Startup person selection

classic Classic list List threaded Threaded
63 messages Options
1234
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

Peter Flynn
On 23/12/2018 17:24, Nick Hall wrote:
> Yes.  Exactly.  We are looking at unstructured text such as newspaper
> articles and notices, obituaries, wills, biographies...
>
> The idea would be to use a special text editor to give meaning to parts
> of a transcription.  

Like TEI does (although overkill here).

> This could be stored as a tagged or marked-up
> note.  Then we could use a tool to automatically generate objects from
> this data.
>
> I have already written a prototype editor, but was too ambitious and
> attempted to add support for tables and lists.  I was not entirely happy
> with the result.

I'm not sure that it's necessary to reinvent this wheel...there are
plenty of data formats and editors for them already. Perhaps a
MarkDown-like format?

P


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Startup person selection

Gary Bussiere
In reply to this post by Dave Scheipers
On 12/23/18 3:59 PM, Dave Scheipers wrote:

> I would prefer to attach the citation to the obituary, a narrative.
> But for some reason, a note cannot be sourced and cited except within
> the body of the text.

I pretty much do the same thing. Attach the Obituary to the Person's Note.

But I also make a pdf copy of the Obituary, keeping that in my Media
folder. Making that the Source with the transcript as a Note.

Then I can attach that Note to the Person. Otherwise it doesn't show in
the People view, except in the Source Citations tab.

Nice thing about Gramps, it's very flexible about how to enter
information. Took me some time to just get the basics down or a workflow
to enter information that would suit me. Still learning new or better ways.


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

GRAMPS - User mailing list
In reply to this post by Peter Flynn
Please don't use an acronym without defining it in a discussion amongst readers spanning all knowledge levels.

I had to visit 5 websites with TEI expert claiming to have introductory texts before anyone defined the term.  One "About TEI" webpage praised how great TEI was, using the acronym 56 times without once defining it, it's purpose or identifying the administering entity. Frustrating.

TEI (Text Encoding Initiative). See http://www.tei-c.org/about/frequently-asked-questions/#section-3 

When you visit that page and then the guidelines, you'll find that it still never identifies the basic purpose of TEI. When it even APPROACHES basic concepts, the documents ALWAYS fallback on jargon.

This is a prime example of the worst thing about computer related specialities. We tend to let ourselves get so caught up in jargon & detail that we alienate the people that might benefit most from tools being built. 

The Wiki is an extensive technical reference. But it presents a nearly unassailably massive wall of text that is too intimidating. Many of the questions asked here daily have answers in the Wiki but there's too much too wade through.

When somebody asks a question here, I try to find the subject in the Wiki and and post the links. On basic concepts, I'm often finding it faster to write a paragraph or 2 from scratch. 

Another few months and it should give enough background to outine an introductory text that is less intimidating but which still links to the tutorials, videos & reference texts.

On Sun, Dec 23, 2018 at 15:47, Peter Flynn
On 23/12/2018 17:24, Nick Hall wrote:
> Yes.  Exactly.  We are looking at unstructured text such as newspaper
> articles and notices, obituaries, wills, biographies...
>
> The idea would be to use a special text editor to give meaning to parts
> of a transcription. 

Like TEI does (although overkill here).

> This could be stored as a tagged or marked-up
> note.  Then we could use a tool to automatically generate objects from
> this data.
>
> I have already written a prototype editor, but was too ambitious and
> attempted to add support for tables and lists.  I was not entirely happy
> with the result.

I'm not sure that it's necessary to reinvent this wheel...there are
plenty of data formats and editors for them already. Perhaps a
MarkDown-like format?


P



_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

ACProctor
This post was updated on .
In reply to this post by Peter Flynn
CONTENTS DELETED
The author has deleted this message.
Reply | Threaded
Open this post in threaded view
|

Re: Startup person selection

Ron Johnson
In reply to this post by Dave Scheipers

But the obituary is the citation.

What you can do is attach the Note with the transcript to both the Citation and the Person.

