Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Troy A. Griffitts]

I've updated migratetags with my latest updates and extracted the
"matcher" logic out to its own class to make it easier to adjust per job.

http://crosswire.org/svn/sword-tools/trunk/migratetags/


On 4/14/19 7:41 AM, Tobias Klein wrote:
>
> Thanks, Troy! I'll have a look.
>
> Best regards,
> Tobias
>
> On 13.04.19 17:34, Troy A. Griffitts wrote:
>> The code I use for mapping across translations is in our tools repo
>> here. It does a reasonable job on same language literal Bibles and
>> isolates to word alignment algorithm to a single method so it can be
>> adjusted without understanding all the other code in there.
>>
>> http://crosswire.org/svn/sword-tools/trunk/migratetags/
>>
>> I think I have a newer matching algorithm I've used for adding
>> strongs to the Tyndale Greek New Testament and to NA28, which I still
>> need to check in, but the shell to do the work is generally there.
>> I'll try to checkin any updates soon.
>>
>> For what it's worth,
>>
>> Troy
>>
>>
>>
>> On April 13, 2019 6:12:28 AM MST, Jamie  wrote:
>>
>> David H is correct that David Instone Brewer has worked on this
>> kind of thing.  I won’t attempt to characterise how things stand
>> at present, nor the approach he has adopted, since I may be out
>> of date, and may end up misrepresenting things.  I did notice
>> this thread, though, and have drawn it to his attention, so I
>> imagine he’ll be in touch (he’s not answering my emails currently
>> so I assume he may be out for the day).
>>
>>  
>>
>> Jamie
>>
>>  
>>
>>  
>>
>> *From:*David Haslam [mailto:dfh...@protonmail.com]
>> *Sent:* 13 April 2019 13:37
>> *To:* SWORD Developers' Collaboration Forum
>> 
>> *Subject:* Re: [sword-devel] Mapping Strongs numbers to
>> translations that do not come with Strongs support
>>
>>  
>>
>> Further stuff written by memory from my hospital bed:
>>
>>  
>>
>> I myself developed a worksheet based editing environment so that
>> Vince LaRue has been able to start the big task of manually
>> adding Strong’s to the Spanish RV1865 (modern orthography) Source
>> Text. 
>>
>>  
>>
>> The WIP is in a shared folder in my Box account. 
>>
>>  
>>
>> The pre-&-postprocessing uses bespoke TextPipe filters. 
>>
>>  
>>
>> But there’s one more gain in the worksheet WPL environment. 
>>
>>  
>>
>> Providing the WPL worksheet includes a column with line numbers,
>> the text can be sorted on the Word column. 
>>
>>  
>>
>> This facilitated adding Strong’s to multiple instances of the
>> same Spanish word (ignoring context). The original textual order
>> can be restored by a sort on the line numbers column. 
>>
>>  
>>
>> Much more detail is involved in how (eg) punctuation is dealt
>> with and preserved. 
>>
>>  
>>
>> Even so, the trial run on 2JN was a success and Vince has since
>> done MAT 1-11 or more when I last looked at the progress. 
>>
>>  
>>
>> Aside: The same environment also facilitated adding markup for
>> \wj_...\wj*
>>
>>  
>>
>> Best regards,
>>
>>  
>>
>> David
>>
>>  
>>
>> Sent from ProtonMail Mobile
>>
>>  
>>
>>  
>>
>> On Sat, Apr 13, 2019 at 13:16, ref...@gmx.net
>>  mailto:ref...@gmx.net>>
>> wrote:
>>
>> I have been thinking about this for years but not done much yet.
>>
>> I think there are a bunch of steps which could make this a
>> machine driven process
>>
>> 1) use verse,mapping from KJV or other already tagged texts
>> (Synodal) and drop on each verse the relevant numbers
>>
>> 2) use names of places and people to tag those first
>>
>> 3) use dictionaries to align further
>>
>> 4) check by hand, but use any realignment to further inform
>> not yet checked verses
>>
>>
>> Sent from my mobile. Please forgive shortness, typos and
>> weird autocorrects.
>>
>>
>>
>>  Original Message 
>> Subject: Re: [sword-devel] Mapping Strongs numbers to
>> translations that do not come with Strongs support
>> From: Michael H
>> To: SWORD Developers' Collaboration Forum
>> CC:
>>
>>
>> The Unfolding Word Team is using Autographa for its
>> "alignment" process (which means adding strongs numbers,
>> but they also are working on fixing stray words into a
>> common (UnfoldingWord) versification across languages, if
>> I understand chat room babble. 
>>
>> https://forum.ccbt.bible/t/gl-ugnt-alignment-process/101
>>
>> http://www.autographa.com/about/
>>
>>  
>>
>> On Sat, Apr 13, 2019 at 6:52 AM Tobias Klein
>> mailto:cont...@tklein.info>> wrote:
>>
>> Hi David,
>>
>> Cool! Thanks for the hint! Do you 

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Robert Hunt]

  
  
Yes, this is what unfoldingWord is doing with (expanded) Strongs
  numbers as well as alignment to our open-licensed Greek text.
e.g.,
https://git.door43.org/unfoldingWord/en_ult/src/branch/master/57-TIT.usfm
Do however, please note that the unfoldingWord Literal Text above
  and our other texts are created as resources for Bible
  translators, not as end user English Bibles (which we tried to
  indicate by naming them "Text" rather than "Bible").

