Comprehensive NWT Comparison Project (calling all technically skilled members) (page 10)

RayPublisher

I said:

"Couldn't you just generate a two-column document that has each translation matching up verse for verse next to each other like a parallel Bible?"

You said:

"Unfortunately we won't be able to legally distribute that since it will be composed of the Society's copyrighted material."

But if someone already has obtained both copies legally (Like many of us have, by downloading them off JW.org) then they can legally view such a document.

It's like someone running MAME on their PC and playing a ROM image of Space Invaders. It is legal to do so as long as you own the original ROM. (And yes I do have a rom chip from space invaders believe it or not!)

MeanMrMustard

Why not just publish the document via a torrent... They couldn't do anything even if they wanted to.

Apognophos

Yes RayPublisher, I was thinking along those lines myself. If you look at my option B above, it would be like a transparency that you lay over the Bible PDF that you already downloaded from the Society.

Re: torrents, I guess that's an option, but it's not really user-friendly, and if we can make a legitimate release that is user-friendly, all the better, right?

MeanMrMustard

@castthefirststone:

That line of text that you have there really really looks like a hex encoding for unicode... but its not decoding with the standard decode functions.

MMM

castthefirststone

You have a PM MeanMRMustard.

Here are the first few lines of the table def:

CREATE TABLE bible_verse ( id INTEGER PRIMARY KEY, label TEXT,content BLOB );
PRAGMA foreign_keys=OFF;
BEGIN TRANSACTION;
CREATE TABLE bible_verse ( id INTEGER PRIMARY KEY, label TEXT,content BLOB );
INSERT INTO "bible_verse" VALUES(0,'1',X'F24BE36C0A668F8AFF3A76BACB40E612ED1E7BEFC5BDC3C04133FA424568A2A5AF6DC3DF1465C064A3B01E473A65AC3DFE09C541B734C65CE3C732CE467B5CFEA4BF8F443272C5AEAB2526F862FE57830ACB62E2E86B4BD2A565F0533F92888A');
INSERT INTO "bible_verse" VALUES(1,'2',X'5F4B2966493DDAFCAE578A54E7495F74368DFA89E366C74AE2484F8FDB9BBA50D8F55CF0F6282A2E5518F67D55191DB7850C0EDF7828435DE1C52BA43E10F191CBFBB833CC6151C3E0E39F7783330CC2442BD106348EB8D8E234C410EE419B0782780A81F3CFB496B82BFB8156E9A4EE4DDB561E74617294F4E0A13A37F933E4A276253A45CF42B06A4CA946824027CCCBD5B94618E8F1ED9C20DFA494785844');
INSERT INTO "bible_verse" VALUES(2,'3',X'C7A211EDE177A30E4317AD65BFFD9139871073CCEDAE5B852D00B59D4BB7BB4B5EA3FD65FA40AB2CD30B8CC624A90CEEB85EAD617BB831878AA074EA105ECA0EF66F2596A0C3FA42EF758282A03BFB13EFCCBC7E9AD386D3DB806605EB1FCA64');

MeanMrMustard

castthefirststone,

Thanks! I have the DB now. I'll parse through it tonight. I sent back another PM - I have some family business to attend to.

MMM

MeanMrMustard

castthefirststone,

Yes, I have the database, and I can see all the tables and structures there. A lot of the tables and fields are exactly what I would expect to find in a DB for any bible. But the actual content is a different story. The content field on the bible_book and bible_verse tables is just a binary BLOB field. The question is how to interpret that field. It is interesting that the hex length (display the field as hex, and then take the length) is always an even number. It would suggest that each character is 2 hex digits, or just a byte (encode each char as UTF-8, plus some for markup of some sort?). That would make sense too because the length of the hex string is often 25% larger than actual character count of any verse. But alas, it doesn't seem to be encoded in any standard way... I've even tried changing the endianness of each byte (just in case they did a casting or something during the display process).

I wonder - and I don't know for sure, but I'm leaning toward it - if this field is actually meant to be a binary MEPS blob. That is, it's the WT's proprietary MEPS format, and I don't know how they encode the data with that format. I don't even know if anyone outside the WT knows how to decode MEPS.

MMM

slimboyfat

Can you translate that into: this is feasible/not feasible and you can do it soon/can't do it ever? For those of us who don't understand the details. Thanks for your hard work either way.

MeanMrMustard

slimboyfat,

The quick translation - In the WT DB, they have the entire bible encoded as a binary blob of data. It is probably MEPS, but I don't know for sure. I don't think it is any of the standard way of encoding text. I may fuss with it some more, or try to pull it out of the PDF next. Can I do it? Probably. Time table? We'll have to see what else comes up in the PDF parse.

MMM

Simon

Just a quick scan of the text makes it look like it's encoded. There are no obvious repeating characters as would be the case for space and 'e' etc... and the range of codes means its less likely that there is a one-to-one translation from hex pair to character.

Maybe it's compressed (which for shorter texts could be slightly larger than the original / plain) but it doesn't have the gzip prefix (maybe stripped, maybe some other compression).

I'm sure people on some encryption / encoding technical forum would be able to help work it out (esp if you can provide the target text)

Comprehensive NWT Comparison Project (calling all technically skilled members)

Share this