On 12/23/18 2:59 PM, Dave Scheipers wrote:
For me, the obituary is nothing more than a narrative of the person's life.

While I will attempt to glean as much information as possible about
the person's family from the obituary, I view that information as a
clue, not a fact. So I do not attach a citation to those clues.

I would prefer to attach the citation to the obituary, a narrative.
But for some reason, a note cannot be sourced and cited except within
the body of the text.

On Sun, Dec 23, 2018 at 2:36 PM Ron Johnson [hidden email] wrote:
Out of curiosity, why?  After all, the obit is a citation.

On 12/23/18 10:00 AM, Dave Scheipers wrote:
Sorry for the empty email. What I meant to send....

I put Obituaries in the Person's Note tab. Not events, etc.

But to get around keeping the note active while the person and events
are closed.... Copy the Note to the Clipboard and then open that note.
It stays active while you search other people and fill in their
portion of the information from  the obit.

Dave




On Sun, Dec 23, 2018 at 10:54 AM Dave Scheipers
[hidden email] wrote:
On Sun, Dec 23, 2018 at 10:44 AM Emyoulation--- via Gramps-users
[hidden email] wrote:
Yay!!!!!!!

Can I suggest an Obituary/Biography as the test case?  I often find myself expanding/refining nearby Persons to the active Relationship using these.

Althoigh error-prone, a newspaper obit suggests many, many things: where the living relatives resided by that date, the sibs & their mate's given name.

Obits can refine a blank death date for the sibs. Estimating a death date for a Living sib goes from "ESTIMATED BETWEEN puberty (or maybe marriage date minus shotgun leadtime?) of mother AND death of mother plus max. lifespan" to "CALCULATED BETWEEN publication date AND death of mother plus max. lifespan". The upper range limit can be refined by the obit publication date when a sib/spouse/parent/child preceded in death.

But it's VERY awkward keeping the obit Text visible as you manually data-mine such information.  Since it is a Note in a Citation in a Source of an Event of a Person in a Relationship, there are 5 Windows open to keep the Note visible for data-mining, plus the clipboard for the citation, plus the 5 Windows to drill down on the Family member to refine the Event & attach the source to it. (I think there's a Note gramplet that could help but I never think about opening it until I'm too deeply engaged in the process.) Re-arranging all those windows so they don't hide the Note text is an incredible time-suck.

I often lose track of who's been refined when working a large family obit. And the Reference on the transcription listed the Citations, their References list the Source and the (periodical) Source lists all the Events related to any Obit logged for that Periodical. (My hometown newspapers Source has a LOT of Obits, Social reports, news clippings which span generations of extended family.

On Sun, Dec 23, 2018 at 8:30, Nick Hall
[hidden email] wrote:

We plan to move more towards source-based data entry in the future.

Nick.

--
Angular momentum makes the world go 'round.


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

ACProctor
In reply to this post by Peter Flynn
Apologies for the posts and deletions folks. My post contained some sample
HTML and it got formatted, even though the "message contains HTML" checkbox
was NOT ticked. I tried various alternatives but they were either screwed up
in the email notifications or eliminated completely in the forum. If there
is a way of including sample HTML without it getting interpreted then I
haven't worked it out.

Tony



--
Sent from: http://gramps.1791082.n4.nabble.com/GRAMPS-User-f1807095.html


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Startup person selection

Dave Scheipers
In reply to this post by Ron Johnson
Hi Ron,

I was just stating how, and why I handle obituaries. This is what
works for me. You have found a practice that works for you.

And I will give another benefit of the person note.... the obituary
will print within the person's record on many reports. Citations are
aggregated at the end of the person's record or at the end of a
report. And putting the obit text in the person's record as well as a
citation that is then attached to either the person or to one or more
events, has the obit printing in both places. Once is enough for me
and I prefer it with the person.

We each do what works for ourselves. The one flaw I find with Gramps
is that a proper citation cannot be attached to text in a note record
that has come from someone else other than me.