Our still rapidly developing translationCore software is used to
  create this mapping: https://www.translationcore.com/
We are still grappling with interpreting potential ambiguities in
  the new USFM 3.0 standard, so still a bit of a moving target.

Blessings,
  Robert.


On 13/04/19 11:08 PM, David Haslam
  wrote:


  
  For a translation already in USFM format, the target should
be to produce Strong’s added text in USFM 3.0 format using Word
Level Attributes. 
  
  
  Refer to the online documentation for USFM 3 at UBSICAP. 
  
  
  In other words, don’t leave the addition of Strongs’s to a
derivative output format such as OSIS. 
  
  
  
  
  Best regards,
  
  
  David
  
  
  Sent from ProtonMail
Mobile
  
  
  
  
  On Sat, Apr 13, 2019 at 11:33, Tobias Klein  wrote:
   Hi,

I have an idea that I would like to run by you guys.

Would it be possible to automatically map Strongs numbers to
translations that do not come with Strongs support?

The approach would be like this:
- Take a translation that comes with Strongs numbers
- Map each word (and Strongs number) to a corresponding word in
the
target translation, by using a regular dictionary

There may be some validation / manual checking needed when there
is not
a clear match between a word in the source translation and the
target
translation.
Furthermore, this would probably only work with pairs of
translations
that are both aiming to be literal. In that case the order of
words
would be very similar and would increase chances of mapping
words/Strongs correctly.

What do you think?
I'd be very happy to see Strongs mapped to German translations
specifically. But it could technically even work for
English/English
translation mapping.

Have a nice weekend!

Best regards,
Tobias


___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page
  
  
  
  
  
  
  
  ___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

  


___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Michael H]

I looked at the Linux version and its limited.  If you install the windows
version and find success, I'd like to know.

On Sun, Apr 14, 2019, 9:49 AM Tobias Klein  wrote:

> Thanks for the links, Michael. I'll have a look at Autographa.
>
> Best regards,
> Tobias
> On 13.04.19 14:03, Michael H wrote:
>
> The Unfolding Word Team is using Autographa for its "alignment" process
> (which means adding strongs numbers, but they also are working on fixing
> stray words into a common (UnfoldingWord) versification across languages,
> if I understand chat room babble.
>
> https://forum.ccbt.bible/t/gl-ugnt-alignment-process/101
>
> http://www.autographa.com/about/
>
>
>
> On Sat, Apr 13, 2019 at 6:52 AM Tobias Klein  wrote:
>
>> Hi David,
>>
>> Cool! Thanks for the hint! Do you happen to know whether that software
>> used by the STEP team is open source?
>>
>> Best regards,
>> Tobias
>> On 13.04.19 13:00, David Haslam wrote:
>>
>> This was done already by the Tyndale STEP team for adding Strong’s
>> Numbers to the ESV.
>>
>> They used bespoke software followed by manual adjustments.
>>
>> Ask David Instone-Brewer for details.
>>
>> Best regards,
>>
>> David
>>
>> Sent from ProtonMail Mobile
>>
>>
>> On Sat, Apr 13, 2019 at 11:33, Tobias Klein  wrote:
>>
>> Hi,
>>
>> I have an idea that I would like to run by you guys.
>>
>> Would it be possible to automatically map Strongs numbers to
>> translations that do not come with Strongs support?
>>
>> The approach would be like this:
>> - Take a translation that comes with Strongs numbers
>> - Map each word (and Strongs number) to a corresponding word in the
>> target translation, by using a regular dictionary
>>
>> There may be some validation / manual checking needed when there is not
>> a clear match between a word in the source translation and the target
>> translation.
>> Furthermore, this would probably only work with pairs of translations
>> that are both aiming to be literal. In that case the order of words
>> would be very similar and would increase chances of mapping
>> words/Strongs correctly.
>>
>> What do you think?
>> I'd be very happy to see Strongs mapped to German translations
>> specifically. But it could technically even work for English/English
>> translation mapping.
>>
>> Have a nice weekend!
>>
>> Best regards,
>> Tobias
>>
>>
>> ___
>> sword-devel mailing list: sword-devel@crosswire.org
>> http://www.crosswire.org/mailman/listinfo/sword-devel
>> Instructions to unsubscribe/change your settings at above page
>>
>>
>>
>>
>> ___
>> sword-devel mailing list: 
>> sword-devel@crosswire.orghttp://www.crosswire.org/mailman/listinfo/sword-devel
>> Instructions to unsubscribe/change your settings at above page
>>
>> ___
>> sword-devel mailing list: sword-devel@crosswire.org
>> http://www.crosswire.org/mailman/listinfo/sword-devel
>> Instructions to unsubscribe/change your settings at above page
>
>
> ___
> sword-devel mailing list: 
> sword-devel@crosswire.orghttp://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page
>
> ___
> sword-devel mailing list: sword-devel@crosswire.org
> http://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page
___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Tobias Klein]

Thanks for the links, Michael. I'll have a look at Autographa.

