[sword-devel] [bd-users] Map Idea
jonmmorgan at gmail.com
Tue Mar 25 23:16:23 MST 2008
On Wed, Mar 26, 2008 at 4:58 PM, coby ingram <coby at peru.com> wrote:
> I'm just replying to this because I've been researching a similar idea for a
> while and have some ideas from my own hard knocks.
> I will insert a few responses to what has been written here. I invite you
> to check out what I say for yourself.
> ------- Forwarded Message --------
> From: peter <refdoc at gmx.net>
> To: SWORD Developers' Collaboration Forum <sword-devel at crosswire.org>
> Cc: bibledesktop-users at crosswire.org
> Subject: Re: [bd-users] [sword-devel] Map Idea
> Date: Sun, 23 Mar 2008 11:04:55 +0000
> Looking at the KML files distributed by Openbible.info (under a free
> licence) - these contain all info one ever would want.
> I assume you meant to type XML... And info is great, it's accessing it
> smoothly and dependably that's the rub. Read on.
KML was meant. KML is the format used by Google Maps and other
mapping software (see http://code.google.com/apis/kml/documentation/).
> The coordinates work in Googlemaps flawlessly as soon as one appends
> One issue that needs to be addressed is whether you want to be dependent on
> Google, or whether you would like to use other maps as well, e.g. from a
> Christian site, or on your local drive.
> 12:8</a>, <a
> search in googlemaps for E35.261226, N31.916978 and you end up in Ai.
> So a genbook or dictionary or whatever made out of this data should be a
> fairly straight forward matter as such - but for the fact that currently
> none of the interfaces will currently load images etc from the internet.
> SwordWeb won't? It's HTML-based, after all. IMHO HTML is one of the best
> ways to go, on a lot of levels.
> Jonathan Morgan wrote:
> > Note that I am CC'ing sword-devel on this. I suspect that that would
> > be a better place to discuss this.
> I'm not subscribed but I assume that's where the discussion is going to take
> > On Sun, Mar 23, 2008 at 10:38 AM, Brent Coffey <bcoffe at hotmail.com> wrote:
> >> Well I checked out the web bible mapping web site and the ccel web site.
> >> The bible mapping site is similar to what I'm thinking but far far
> >> than what I was imagining. But could be a valid way of starting...just
> >> known locations from verses that mention a location. I think I know how
> >> do the GUI, conceptually at least. A really slick tool might be to do
> >> a flash/flex front-end instead of using swing or any of the other
> >> java toolkits. But that would get away from the free tools. Then again
> >> using SWT or even Apples JVM can make so pretty nice GUI's, IMO.
> HTML makes your GUI flexible, customizable, and understood by every
> platform. Just for starters. And Web apps more than others speak other
> (software) languages. Most every app has a window on the web. Especially
> Java. The nonfree tools you mention, especially Flash, not only exclude many
> users, but are unneccesarily complex and do not play well with others --
> like HTML does.
HTML may also have major browser incompatibilities, obscure issues,
and strange results. I do not say that it shouldn't be used, just
that it isn't best for everything.
> >> I figured it may not be so bad to pull in map data from the java
> >> project (NASA project, like google). But yes it is defiantly ambitious
> and I
> >> have no idea how much I want or am willing to invest in the effort. Just
> >> kicking some more "unique" ideas around. Would be nice if BibleDesktop
> >> could out shine logos, then maybe Logos would open up their Libronix
> >> which I believe is a closed standard? Would be great if any piece of
> >> software commercial or open source could read Libronix files. The price
> >> logos is so high and with no major commercial competition it seems like
> >> is not as innovative as it could be.
> >> But it does seem to me that the real challenge and the most useful thing
> >> the idea of "tagging" meta data to books, chapters, verses, words, or
> >> phrases. This technology would be foundational to doing any of the
> >> ideas as well as a host of other things it seems. Where would I start to
> >> even get requirements for such a huge undertaking? I checked out CCEL but
> >> didn't see anything that jumped out for annotations? Perhaps I missed
> >> something?
> I sent a link that lists most known annotation-related projects. A thorough
> knowledge of the subject is important before going much further.
> On CCEL you have to open to a page in a book (Try the KJV) that contains
> text. A toolbar pops up that allows you to put notes on their server if you
> sign in. It has many nice features. Crosswalk has a similar service.
> Obviously these guys would be a great resource.
> > If you are interested in getting location data, have a look at
> > http://www.openbible.info/.
> > A few things that are worth considering:
> > 1. Generic and Useful are not necessarily connected. The mere fact
> > that I can attach arbitrary information to verses doesn't mean that
> > that information will be easy to use afterwards. A format specific to
> > something like maps may be easier to implement and use.
> It's called RDF and it is a universal linking format for metadata. All
> linking, all formats, all data. The trouble is putting together a library
> with enough heavy lifting power to handle it. Most web annotation services
> put their notes on a server like Apache using a MySQL database and an
> interpreted control language like Perl. I have been looking for a way to do
> this on a smaller scale but no one else seems to be. BD might have the seeds
> of a way to work on both a large and a small scale since it, like most
> annotation, many database, and some interpreter and server apps, is in Java.
It still hasn't solved your problem. We can create map data in RDF
with little dificulty. We can link it to verses with little
difficulty. We can probably use that data with a tool like Google
Maps fairly easily, or draw maps using other methods with somewhat
more difficulty. But we still need to add support for processing map
annotation data. Just adding a generalised annotation framework
hasn't solved our specific case. Developing a good UI to annotations
means more than showing the user that we have lots of KML data (or
other data) associated with this verse. It means interpreting the
data in some way, and the way of interpreting it is likely to be
somewhat different for each kind of data.
> > 2. A distinction would need to be made between making an annotation to
> > a particular version of the Bible and making an annotation to a
> > particular verse in any Bible (for example, I don't really want to tie
> > map locations to a particular Bible, but I do want to tie them to a
> > particular verse or verse range - which may not be present in all
> > Bibles).
> If the annotations are tied to the OSIS (XML) tags first, and only second to
> the (localized) text, you might be able to annotate many Bibles with one
> note. Most annotation seems to be XML-based.
> Please let's be clear that you do not want to touch your original text. The
> annotations stay in a separate file. RDF/XML links the two files. Others
> running the same study platform can import and use your notes as easily as
> opening an email.
You certainly don't want to touch the original text. However, then
you are still left with problems of distribution and relevancy (if I
link to a GenBook key, it is likely to only apply to that GenBook,
whereas if I link to a verse then it may be intended to relate to many
versions / commentaries, or it may be only intended to link to that
one. The UI would need in some form to manage these user intentions).
> > 3. How are you planning for annotations to be distributed? For
> > example, your Bible maps annotations should not be version specific,
> > and as such shouldn't be distributed with a version, so they must be
> > installed in some other way (either automatically or explicitly).
> > 4. Are you also wishing to annotate dictionaries, commentaries,
> > GenBooks and so on, or just Bibles?
> A Bible in OSIS format is one of the more complex XML documents in your
> list. The others should be trivial to add.
As stated before, while it is trivial in theory to create links
between resources and annotations and so on, using those resources
(including presenting them in a good UI) and developing a UI that will
handle user intentions correctly is not so trivial.
More information about the sword-devel