On Sun, Dec 23, 2018 at 6:18 PM Ron Johnson <[hidden email]> wrote:

>
>
> But the obituary is the citation.
>
> What you can do is attach the Note with the transcript to both the Citation and the Person.
>
> On 12/23/18 2:59 PM, Dave Scheipers wrote:
>
> For me, the obituary is nothing more than a narrative of the person's life.
>
> While I will attempt to glean as much information as possible about
> the person's family from the obituary, I view that information as a
> clue, not a fact. So I do not attach a citation to those clues.
>
> I would prefer to attach the citation to the obituary, a narrative.
> But for some reason, a note cannot be sourced and cited except within
> the body of the text.
>
> On Sun, Dec 23, 2018 at 2:36 PM Ron Johnson <[hidden email]> wrote:
>
> Out of curiosity, why?  After all, the obit is a citation.
>
> On 12/23/18 10:00 AM, Dave Scheipers wrote:
>
> Sorry for the empty email. What I meant to send....
>
> I put Obituaries in the Person's Note tab. Not events, etc.
>
> But to get around keeping the note active while the person and events
> are closed.... Copy the Note to the Clipboard and then open that note.
> It stays active while you search other people and fill in their
> portion of the information from  the obit.
>
> Dave
>
>
>
>
> On Sun, Dec 23, 2018 at 10:54 AM Dave Scheipers
> <[hidden email]> wrote:
>
> On Sun, Dec 23, 2018 at 10:44 AM Emyoulation--- via Gramps-users
> <[hidden email]> wrote:
>
> Yay!!!!!!!
>
> Can I suggest an Obituary/Biography as the test case?  I often find myself expanding/refining nearby Persons to the active Relationship using these.
>
> Althoigh error-prone, a newspaper obit suggests many, many things: where the living relatives resided by that date, the sibs & their mate's given name.
>
> Obits can refine a blank death date for the sibs. Estimating a death date for a Living sib goes from "ESTIMATED BETWEEN puberty (or maybe marriage date minus shotgun leadtime?) of mother AND death of mother plus max. lifespan" to "CALCULATED BETWEEN publication date AND death of mother plus max. lifespan". The upper range limit can be refined by the obit publication date when a sib/spouse/parent/child preceded in death.
>
> But it's VERY awkward keeping the obit Text visible as you manually data-mine such information.  Since it is a Note in a Citation in a Source of an Event of a Person in a Relationship, there are 5 Windows open to keep the Note visible for data-mining, plus the clipboard for the citation, plus the 5 Windows to drill down on the Family member to refine the Event & attach the source to it. (I think there's a Note gramplet that could help but I never think about opening it until I'm too deeply engaged in the process.) Re-arranging all those windows so they don't hide the Note text is an incredible time-suck.
>
> I often lose track of who's been refined when working a large family obit. And the Reference on the transcription listed the Citations, their References list the Source and the (periodical) Source lists all the Events related to any Obit logged for that Periodical. (My hometown newspapers Source has a LOT of Obits, Social reports, news clippings which span generations of extended family.
>
> On Sun, Dec 23, 2018 at 8:30, Nick Hall
> <[hidden email]> wrote:
>
> We plan to move more towards source-based data entry in the future.
>
> Nick.
>
>
> --
> Angular momentum makes the world go 'round.
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Startup person selection

Ron Johnson
On 12/23/18 6:05 PM, Dave Scheipers wrote:
Hi Ron,

I was just stating how, and why I handle obituaries. This is what
works for me. You have found a practice that works for you.

And I will give another benefit of the person note.... the obituary
will print within the person's record on many reports. Citations are
aggregated at the end of the person's record or at the end of a
report. And putting the obit text in the person's record as well as a
citation that is then attached to either the person or to one or more
events, has the obit printing in both places. Once is enough for me
and I prefer it with the person.

We each do what works for ourselves. The one flaw I find with Gramps
is that a proper citation cannot be attached to text in a note record
that has come from someone else other than me.


Notes are just that... notes.  You could attach the obit citation to the Person.

On Sun, Dec 23, 2018 at 6:18 PM Ron Johnson [hidden email] wrote:

But the obituary is the citation.