Best regards,
Tobias

On 13.04.19 14:03, Michael H wrote:
The Unfolding Word Team is using Autographa for its "alignment" 
process (which means adding strongs numbers, but they also are working 
on fixing stray words into a common (UnfoldingWord) versification 
across languages, if I understand chat room babble.


https://forum.ccbt.bible/t/gl-ugnt-alignment-process/101

http://www.autographa.com/about/



On Sat, Apr 13, 2019 at 6:52 AM Tobias Klein > wrote:


Hi David,

Cool! Thanks for the hint! Do you happen to know whether that
software used by the STEP team is open source?

Best regards,
Tobias

On 13.04.19 13:00, David Haslam wrote:

This was done already by the Tyndale STEP team for adding
Strong’s Numbers to the ESV.

They used bespoke software followed by manual adjustments.

Ask David Instone-Brewer for details.

Best regards,

David

Sent from ProtonMail Mobile


On Sat, Apr 13, 2019 at 11:33, Tobias Klein mailto:cont...@tklein.info>> wrote:

Hi,

I have an idea that I would like to run by you guys.

Would it be possible to automatically map Strongs numbers to
translations that do not come with Strongs support?

The approach would be like this:
- Take a translation that comes with Strongs numbers
- Map each word (and Strongs number) to a corresponding word in the
target translation, by using a regular dictionary

There may be some validation / manual checking needed when there
is not
a clear match between a word in the source translation and the
target
translation.
Furthermore, this would probably only work with pairs of
translations
that are both aiming to be literal. In that case the order of words
would be very similar and would increase chances of mapping
words/Strongs correctly.

What do you think?
I'd be very happy to see Strongs mapped to German translations
specifically. But it could technically even work for English/English
translation mapping.

Have a nice weekend!

Best regards,
Tobias


___
sword-devel mailing list: sword-devel@crosswire.org

http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page




___
sword-devel mailing list:sword-devel@crosswire.org  

http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

___
sword-devel mailing list: sword-devel@crosswire.org

http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page


___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page
___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Tobias Klein]

Thanks, Troy! I'll have a look.

Best regards,
Tobias

On 13.04.19 17:34, Troy A. Griffitts wrote:
The code I use for mapping across translations is in our tools repo 
here. It does a reasonable job on same language literal Bibles and 
isolates to word alignment algorithm to a single method so it can be 
adjusted without understanding all the other code in there.


http://crosswire.org/svn/sword-tools/trunk/migratetags/

I think I have a newer matching algorithm I've used for adding strongs 
to the Tyndale Greek New Testament and to NA28, which I still need to 
check in, but the shell to do the work is generally there. I'll try to 
checkin any updates soon.


For what it's worth,

Troy



On April 13, 2019 6:12:28 AM MST, Jamie  wrote:

David H is correct that David Instone Brewer has worked on this
kind of thing.  I won’t attempt to characterise how things stand
at present, nor the approach he has adopted, since I may be out of
date, and may end up misrepresenting things.  I did notice this
thread, though, and have drawn it to his attention, so I imagine
he’ll be in touch (he’s not answering my emails currently so I
assume he may be out for the day).

Jamie

*From:*David Haslam [mailto:dfh...@protonmail.com]
*Sent:* 13 April 2019 13:37
*To:* SWORD Developers' Collaboration Forum

*Subject:* Re: [sword-devel] Mapping Strongs numbers to
translations that do not come with Strongs support

Further stuff written by memory from my hospital bed:

I myself developed a worksheet based editing environment so that
Vince LaRue has been able to start the big task of manually adding
Strong’s to the Spanish RV1865 (modern orthography) Source Text.

The WIP is in a shared folder in my Box account.

The pre-&-postprocessing uses bespoke TextPipe filters.

But there’s one more gain in the worksheet WPL environment.

Providing the WPL worksheet includes a column with line numbers,
the text can be sorted on the Word column.

This facilitated adding Strong’s to multiple instances of the same
Spanish word (ignoring context). The original textual order can be
restored by a sort on the line numbers column.

Much more detail is involved in how (eg) punctuation is dealt with
and preserved.

Even so, the trial run on 2JN was a success and Vince has since
done MAT 1-11 or more when I last looked at the progress.

Aside: The same environment also facilitated adding markup for
\wj_...\wj*

Best regards,

David

Sent from ProtonMail Mobile

On Sat, Apr 13, 2019 at 13:16, ref...@gmx.net
 mailto:ref...@gmx.net>>
wrote:

I have been thinking about this for years but not done much yet.

I think there are a bunch of steps which could make this a
machine driven process

1) use verse,mapping from KJV or other already tagged texts
(Synodal) and drop on each verse the relevant numbers

2) use names of places and people to tag those first

3) use dictionaries to align further

4) check by hand, but use any realignment to further inform
not yet checked verses


Sent from my mobile. Please forgive shortness, typos and weird
autocorrects.



 Original Message 
Subject: Re: [sword-devel] Mapping Strongs numbers to
translations that do not come with Strongs support
From: Michael H
To: SWORD Developers' Collaboration Forum
CC:


The Unfolding Word Team is using Autographa for its
"alignment" process (which means adding strongs numbers,
but they also are working on fixing stray words into a
common (UnfoldingWord) versification across languages, if
I understand chat room babble.

https://forum.ccbt.bible/t/gl-ugnt-alignment-process/101

http://www.autographa.com/about/

On Sat, Apr 13, 2019 at 6:52 AM Tobias Klein
mailto:cont...@tklein.info>> wrote:

Hi David,

Cool! Thanks for the hint! Do you happen to know
whether that software used by the STEP team is open
source?

