in consecutive order (on separate lines)
  • Add hyperlinked bookmarks (page numbers) at the start of every page (for html edition)
  • Create a paragraph version and remove the carriage returns and hyphenation at the end of most lines.
  • Add links to the Additional notes in the main body of the text
We need a person skilled in HTML management, a set of code tools, etc. The easiest way to create pages would be to write HTML extraction code and create new html pages directly from MS Word, managing links and web-formatting styles at the same time. If you would like to participate in this B-Greek project, send Louis Sorenson a private message or post a reply here. It is a great way to read in depth Robertson's grammar.

Re: AT Robertson Grammar

Posted: May 27th, 2011, 2:46 pm
by Jonathan Robie
At this point, are you making your changes to the .doc or to the html? Which serves as the current source?

Re: AT Robertson Grammar

Posted: May 27th, 2011, 3:13 pm
by Louis L Sorenson
The Word Doc is the source at this point. All exports are full of crud,even when they are cleaned up by MS Word HTML cleaners. What is really needed is
1) Clear styles which are compatible with web and linked to a CSS stylesheet. No mso junk.
2) Decision to export the Greek/Hebrew as html entities e.g. &0952 or as Unicode characters. I'm not sure the benefits and detriments of both.
3) Chunk size. - chapter?
4) Page links - some mechanism for querying the correct page, as that is the way most books refer to the Grammar.
5) Some really good export programs (they look really good) use absolute positioning. But I don't think that would be best. (I'm not a web programmer).

Re: AT Robertson Grammar

Posted: May 27th, 2011, 4:29 pm
by Jonathan Robie
llsorenson wrote:The Word Doc is the source at this point. All exports are full of crud,even when they are cleaned up by MS Word HTML cleaners. What is really needed is
1) Clear styles which are compatible with web and linked to a CSS stylesheet. No mso junk.
I think you've used html tidy and still have issues here, right? I haven't tried it on your html, it usually works well for the smaller things I work with.

One odd thing about your html is that every line is a separate paragraph, I assume that's true of the .doc file as well. I see similar things using standard tools to convert to DocBook and other formats with your .doc file. I suspect this was created using an OCR reader and the paragraphs are used to avoid line flow.

Would you be OK with a different paragraph flow than the original, assuming you could have references to the start of each page?
llsorenson wrote:2) Decision to export the Greek/Hebrew as html entities e.g. &0952 or as Unicode characters. I'm not sure the benefits and detriments of both.
I'm a big fan of Unicode, specifically utf-8 precombined characters, for this kind of thing.
llsorenson wrote:3) Chunk size. - chapter?
Yes, I would think so.
llsorenson wrote:4) Page links - some mechanism for querying the correct page, as that is the way most books refer to the Grammar.
I suppose that's going to be somewhat manual. Neither HTML nor XML have a well defined concept of links. I'd probably suggest adding anchors with names like #page1, #page2, etc. at appropriate places, and linking to them.
llsorenson wrote:5) Some really good export programs (they look really good) use absolute positioning. But I don't think that would be best. (I'm not a web programmer).
Not sure how you're using the term absolute positioning here. Do you mean page images that are visually identical, with the same positioning, as opposed to simply containing the same content?

Re: AT Robertson Grammar

Posted: May 27th, 2011, 5:17 pm
by Louis L Sorenson
I meant like what is seen in the ClickToConvert version http://www.letsreadgreek.org/resources/ ... rChap3.htm

Re: AT Robertson Grammar

Posted: May 27th, 2011, 6:06 pm
by Jonathan Robie
What about a PDF formatted something like this?

http://www.ibiblio.org/bgreek/project/r ... rammar.pdf

Or is it important to have this as html pages?

Re: AT Robertson Grammar

Posted: May 27th, 2011, 6:26 pm
by Louis L Sorenson
I think small web pages would be best - the quickest. Not all can handle a 30meg download. Think of those in the third world. Plus the fact that html pages can be used inside other applications.

A pdf is useful on one's own system. The problem I've had with the pdf is that I cannot get any pdf printer to get all the bookmarks correct. (Ted Hildebrandt has an indexed one, but the text is not Unicode but uses non-unicode thGreek/thHebrew fonts). I've tried quite a number pdf exporters about a year ago, many could not handle the Greek/Hebrew. None get the indexes correct. The Bookmarks are all based off of the headings in the Word document. There are 6-7 levels of headings.

Those headings need to be styles - some contain Greek or Hebrew words, which also need styles in order to specify the font. That's also one of the issues I've been struggling with. How many styles? In a pdf you can embed the fonts. It's a little different with an html version.

Re: AT Robertson Grammar

Posted: May 28th, 2011, 7:18 am
by Jonathan Robie
I see 241 headers in the generated HTML. OOWriter reports 247 headings - is that the discrepancy you see, or is there a greater discrepancy?

Re: AT Robertson Grammar

Posted: May 28th, 2011, 9:54 am
by Louis L Sorenson
I've only gotten 1 out of 5 to show up. The outline in the early pages is what needs to be counted - every one of those outline points has a reference. Have you added those Bookmarks to the pdf?

Re: AT Robertson Grammar

Posted: May 30th, 2011, 8:27 pm
by Jonathan Robie
llsorenson wrote:I've only gotten 1 out of 5 to s]how up. The outline in the early pages is what needs to be counted - every one of those outline points has a reference. Have you added those Bookmarks to the pdf?
I didn't add anything, but I may have gotten the bookmarks to generate correctly.

Did you check the bookmarks in the PDF I posted earlier in this thread?

http://www.ibiblio.org/bgreek/project/r ... rammar.pdf

The outline of bookmarks looks essentially like the detailed table of contents - I haven't checked them all, but none of the ones I have checked so far are missing.