What you can do is attach the Note with the transcript to both the Citation and the Person.

On 12/23/18 2:59 PM, Dave Scheipers wrote:

For me, the obituary is nothing more than a narrative of the person's life.

While I will attempt to glean as much information as possible about
the person's family from the obituary, I view that information as a
clue, not a fact. So I do not attach a citation to those clues.

I would prefer to attach the citation to the obituary, a narrative.
But for some reason, a note cannot be sourced and cited except within
the body of the text.

On Sun, Dec 23, 2018 at 2:36 PM Ron Johnson [hidden email] wrote:

Out of curiosity, why?  After all, the obit is a citation.

On 12/23/18 10:00 AM, Dave Scheipers wrote:

Sorry for the empty email. What I meant to send....

I put Obituaries in the Person's Note tab. Not events, etc.

But to get around keeping the note active while the person and events
are closed.... Copy the Note to the Clipboard and then open that note.
It stays active while you search other people and fill in their
portion of the information from  the obit.

Dave




On Sun, Dec 23, 2018 at 10:54 AM Dave Scheipers
[hidden email] wrote:

On Sun, Dec 23, 2018 at 10:44 AM Emyoulation--- via Gramps-users
[hidden email] wrote:

Yay!!!!!!!

Can I suggest an Obituary/Biography as the test case?  I often find myself expanding/refining nearby Persons to the active Relationship using these.

Althoigh error-prone, a newspaper obit suggests many, many things: where the living relatives resided by that date, the sibs & their mate's given name.

Obits can refine a blank death date for the sibs. Estimating a death date for a Living sib goes from "ESTIMATED BETWEEN puberty (or maybe marriage date minus shotgun leadtime?) of mother AND death of mother plus max. lifespan" to "CALCULATED BETWEEN publication date AND death of mother plus max. lifespan". The upper range limit can be refined by the obit publication date when a sib/spouse/parent/child preceded in death.

But it's VERY awkward keeping the obit Text visible as you manually data-mine such information.  Since it is a Note in a Citation in a Source of an Event of a Person in a Relationship, there are 5 Windows open to keep the Note visible for data-mining, plus the clipboard for the citation, plus the 5 Windows to drill down on the Family member to refine the Event & attach the source to it. (I think there's a Note gramplet that could help but I never think about opening it until I'm too deeply engaged in the process.) Re-arranging all those windows so they don't hide the Note text is an incredible time-suck.

I often lose track of who's been refined when working a large family obit. And the Reference on the transcription listed the Citations, their References list the Source and the (periodical) Source lists all the Events related to any Obit logged for that Periodical. (My hometown newspapers Source has a LOT of Obits, Social reports, news clippings which span generations of extended family.

On Sun, Dec 23, 2018 at 8:30, Nick Hall
[hidden email] wrote:

We plan to move more towards source-based data entry in the future.

Nick.

--
Angular momentum makes the world go 'round.


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

Janusz S. Bień
In reply to this post by GRAMPS - User mailing list
On Sun, Dec 23 2018 at 22:53 GMT, Emyoulation--- via Gramps-users wrote:
> Please don't use an acronym without defining it in a discussion amongst readers spanning all knowledge levels.

TEI was mentioned on the list not long ago.

>
> I had to visit 5 websites with TEI expert claiming to have
>introductory texts before anyone defined the term. One "About TEI"
>webpage praised how great TEI was, using the acronym 56 times without
>once defining it, it's purpose or identifying the administering
>entity. Frustrating.
>
> TEI (Text Encoding Initiative). See
> http://www.tei-c.org/about/frequently-asked-questions/#section-3
>
> When you visit that page and then the guidelines, you'll find that it
> still never identifies the basic purpose of TEI.  When it even
> APPROACHES basic concepts, the documents ALWAYS fallback on jargon.