Best regards,
Tobias

On 13.04.19 13:00, David Haslam wrote:

This was done already by the Tyndale STEP team for
adding Strong’s Numbers to the ESV.

They used bespoke software followed by manual
adjustments.

Ask David Instone-Brewer for details.

Best regards,

David

Sent from ProtonMail Mobile

On Sat, Apr 13, 2019 at 11:33, Tobias Klein
mailto:cont...@tklein.info>>
wrote:

Hi,

I 

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Troy A. Griffitts]

The code I use for mapping across translations is in our tools repo here. It 
does a reasonable job on same language literal Bibles and isolates to word 
alignment algorithm to a single method so it can be adjusted without 
understanding all the other code in there.

http://crosswire.org/svn/sword-tools/trunk/migratetags/

I think I have a newer matching algorithm I've used for adding strongs to the 
Tyndale Greek New Testament and to NA28, which I still need to check in, but 
the shell to do the work is generally there. I'll try to checkin any updates 
soon.

For what it's worth,

Troy



On April 13, 2019 6:12:28 AM MST, Jamie  wrote:
>David H is correct that David Instone Brewer has worked on this kind of
>thing.  I won’t attempt to characterise how things stand at present,
>nor the approach he has adopted, since I may be out of date, and may
>end up misrepresenting things.  I did notice this thread, though, and
>have drawn it to his attention, so I imagine he’ll be in touch (he’s
>not answering my emails currently so I assume he may be out for the
>day).
>
> 
>
>Jamie
>
> 
>
> 
>
>From: David Haslam [mailto:dfh...@protonmail.com] 
>Sent: 13 April 2019 13:37
>To: SWORD Developers' Collaboration Forum 
>Subject: Re: [sword-devel] Mapping Strongs numbers to translations that
>do not come with Strongs support
>
> 
>
>Further stuff written by memory from my hospital bed:
>
> 
>
>I myself developed a worksheet based editing environment so that Vince
>LaRue has been able to start the big task of manually adding Strong’s
>to the Spanish RV1865 (modern orthography) Source Text. 
>
> 
>
>The WIP is in a shared folder in my Box account. 
>
> 
>
>The pre-&-postprocessing uses bespoke TextPipe filters. 
>
> 
>
>But there’s one more gain in the worksheet WPL environment. 
>
> 
>
>Providing the WPL worksheet includes a column with line numbers, the
>text can be sorted on the Word column. 
>
> 
>
>This facilitated adding Strong’s to multiple instances of the same
>Spanish word (ignoring context). The original textual order can be
>restored by a sort on the line numbers column. 
>
> 
>
>Much more detail is involved in how (eg) punctuation is dealt with and
>preserved. 
>
> 
>
>Even so, the trial run on 2JN was a success and Vince has since done
>MAT 1-11 or more when I last looked at the progress. 
>
> 
>
>Aside: The same environment also facilitated adding markup for
>\wj_...\wj*
>
> 
>
>Best regards,
>
> 
>
>David
>
> 
>
>Sent from ProtonMail Mobile
>
> 
>
> 
>
>On Sat, Apr 13, 2019 at 13:16, ref...@gmx.net  
>mailto:ref...@gmx.net> > wrote:
>
>I have been thinking about this for years but not done much yet. 
>
>I think there are a bunch of steps which could make this a machine
>driven process
>
>1) use verse,mapping from KJV or other already tagged texts (Synodal)
>and drop on each verse the relevant numbers
>
>2) use names of places and people to tag those first
>
>3) use dictionaries to align further
>
>4) check by hand, but use any realignment to further inform not yet
>checked verses
>
>
>Sent from my mobile. Please forgive shortness, typos and weird
>autocorrects.
>
>
>
> Original Message 
>Subject: Re: [sword-devel] Mapping Strongs numbers to translations that
>do not come with Strongs support
>From: Michael H 
>To: SWORD Developers' Collaboration Forum 
>CC: 
>
>
>
>
>The Unfolding Word Team is using Autographa for its "alignment" process
>(which means adding strongs numbers, but they also are working on
>fixing stray words into a common (UnfoldingWord) versification across
>languages, if I understand chat room babble. 
>
>https://forum.ccbt.bible/t/gl-ugnt-alignment-process/101
>
>http://www.autographa.com/about/
>
>
>
> 
>
>On Sat, Apr 13, 2019 at 6:52 AM Tobias Klein  > wrote:
>
>Hi David,
>
>Cool! Thanks for the hint! Do you happen to know whether that software
>used by the STEP team is open source?
>
>Best regards,
>Tobias
>
>On 13.04.19 13:00, David Haslam wrote:
>
>This was done already by the Tyndale STEP team for adding Strong’s
>Numbers to the ESV. 
>
> 
>
>They used bespoke software followed by manual adjustments. 
>
> 
>
>Ask David Instone-Brewer for details. 
>
> 
>
>Best regards,
>
> 
>
>David
>
> 
>
>Sent from ProtonMail Mobile
>
> 
>
> 
>
>On Sat, Apr 13, 2019 at 11:33, Tobias Klein  > wrote: 
>
>Hi,
>
>I have an idea that I would like to run by you guys.
>
>Would it be possible to automatically map Strongs numbers to
>translations that do not come with Strongs support?
>
>The approach would be like this:
>- Take a translation that comes with Strongs numbers
>- Map each word (and Strongs number) to a corresponding word in the
>target translation, by using a regular dictionary
>
>There may be some validation / manual checking needed when there is not
>a clear match between a word in the source translation and the target
>translation.
>Furthermore, this would probably only work with pairs of 

