[sword-devel] KJV 1611
dmsmith at crosswire.org
Sat Dec 14 10:07:25 MST 2013
There are no GUI tools. Personally, I find that a sometimes GUI makes it harder not easier. The command line is really easy, but some break out in cold sweat thinking about it. I find that the command line for osis2mod is incredibly easy. But I avoid GUI interfaces if I can. But the module making tool will be the least of your effort.
The information is in the wiki (www.crosswire.org/wiki). There's a links on the home page to pretty much everything that you need to know. You can also find a transcription of Gen 1 for the 1611 KJV.
Regarding the effort, go for it. I've often wanted for us to have it, but we have not found a good, unencumbered* e-text. (* people often claim rights to the e-text for their effort in producing it and are unwilling to share)
Regarding encoding, there are a few unicode characters to use beyond what is currently encoded in ASCII. These include the long s and letters that are tied together, ae, oe, ff, ....
Note that using these makes the text more like the original, but make search hard. Currently the use will have to know which Unicode character is used and how to compose that on their keyboard. SWORD/JSword could normalize these when building the search index, but currently don't.
Presuming that the 1611 has the same words (perhaps w/ different spelling) in the same order as the 1769, we can apply the morphology and lemmas from our KJV text to it.
So the suggestions here are good.
First: Find an unencumbered text and get it into electronic form. Ensure that you have a reference text that you'll use as an authority. In working on our 1769 KJV, I found that there are differences between various printings of the KJV, even those purporting to be the "authoritative" version. Fortunately, there are several facsimiles of the 1611 KJV that can be used. Even in the first few printings of the 1611 there were textual errors. Some embarrassing and some amusing. If you do the transcription, avoid the temptation to fix what you think is wrong, but faithfully reproduce what you see. Example, the 1611 KJV is inconsistent in its use of punctuation. You may find spaces before punctuation or after it. Or maybe not.
As Chris said, if we had such a text, it would be a module already.
Second, proof read the text to ensure that it matches the reference text. It is very easy to have transcription errors. If possible, get a couple of different e-texts and compare them for differences. If they have independent histories of creation, it will help identify transcription errors. We did this for the KJV we have. It may be interesting to compare it to the 1769 KJV.
Third, encode it with an appropriate markup. We recommend OSIS. If the comparison to the 1769 KJV is favorable, then the 1769 markup could be laid over your effort.
Fourth, create a module. We recommend osis2mod. The effort to use osis2mod is so far easier than creating the text and validating it.
On Dec 14, 2013, at 11:01 AM, Israel <israeldahl at gmail.com> wrote:
> Sorry for the misinformation. I had thought someone had said there were
> no GUI tools, so making a module was not as simple as it could be.
> Please forgive my ignorance. I didn't realize it could be done so
> easily, do you have links to the info on how to do it, so I can be
> better informed and not make such misinformed comments in the future,
> and rather point someone to the right place? Thanks for the correction!!
> On 12/14/2013 09:24 AM, Chris Little wrote:
>> Hi Jeffrey,
>> There is literally no possibility that any characters in the 1611 KJV
>> have not already been encoded in Unicode. Don't concern yourself with
>> fonts at all. You don't need to make your own font. And we will not
>> distribute fonts with modules. Your only concern in this area is
>> encoding in Unicode, using the correct characters.
>> Once you have a document created in some standard format (OSIS or
>> USFM) or just a very regular text file that we can convert to OSIS,
>> compiling a module is trivial. It requires one command line call to
>> the appropriate module creation tool. (So ignore people who say
>> something is "a bit compilcated" followed by an admission that they've
>> never done it.)
>> We don't have a 1611 KJV text for you to work with. If we did, there
>> would already be a 1611 KJV module. So if you take this on, the bulk
>> of your work will probably be in creating the text (via OCR,
>> keyboarding, or some other method of procurement).
>> On 12/14/2013 5:15 AM, David Troidl wrote:
>>> The character thorn can be found in the Latin-1 Supplement block of
>>> Unicode: Þ
>>> I would suggest investigating the Latin Extended blocks. There are many
>>> fonts that support them. You may find exactly the characters you are
>>> looking for.
>>> You could also check with the Medieval Unicode Font Initiative:
>>> On 12/13/2013 8:51 PM, Israel wrote:
>>>> If you wanted to do this you would need an appropriate font. You may
>>>> have to actually make a font. I suggest using fontforge as it is free
>>>> (as in freedom, as well as no cost).
>>>> depending on your platform this could be extremely easy to install, or
>>>> you may have to go to the website to download it.
>>>> This work would be helpful if someone wanted to make the Tyndale's
>>>> version true to the original (as it has some errors, such as the
>>>> problems with "them" and other words that are mainly contained in an
>>>> archaic symbol).
>>>> Making a module is a bit complicated as of right now. But I am sure
>>>> someone can provide you with some scripts to make easier. Though I
>>>> have never made a module, so that is just my outside opinion
>>>> On 12/13/2013 06:39 PM, Barnes, Jeffrey wrote:
>>>>> Hi Swordsmen,
>>>>> One thing I’ve been hoping to see is someone interested enough to
>>>>> make a KJV 1611 module. I like the version for a few reasons, like
>>>>> the natural flow of the text, it’s poetic nature is beautiful to me,
>>>>> and the Reformation principle of clarity is followed imho.
>>>>> So since the Sword project doesn’t have a 1611 module yet, I’d like
>>>>> to investigate what’s involved in making it.
>>>>> One thing is that to be true to the 17th century printing, the
>>>>> alternate spellings of the text would have to be followed. The
>>>>> typeface used in the facsimiles I’ve found is a Gothic black letter
>>>>> face. I don’t want to use that, because it would make it unduly hard
>>>>> to read, especially on mobile and computer screens. I think one would
>>>>> need to use a roman, perhaps sans face for readability.
>>>>> But the roman faces that are used to render the text don’t use glyphs
>>>>> like the long ’s’, the ‘thorn’, the rotunda ‘r’, etc. I think those
>>>>> are important visual cues to the reading of the text. So if I would
>>>>> write a parser, it would replace the roman text source (probably
>>>>> ascii range code points) to replace glyphs with unicode according to
>>>>> the printing rules of the era.
>>>>> Is this work happening currently?
>>>>> If so, could I help?
>>>>> If not, where could I get a text source? I’ve seen facsimile
>>>>> renderings with roman parallel renderings, but they are a page at a
>>>>> time. It would be good if there was one or two files already with the
>>>>> roman characters.
>>>>> After the parser, then there’s the work of making it a module. Where
>>>>> can I find a procedure for doing that? Is it a manual process?
>>>>> I haven’t started any work yet, just thinking.
>>>>> Any help appreciated.
>>>>> sword-devel mailing list: sword-devel at crosswire.org
>>>>> Instructions to unsubscribe/change your settings at above page
>>> This email is free from viruses and malware because avast! Antivirus
>>> protection is active.
>>> sword-devel mailing list: sword-devel at crosswire.org
>>> Instructions to unsubscribe/change your settings at above page
>> sword-devel mailing list: sword-devel at crosswire.org
>> Instructions to unsubscribe/change your settings at above page
> sword-devel mailing list: sword-devel at crosswire.org
> Instructions to unsubscribe/change your settings at above page
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 4145 bytes
Desc: not available
More information about the sword-devel