What about

http://www.tei-c.org/

The Text Encoding Initiative (TEI) is a consortium which collectively
develops and maintains a standard for the representation of texts in
digital form. Its chief deliverable is a set of Guidelines which specify
encoding methods for machine-readable texts, chiefly in the humanities,
social sciences and linguistics. Since 1994, the TEI Guidelines have
been widely used by libraries, museums, publishers, and individual
scholars to present texts for online research, teaching, and
preservation.

?

BTW, an XML editor with the best support of TEI is oxygen:

https://www.oxygenxml.com/

It's a commercial one, but you can test the full version free for a
month (unless this changed recently).

Best regards

Janusz


--
             ,  
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

Nick Hall
In reply to this post by Peter Flynn
On 23/12/2018 21:46, Peter Flynn wrote:
I have already written a prototype editor, but was too ambitious and attempted to add support for tables and lists.  I was not entirely happy with the result.

I'm not sure that it's necessary to reinvent this wheel...there are plenty of data formats and editors for them already. Perhaps a MarkDown-like format?

Possibly, but STEMMA is probably the best I've found so far.  TEI is overkill, as you say, and not really suited to our purpose.

We should specify a format that is easy to write in an external editor, but we should also give users the ability to create "tagged" transcriptions within Gramps.

Nick.




_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

Peter Flynn
In reply to this post by GRAMPS - User mailing list
On 23/12/2018 22:53, [hidden email] wrote:
> Please don't use an acronym without defining it in a discussion amongst
> readers spanning all knowledge levels.

Quite right, sorry about that. I've spent too long working with that
stuff...I should get out more.

> Frustrating.[...] TEI (Text Encoding Initiative).
> See http://www.tei-c.org/about/frequently-asked-questions/#section-3
>
> When you visit that page and then the guidelines, you'll find that it
> still never identifies the basic purpose of TEI. When it even
> APPROACHES basic concepts, the documents ALWAYS fall back on jargon.
First sentence: "a standard for the representation of texts in digital
form". No jargon that I can see, but it's not very informative.

The Text Encoding Initiative is an XML vocabulary for marking up
("describing") literary, linguistic, and historical texts in the
Humanities. (XML is the Extensible Markup Language, the same thing used
for defining HTML.)

On 23/12/2018 17:24, Nick Hall wrote:
> The idea would be to use a special text editor to give meaning to
> parts of a transcription.
TEI does EXACTLY this, as do many other XML systems like DocBook, and
even to some extent HTML. A "special text editor" in these terms only
has to be an ordinary XML editor.

[Emyoulation]
But yes, later on it gets very messy and IMNSHO needs a full rewrite.

> This is a prime example of the worst thing about computer related
> specialities. We tend to let ourselves get so caught up in jargon &
> detail that we alienate the people that might benefit most from tools
> being built.

The underlying principle (historically since before the start of the
Internet) is that if you think it might be of use, you have to read the
fine documentation and learn it. This is an inadequate approach
nowadays, as most people are too busy surviving and don't have the time
to read and learn in that way. Learning by doing is more fun anyway.

> The Wiki is an extensive technical reference. But it presents a nearly
> unassailably massive wall of text that is too intimidating. Many of the
> questions asked here daily have answers in the Wiki but there's too much
> too wade through.

Wikis — in general — are usually written in systems with poor indexing,
so finding anything is virtually impossible, even when the word you are
looking for exists in an entry, it doesn't get returned as one of the hits.

Worse, because of the problem you describe above, the beginner simply
doesn't have the specialist vocabulary needed to search for what they
want. Try searching the Gramps Wiki for 'add second wife' :-) which is a
perfectly normal thing for a new user to want to do...but that's not how
it's phrased.

> When somebody asks a question here, I try to find the subject in the
> Wiki and and post the links. On basic concepts, I'm often finding it
> faster to write a paragraph or 2 from scratch.

When someone asks a question in my own fields, I do much the same.
I don't think anyone has yet come up with a workable way to proceed,
other than asking the new user to read a long document describing what
the system is and what it can do.

There *is* of course a better way: make the software interface so
obvious that you don't need the manual. But that is *hard*  and it goes
against the grain of many people, who view things in a different light,
and want the process to be formative.