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Jamie]

David H is correct that David Instone Brewer has worked on this kind of thing.  
I won’t attempt to characterise how things stand at present, nor the approach 
he has adopted, since I may be out of date, and may end up misrepresenting 
things.  I did notice this thread, though, and have drawn it to his attention, 
so I imagine he’ll be in touch (he’s not answering my emails currently so I 
assume he may be out for the day).

 

Jamie

 

 

From: David Haslam [mailto:dfh...@protonmail.com] 
Sent: 13 April 2019 13:37
To: SWORD Developers' Collaboration Forum 
Subject: Re: [sword-devel] Mapping Strongs numbers to translations that do not 
come with Strongs support

 

Further stuff written by memory from my hospital bed:

 

I myself developed a worksheet based editing environment so that Vince LaRue 
has been able to start the big task of manually adding Strong’s to the Spanish 
RV1865 (modern orthography) Source Text. 

 

The WIP is in a shared folder in my Box account. 

 

The pre-&-postprocessing uses bespoke TextPipe filters. 

 

But there’s one more gain in the worksheet WPL environment. 

 

Providing the WPL worksheet includes a column with line numbers, the text can 
be sorted on the Word column. 

 

This facilitated adding Strong’s to multiple instances of the same Spanish word 
(ignoring context). The original textual order can be restored by a sort on the 
line numbers column. 

 

Much more detail is involved in how (eg) punctuation is dealt with and 
preserved. 

 

Even so, the trial run on 2JN was a success and Vince has since done MAT 1-11 
or more when I last looked at the progress. 

 

Aside: The same environment also facilitated adding markup for \wj_...\wj*

 

Best regards,

 

David

 

Sent from ProtonMail Mobile

 

 

On Sat, Apr 13, 2019 at 13:16, ref...@gmx.net   
mailto:ref...@gmx.net> > wrote:

I have been thinking about this for years but not done much yet. 

I think there are a bunch of steps which could make this a machine driven 
process

1) use verse,mapping from KJV or other already tagged texts (Synodal) and drop 
on each verse the relevant numbers

2) use names of places and people to tag those first

3) use dictionaries to align further

4) check by hand, but use any realignment to further inform not yet checked 
verses


Sent from my mobile. Please forgive shortness, typos and weird autocorrects.



 Original Message 
Subject: Re: [sword-devel] Mapping Strongs numbers to translations that do not 
come with Strongs support
From: Michael H 
To: SWORD Developers' Collaboration Forum 
CC: 




The Unfolding Word Team is using Autographa for its "alignment" process (which 
means adding strongs numbers, but they also are working on fixing stray words 
into a common (UnfoldingWord) versification across languages, if I understand 
chat room babble. 

https://forum.ccbt.bible/t/gl-ugnt-alignment-process/101

http://www.autographa.com/about/



 

On Sat, Apr 13, 2019 at 6:52 AM Tobias Klein mailto:cont...@tklein.info> > wrote:

Hi David,

Cool! Thanks for the hint! Do you happen to know whether that software used by 
the STEP team is open source?

Best regards,
Tobias

On 13.04.19 13:00, David Haslam wrote:

This was done already by the Tyndale STEP team for adding Strong’s Numbers to 
the ESV. 

 

They used bespoke software followed by manual adjustments. 

 

Ask David Instone-Brewer for details. 

 

Best regards,

 

David

 

Sent from ProtonMail Mobile

 

 

On Sat, Apr 13, 2019 at 11:33, Tobias Klein mailto:cont...@tklein.info> > wrote: 

Hi,

I have an idea that I would like to run by you guys.

Would it be possible to automatically map Strongs numbers to
translations that do not come with Strongs support?

The approach would be like this:
- Take a translation that comes with Strongs numbers
- Map each word (and Strongs number) to a corresponding word in the
target translation, by using a regular dictionary

There may be some validation / manual checking needed when there is not
a clear match between a word in the source translation and the target
translation.
Furthermore, this would probably only work with pairs of translations
that are both aiming to be literal. In that case the order of words
would be very similar and would increase chances of mapping
words/Strongs correctly.

What do you think?
I'd be very happy to see Strongs mapped to German translations
specifically. But it could technically even work for English/English
translation mapping.

Have a nice weekend!

Best regards,
Tobias


___
sword-devel mailing list: sword-devel@crosswire.org 
 
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

 

 

 

___
sword-devel mailing list: sword-devel@crosswire.org 
 

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[David Haslam]

Further stuff written by memory from my hospital bed:

I myself developed a worksheet based editing environment so that Vince LaRue 
has been able to start the big task of manually adding Strong’s to the Spanish 
RV1865 (modern orthography) Source Text.