P


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Markup test (was: Re: Source-based data entry

Peter Flynn
In reply to this post by ACProctor
On 23/12/2018 23:21, ACProctor wrote:
> Apologies for the posts and deletions folks. My post contained some sample
> HTML and it got formatted,

Was that done by mailman or your mail user agent?

I'm using Thunderbird in plaintext mode. Here is the word 'text'
enclosed in a HTML span with the class attribute set to 'foo':

<span class="foo">text</span>

Does that appear as markup or has the markup been stripped?

P


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

enno
In reply to this post by Nick Hall
Hello all,

For years, at least since I saw my late father using Microsoft Word to
save data, research notes, and stories that he couldn't store in
Brother's Keeper, I have dreamt of linking documents to my tree, using
bookmarks, special styles, whatever.

Today, I use LibreOffice on Linux, and I have also tried to save web
pages on Evernote, until they became too commercial, so now, I often
save pages in Pocket, which comes as a standard add-on in Firefox.

Either way, most of the text that I encounter today comes from the web,
and modern word processors can all save data using XML, so for me it
seems pretty straightforward to have an import function, or a simple
editor that stores data as XML, or XHTML, which is a term I learned from
Evernote, I think.

This also reminds me of a subject called Microdata, which was mentioned
on the RootsDev mailing list, I guess. That's many years ago, when we
still thought that we could collaborate to create a new standard.

Anyway, this Microdata is documented on https://schema.org/docs/gs.html 
and it is already used by FamilySearch and Geni, just to name a few. You
can check that in some page sources on these sites.

As I understand it, Microdata is just a way to put data on a web page,
by enclosing it with span tags to avoid interpretation, but the schema
behind it can be used everywhere, so why not use it in a word processor.

In some respects its not as powerful as STEMMA, but it is a standard, right?

Regards,

Enno



_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

Peter Flynn
On 24/12/2018 20:33, Enno Borgsteede wrote:
> Either way, most of the text that I encounter today comes from the
> web, and modern word processors can all save data using XML, so for
> me it seems pretty straightforward to have an import function,

It would depend heavily on what XML vocabulary it was. Word can only
export WordML/OOXML or XHTML; Libre Office exports ODF, XHTML, or Writer
Layout XML; Google Docs exports OOXML, ODF, and EPUB (XHTML); and
AbiWord exports OOXML, ODF, EPUB (XHTML), WML, and DocBook. So any
import function would need to be able to make sense of at least a few of
those.

(I won't define all of those now or I'd be here until morning :-)

> or a simple editor that stores data as XML, or XHTML, which is a term
> I learned from Evernote, I think.

The number of XML editors that could be classed as 'simple' is very
small, unfortunately, as most of them are designed to be able to handle
anything the XML standard defines (which is a LOT). There are only two
open source ones (XMLCopyEditor and Emacs with psgml or nxml).

XHTML is just the modern version of HTML, which was originally defined
before XML was invented.

> This also reminds me of a subject called Microdata, which was
> mentioned on the RootsDev mailing list, I guess. That's many years
> ago, when we still thought that we could collaborate to create a new
> standard.

Alas yes.

> Anyway, this Microdata is documented on
> https://schema.org/docs/gs.html and it is already used by
> FamilySearch and Geni, just to name a few. You can check that in some
> page sources on these sites.

I'm not sure how much Microdata is used these days. RDFa seems to have
gained a lot of traction.

> As I understand it, Microdata is just a way to put data on a web
> page, by enclosing it with span tags to avoid interpretation,

Yes, browsers will ignore them, but the span tags could be interpreted
by any conforming software that can read XML.

> but the schema behind it can be used everywhere, so why not use it in
> a word processor.
I think it does get used by some of the specialist embedded
(Javascript?) text editor modules used by publishers. So yes, it would
be useful if it could be used in wordprocessors.

> In some respects its not as powerful as STEMMA, but it is a standard,
> right?

No, just an application within XML. But it could be used to add value to
XHTML notes.

P


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Markup test

ACProctor
In reply to this post by Peter Flynn
Thanks for the advice, Peter (yes, that came through fine). Although I normally use Thunderbird, on this occasion I was doing it through the Web page.
When I wake up a bit (long night, last night) I'll try and send it the normal way. It was merely indicating that TEI is not ideal for historical text,
and that off-the-shelf HTML editors can be made to edit custom XML mark-up. I hope that's not too technical for this list.

Tony

On 24/12/2018 20:08, Peter Flynn wrote:

> On 23/12/2018 23:21, ACProctor wrote:
>> Apologies for the posts and deletions folks. My post contained some sample
>> HTML and it got formatted,
>
> Was that done by mailman or your mail user agent?
>
> I'm using Thunderbird in plaintext mode. Here is the word 'text' enclosed in a HTML span with the class attribute set to 'foo':
>
> <span class="foo">text</span>
>
> Does that appear as markup or has the markup been stripped?
>
> P
>
>
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org
>


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Markup test

Janusz S. Bień
On Tue, Dec 25 2018 at 10:59 GMT, Tony Proctor wrote:

[...]

> It was merely indicating that TEI is not ideal
> for historical text

I think nobody claims that TEI is ideal for anything.

As for historical texts, looks like almost all numerous projects

http://www.tei-c.org/activities/projects/

deal with historical texts, not necessarily in diplomatic
transcription. Perhaps in "Journal of the Text Encoding Initiative" you
can find more example.

I think it is true that "the best tool is the tool you know best". For
somebody who has to learn a tool from the scratch TEI seems a viable
option. It is complicated because it tries to be comprehensive. For a
specific application you need only a subset of it.

Merry Christmas!

Janusz  

--
             ,  
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Markup test

ACProctor
In reply to this post by ACProctor
Although I've probably posted these notes in a different list already, that would have been a while back.

STEMMA has comprehensive mark-up designed specifically for historical text (i.e. semantic content rather than presentation) and so I've been down this road already. I abandoned TEI becuse it is very complicated and designed primarily for diplomatic transcription.

However, I have experimented with customisable HTML editors such as CKEditor and Textbox.io. Although my mark-up is XML, these editors can easily be fooled into editing it via HTML using "data attributes". For instance, consider a hypothetical mark-up example of

<Place Id='place-id'>place reference</Place>
 
Each piece of custom mark-up will generally have both semantic aspects (the place reference and the ID in my example above) and presentational aspects (how do I want to see that marked-up reference depicted). If the custom mark-up is replaced with a <span> or <div> (dependent upon whether it's inline or block oriented) of a specific class then it can represent both types of aspect, and it should be acceptable to any good HTML editor, such as CKEditor.

As a more complicated example, consider this STEMMA XML mark-up that has several attribute values:

<Anom Mode='Intralinear' Posn='T' Ref='^'> some text </Anom>

This could be converted as follows to make it acceptable to the editor (where "xx" is some unique proprietary name to avoid name clashes):

<span class="xx-anom" data-xx-mode="Intralinear" data-xx-posn="T" data-xx-ref="^"> some text >/span>

As well as capturing the attribute values, this allows the editor to display different pieces of marked-up text according to different styles, using some CSS that has selectors such as span.xx-anom.

Apologies for the technical information but I though it might be useful, somewhere.

Tony



_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Markup test

Janusz S. Bień
On Tue, Dec 25 2018 at 13:36 GMT, Tony Proctor wrote:
> Although I've probably posted these notes in a different list already,
> that would have been a while back.
>
> STEMMA has comprehensive mark-up designed specifically for historical
> text (i.e. semantic content rather than presentation) and so I've been
> down this road already. I abandoned TEI becuse it is very complicated

That definitely true.

> and designed primarily for diplomatic transcription.

I don't think it is true. If you insist on this statement, please
provide some arguments.

I don't use TEI but I watched its development almost from its beginning
and my impression is that diplomatic transcription is just one of many
possible applications.

Sorry for pressing the point, but I would like to make sure that the
subscribers of this list are not misinformed about TEI.