The WIP is in a shared folder in my Box account.

The pre-&-postprocessing uses bespoke TextPipe filters.

But there’s one more gain in the worksheet WPL environment.

Providing the WPL worksheet includes a column with line numbers, the text can 
be sorted on the Word column.

This facilitated adding Strong’s to multiple instances of the same Spanish word 
(ignoring context). The original textual order can be restored by a sort on the 
line numbers column.

Much more detail is involved in how (eg) punctuation is dealt with and 
preserved.

Even so, the trial run on 2JN was a success and Vince has since done MAT 1-11 
or more when I last looked at the progress.

Aside: The same environment also facilitated adding markup for \wj_...\wj*

Best regards,

David

Sent from ProtonMail Mobile

On Sat, Apr 13, 2019 at 13:16, ref...@gmx.net  wrote:

> I have been thinking about this for years but not done much yet.
>
> I think there are a bunch of steps which could make this a machine driven 
> process
>
> 1) use verse,mapping from KJV or other already tagged texts (Synodal) and 
> drop on each verse the relevant numbers
>
> 2) use names of places and people to tag those first
>
> 3) use dictionaries to align further
>
> 4) check by hand, but use any realignment to further inform not yet checked 
> verses
>
> Sent from my mobile. Please forgive shortness, typos and weird autocorrects.
>
>  Original Message 
> Subject: Re: [sword-devel] Mapping Strongs numbers to translations that do 
> not come with Strongs support
> From: Michael H
> To: SWORD Developers' Collaboration Forum
> CC:
>
>> The Unfolding Word Team is using Autographa for its "alignment" process 
>> (which means adding strongs numbers, but they also are working on fixing 
>> stray words into a common (UnfoldingWord) versification across languages, if 
>> I understand chat room babble.
>>
>> https://forum.ccbt.bible/t/gl-ugnt-alignment-process/101
>>
>> http://www.autographa.com/about/
>>
>> On Sat, Apr 13, 2019 at 6:52 AM Tobias Klein  wrote:
>>
>>> Hi David,
>>>
>>> Cool! Thanks for the hint! Do you happen to know whether that software used 
>>> by the STEP team is open source?
>>>
>>> Best regards,
>>> Tobias
>>>
>>> On 13.04.19 13:00, David Haslam wrote:
>>>
 This was done already by the Tyndale STEP team for adding Strong’s Numbers 
 to the ESV.

 They used bespoke software followed by manual adjustments.

 Ask David Instone-Brewer for details.

 Best regards,

 David

 Sent from ProtonMail Mobile

 On Sat, Apr 13, 2019 at 11:33, Tobias Klein  wrote:

> Hi,
>
> I have an idea that I would like to run by you guys.
>
> Would it be possible to automatically map Strongs numbers to
> translations that do not come with Strongs support?
>
> The approach would be like this:
> - Take a translation that comes with Strongs numbers
> - Map each word (and Strongs number) to a corresponding word in the
> target translation, by using a regular dictionary
>
> There may be some validation / manual checking needed when there is not
> a clear match between a word in the source translation and the target
> translation.
> Furthermore, this would probably only work with pairs of translations
> that are both aiming to be literal. In that case the order of words
> would be very similar and would increase chances of mapping
> words/Strongs correctly.
>
> What do you think?
> I'd be very happy to see Strongs mapped to German translations
> specifically. But it could technically even work for English/English
> translation mapping.
>
> Have a nice weekend!
>
> Best regards,
> Tobias
>
> ___
> sword-devel mailing list: sword-devel@crosswire.org
> http://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page

 ___
 sword-devel mailing list:
 sword-devel@crosswire.org

 http://www.crosswire.org/mailman/listinfo/sword-devel
 Instructions to unsubscribe/change your settings at above page
>>>
>>> ___
>>> sword-devel mailing list: sword-devel@crosswire.org
>>> http://www.crosswire.org/mailman/listinfo/sword-devel
>>> Instructions to unsubscribe/change your settings at above page___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[David Haslam]

No - other than a loose description.

Ask David I-B and/or Jamie (ARA) Jamieson.

David

Sent from ProtonMail Mobile

On Sat, Apr 13, 2019 at 12:50, Tobias Klein  wrote:

> Hi David,
>
> Cool! Thanks for the hint! Do you happen to know whether that software used 
> by the STEP team is open source?
>
> Best regards,
> Tobias
>
> On 13.04.19 13:00, David Haslam wrote:
>
>> This was done already by the Tyndale STEP team for adding Strong’s Numbers 
>> to the ESV.
>>
>> They used bespoke software followed by manual adjustments.
>>
>> Ask David Instone-Brewer for details.
>>
>> Best regards,
>>
>> David
>>
>> Sent from ProtonMail Mobile
>>
>> On Sat, Apr 13, 2019 at 11:33, Tobias Klein  wrote:
>>
>>> Hi,
>>>
>>> I have an idea that I would like to run by you guys.
>>>
>>> Would it be possible to automatically map Strongs numbers to
>>> translations that do not come with Strongs support?
>>>
>>> The approach would be like this:
>>> - Take a translation that comes with Strongs numbers
>>> - Map each word (and Strongs number) to a corresponding word in the
>>> target translation, by using a regular dictionary
>>>
>>> There may be some validation / manual checking needed when there is not
>>> a clear match between a word in the source translation and the target
>>> translation.
>>> Furthermore, this would probably only work with pairs of translations
>>> that are both aiming to be literal. In that case the order of words
>>> would be very similar and would increase chances of mapping
>>> words/Strongs correctly.
>>>
>>> What do you think?
>>> I'd be very happy to see Strongs mapped to German translations
>>> specifically. But it could technically even work for English/English
>>> translation mapping.
>>>
>>> Have a nice weekend!
>>>
>>> Best regards,
>>> Tobias
>>>
>>> ___
>>> sword-devel mailing list: sword-devel@crosswire.org
>>> http://www.crosswire.org/mailman/listinfo/sword-devel
>>> Instructions to unsubscribe/change your settings at above page
>>
>> ___
>> sword-devel mailing list:
>> sword-devel@crosswire.org
>>
>> http://www.crosswire.org/mailman/listinfo/sword-devel
>> Instructions to unsubscribe/change your settings at above page___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[ref...@gmx.net]

I have been thinking about this for years but not done much yet. I think there are a bunch of steps which could make this a machine driven process1) use verse,mapping from KJV or other already tagged texts (Synodal)  and drop on each verse the relevant numbers2) use names of places and people to tag those first3) use dictionaries to align further4) check by hand, but use any realignment to further inform not yet checked versesSent from my mobile. Please forgive shortness, typos and weird autocorrects. Original Message Subject: Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs supportFrom: Michael H To: SWORD Developers' Collaboration Forum CC: The Unfolding Word Team is using Autographa for its "alignment" process (which means adding strongs numbers, but they also are working on fixing stray words into a common (UnfoldingWord) versification across languages, if I understand chat room babble. https://forum.ccbt.bible/t/gl-ugnt-alignment-process/101http://www.autographa.com/about/On Sat, Apr 13, 2019 at 6:52 AM Tobias Klein  wrote:
  

  
  
Hi David,
Cool! Thanks for the hint! Do you happen to know whether that
  software used by the STEP team is open source?
Best regards,
  Tobias

On 13.04.19 13:00, David Haslam wrote:


  
  This was done already by the Tyndale STEP team for adding
Strong’s Numbers to the ESV. 
  
  
  They used bespoke software followed by manual adjustments. 
  
  
  Ask David Instone-Brewer for details. 
  
  
  Best regards,
  
  
  David
  
  
  Sent from ProtonMail
Mobile
  
  
  
  
  On Sat, Apr 13, 2019 at 11:33, Tobias Klein  wrote:
   Hi,

I have an idea that I would like to run by you guys.

Would it be possible to automatically map Strongs numbers to
translations that do not come with Strongs support?

The approach would be like this:
- Take a translation that comes with Strongs numbers
- Map each word (and Strongs number) to a corresponding word in
the
target translation, by using a regular dictionary

There may be some validation / manual checking needed when there
is not
a clear match between a word in the source translation and the
target
translation.
Furthermore, this would probably only work with pairs of
translations
that are both aiming to be literal. In that case the order of
words
would be very similar and would increase chances of mapping
words/Strongs correctly.

What do you think?
I'd be very happy to see Strongs mapped to German translations
specifically. But it could technically even work for
English/English
translation mapping.

Have a nice weekend!

Best regards,
Tobias


___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page
  
  
  
  
  
  
  
  ___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

  

___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page
___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Michael H]

The Unfolding Word Team is using Autographa for its "alignment" process
(which means adding strongs numbers, but they also are working on fixing
stray words into a common (UnfoldingWord) versification across languages,
if I understand chat room babble.

https://forum.ccbt.bible/t/gl-ugnt-alignment-process/101

http://www.autographa.com/about/



On Sat, Apr 13, 2019 at 6:52 AM Tobias Klein  wrote:

> Hi David,
>
> Cool! Thanks for the hint! Do you happen to know whether that software
> used by the STEP team is open source?
>
> Best regards,
> Tobias
> On 13.04.19 13:00, David Haslam wrote:
>
> This was done already by the Tyndale STEP team for adding Strong’s Numbers
> to the ESV.
>
> They used bespoke software followed by manual adjustments.
>
> Ask David Instone-Brewer for details.
>
> Best regards,
>
> David
>
> Sent from ProtonMail Mobile
>
>
> On Sat, Apr 13, 2019 at 11:33, Tobias Klein  wrote:
>
> Hi,
>
> I have an idea that I would like to run by you guys.
>
> Would it be possible to automatically map Strongs numbers to
> translations that do not come with Strongs support?
>
> The approach would be like this:
> - Take a translation that comes with Strongs numbers
> - Map each word (and Strongs number) to a corresponding word in the
> target translation, by using a regular dictionary
>
> There may be some validation / manual checking needed when there is not
> a clear match between a word in the source translation and the target
> translation.
> Furthermore, this would probably only work with pairs of translations
> that are both aiming to be literal. In that case the order of words
> would be very similar and would increase chances of mapping
> words/Strongs correctly.
>
> What do you think?
> I'd be very happy to see Strongs mapped to German translations
> specifically. But it could technically even work for English/English
> translation mapping.
>
> Have a nice weekend!
>
> Best regards,
> Tobias
>
>
> ___
> sword-devel mailing list: sword-devel@crosswire.org
> http://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page
>
>
>
>
> ___
> sword-devel mailing list: 
> sword-devel@crosswire.orghttp://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page
>
> ___
> sword-devel mailing list: sword-devel@crosswire.org
> http://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page
___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Tobias Klein]

Hi David,

Cool! Thanks for the hint! Do you happen to know whether that software 
used by the STEP team is open source?