Best regards

Janusz

--
             ,  
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

ACProctor
In reply to this post by Peter Flynn
Markdown is a simplified mark-up for presentational aspects. I am not aware of what the Gramps scope is, but when I refer to "historical text", I am
implying that "semantic markup" would be used. This is not just formatting (e.g. that something was underlined) otherwise I could use plain HTML. But
it's not even just marking personal names, places references, dates, etc -- sometimes I want to connect those references to corresponding entities in
my data, e.g. connecting a reference to "my grandmother" (which is a person reference albeit not by personal name) to Annie Elizabeth Proctor in my
data. This is part of the reason why I avoided TEI (the other being that it's far too complicated), and why I came up with a private format.

Tony

On 23/12/2018 21:46, Peter Flynn wrote:

> On 23/12/2018 17:24, Nick Hall wrote:
>> Yes.  Exactly.  We are looking at unstructured text such as newspaper articles and notices, obituaries, wills, biographies...
>>
>> The idea would be to use a special text editor to give meaning to parts of a transcription.
>
> Like TEI does (although overkill here).
>
>> This could be stored as a tagged or marked-up note.  Then we could use a tool to automatically generate objects from this data.
>>
>> I have already written a prototype editor, but was too ambitious and attempted to add support for tables and lists.  I was not entirely happy with
>> the result.
>
> I'm not sure that it's necessary to reinvent this wheel...there are plenty of data formats and editors for them already. Perhaps a MarkDown-like
> format?
>
> P
>
>
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Source-based data entry [Was: Startup person selection]

GRAMPS - User mailing list
I suggest that this thread be moved to the Developer mail-list. 

Better yet, rather than flood the Developer mail-list with over-eager pie-in-the-sky comments (I'm guilty of that), maybe there should be a Gramps-RAD mail-list where thick-skinned users (those who are willing to Beta-test, who expect to deal with crash & burns) can bounce ideas off developers & each other. New discussions could be announced here with a subscribe link. Then you unsubscribe when that discussion is done.... or you start to feel the burn out.

That said, RAD (Rapid Application Development) is changing the way to discuss software design.  Perhaps we ought to just TRY the proof-of-concept that Nick built. There's NO commitment to the approach of a RAD prototype and there's no expectation that it will do anything other than corrupt the test Tree.  It just provides a common frame of reference. Then we can have a discussion about the relative merits of one vs. another approach. 

We're still in the brainstorming phase but have already diverged into Standards adoption.

See
https://clearbridgemobile.com/beginners-guide-poc-vs-mvp-vs-prototype/


On Tue, Dec 25, 2018 at 7:56, Tony Proctor
Markdown is a simplified mark-up for presentational aspects. I am not aware of what the Gramps scope is, but when I refer to "historical text", I am
implying that "semantic markup" would be used. This is not just formatting (e.g. that something was underlined) otherwise I could use plain HTML. But
it's not even just marking personal names, places references, dates, etc -- sometimes I want to connect those references to corresponding entities in
my data, e.g. connecting a reference to "my grandmother" (which is a person reference albeit not by personal name) to Annie Elizabeth Proctor in my
data. This is part of the reason why I avoided TEI (the other being that it's far too complicated), and why I came up with a private format.

Tony

On 23/12/2018 21:46, Peter Flynn wrote:

> On 23/12/2018 17:24, Nick Hall wrote:
>> Yes.  Exactly.  We are looking at unstructured text such as newspaper articles and notices, obituaries, wills, biographies...
>>
>> The idea would be to use a special text editor to give meaning to parts of a transcription.
>
> Like TEI does (although overkill here).
>
>> This could be stored as a tagged or marked-up note.  Then we could use a tool to automatically generate objects from this data.
>>
>> I have already written a prototype editor, but was too ambitious and attempted to add support for tables and lists.  I was not entirely happy with
>> the result.
>
> I'm not sure that it's necessary to reinvent this wheel...there are plenty of data formats and editors for them already. Perhaps a MarkDown-like
> format?
>
> P
>
>
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
1234