Best regards,
Tobias

On 13.04.19 13:00, David Haslam wrote:
This was done already by the Tyndale STEP team for adding Strong’s 
Numbers to the ESV.


They used bespoke software followed by manual adjustments.

Ask David Instone-Brewer for details.

Best regards,

David

Sent from ProtonMail Mobile


On Sat, Apr 13, 2019 at 11:33, Tobias Klein > wrote:

Hi,

I have an idea that I would like to run by you guys.

Would it be possible to automatically map Strongs numbers to
translations that do not come with Strongs support?

The approach would be like this:
- Take a translation that comes with Strongs numbers
- Map each word (and Strongs number) to a corresponding word in the
target translation, by using a regular dictionary

There may be some validation / manual checking needed when there is not
a clear match between a word in the source translation and the target
translation.
Furthermore, this would probably only work with pairs of translations
that are both aiming to be literal. In that case the order of words
would be very similar and would increase chances of mapping
words/Strongs correctly.

What do you think?
I'd be very happy to see Strongs mapped to German translations
specifically. But it could technically even work for English/English
translation mapping.

Have a nice weekend!

Best regards,
Tobias


___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page




___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page
___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[David Haslam]

For a translation already in USFM format, the target should be to produce 
Strong’s added text in USFM 3.0 format using Word Level Attributes.

Refer to the online documentation for USFM 3 at UBSICAP.

In other words, don’t leave the addition of Strongs’s to a derivative output 
format such as OSIS.

Best regards,

David

Sent from ProtonMail Mobile

On Sat, Apr 13, 2019 at 11:33, Tobias Klein  wrote:

> Hi,
>
> I have an idea that I would like to run by you guys.
>
> Would it be possible to automatically map Strongs numbers to
> translations that do not come with Strongs support?
>
> The approach would be like this:
> - Take a translation that comes with Strongs numbers
> - Map each word (and Strongs number) to a corresponding word in the
> target translation, by using a regular dictionary
>
> There may be some validation / manual checking needed when there is not
> a clear match between a word in the source translation and the target
> translation.
> Furthermore, this would probably only work with pairs of translations
> that are both aiming to be literal. In that case the order of words
> would be very similar and would increase chances of mapping
> words/Strongs correctly.
>
> What do you think?
> I'd be very happy to see Strongs mapped to German translations
> specifically. But it could technically even work for English/English
> translation mapping.
>
> Have a nice weekend!
>
> Best regards,
> Tobias
>
> ___
> sword-devel mailing list: sword-devel@crosswire.org
> http://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

Re: [sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[David Haslam]

This was done already by the Tyndale STEP team for adding Strong’s Numbers to 
the ESV.

They used bespoke software followed by manual adjustments.

Ask David Instone-Brewer for details.

Best regards,

David

Sent from ProtonMail Mobile

On Sat, Apr 13, 2019 at 11:33, Tobias Klein  wrote:

> Hi,
>
> I have an idea that I would like to run by you guys.
>
> Would it be possible to automatically map Strongs numbers to
> translations that do not come with Strongs support?
>
> The approach would be like this:
> - Take a translation that comes with Strongs numbers
> - Map each word (and Strongs number) to a corresponding word in the
> target translation, by using a regular dictionary
>
> There may be some validation / manual checking needed when there is not
> a clear match between a word in the source translation and the target
> translation.
> Furthermore, this would probably only work with pairs of translations
> that are both aiming to be literal. In that case the order of words
> would be very similar and would increase chances of mapping
> words/Strongs correctly.
>
> What do you think?
> I'd be very happy to see Strongs mapped to German translations
> specifically. But it could technically even work for English/English
> translation mapping.
>
> Have a nice weekend!
>
> Best regards,
> Tobias
>
> ___
> sword-devel mailing list: sword-devel@crosswire.org
> http://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page

[sword-devel] Mapping Strongs numbers to translations that do not come with Strongs support

[Tobias Klein]

Hi,

I have an idea that I would like to run by you guys.

Would it be possible to automatically map Strongs numbers to 
translations that do not come with Strongs support?


The approach would be like this:
- Take a translation that comes with Strongs numbers
- Map each word (and Strongs number) to a corresponding word in the 
target translation, by using a regular dictionary


There may be some validation / manual checking needed when there is not 
a clear match between a word in the source translation and the target 
translation.
Furthermore, this would probably only work with pairs of translations 
that are both aiming to be literal. In that case the order of words 
would be very similar and would increase chances of mapping 
words/Strongs correctly.


What do you think?
I'd be very happy to see Strongs mapped to German translations 
specifically. But it could technically even work for English/English 
translation mapping.


Have a nice weekend!

Best regards,
Tobias


___
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page