[bd-users] CD version of JSword

[DM Smith]

Coby,

Thanks so much for your wish list. Without more Java and Web programmers 
stepping up, this will take a long time. None the less, I will be adding 
this to our TODO list.

See my (semi-technical) comments below: (I too would like others to 
comment on and add to Coby's suggestions and even my response!).

Thanks,
    DM

coby ingram wrote:

>To all users and developers--
>
>I am dreaming on a version of JSword that can truly "run anywhere".
>  
>
This is a great idea!

>Currently I must use an office computer that does not permit installation of private programs, as well as a public access computer at the local library with even more restrictions--for example, it has no CD drive available.  My children just returned from a youth mission to Brazil where they could only use Internet at a cafe, or by borrowing the host organization's office computer by special permission.  Then my home computer has Internet but no printer.
>
>Teen missionaries, office workers, and Christians worldwide with limited computer access need Bible software that is as full-featured as possible, yet that runs on almost any computer with no impact on its secure resources.  This is, in my opinion, an urgent opportunity.
>  
>
This is a big challenge as our application uses Sun's Java. I have heard 
that IBM's works well also. However, GNU's does not. The reason that I 
mention this is that we would have to figure out how to distribute Java 
without violating their licensing agreements. The contexts that you have 
mentioned are not certain to have or allow the installation of Java.

When we re-write BibleDesktop to use SWT and RCP from Eclipse, we will 
be able to use GNU's Java. At that time, we will be able to distribute 
Java without licensing concerns.

>I am imagining a version of JSword that:
>
>--is based on using Explorer/Netscape as the operating window (or something even simpler--for example, see the HTML commentary on Amos from NZ--I forgot the URL).
>  
>
JSword is really well suited to having different user interfaces. As GUI 
writers need find lacks in JSword for their particular application we 
try to change JSword to accommodate.

In our next major release of BibleDesktop (post 1.0) we are hoping to 
embed a native browser as the heart of the application. I know that this 
is not exactly what your are driving at. But, I'll be working on this.

Joe has a demo, (which he posted on jsword-devel earlier today), at: 
http://eireneh.com/studious/

Yiguang has posted that he wants to create a Java Server Faces front end 
to JSword.

>--Reconfigures the window temporarily to add menus and buttons, as well as multiple viewing windows, automatically
>  
>
I do web development and find the confines of a browser to be too 
limiting for what I want to do. Part of the limitation is my own ability.

Managing multiple windows in a browser context is best when they are 
independent or tightly coordinated (e.g. dialog boxes to do a particular 
function.) When changes to one is expected to update the other, then the 
independency of browser windows becomes a problem. The user can simply 
close any window.

The other limitation is that browsers are more or less stateless. The 
state they maintain is via cookies. Standalone applications don't have 
this challenge. Fortunately, JSword does not require an application to 
maintain state.

>--Requires NO install, but can run from a CD or USB stick, just like the Autorun CD's you get bundled with a new computer
>  
>
I'll add Autorun as a TODO.

Interestingly, many autorun applications maintain some state on the 
local computer. Some actually install to and run from the computer.

>--Uses XML/OSIS to store Bible and book files in a format that can be read directly by any decent text editor or HTML interpreter
>  
>
We don't have control over the format of the "modules," (i.e. the Bibles 
and books). Many of these are compressed to conserve disk space. However 
they are not simply zip files. Most are not in OSIS or ThML (the other 
xml format). We could add an export functionality for modules. This 
would be to OSIS and possibly passed through XSL to create a preferred 
format.

Of course, locked modules would not be allowed to be exported.

>--Uses XML/OSIS to enable Search and reference linking functions (Strong's, commentaries, etc.) despite possible limitations of interface
>  
>
I'm not sure what you want here. This is the heart of how JSword 
actually works. At this time, the linking from a Bible to other verses, 
to Strongs and to Robinsons is robust. So is linking from commentaries 
and our readings to the user's current Bible. Linking within a 
dictionary or commentary needs help and the daily devotions are not all 
that daily yet.

>--Where possible circumvents the tendency of Java to run complicated functions slowly (e.g. Search)
>  
>
I'm not sure what you mean here. JSword is very fast when searching. The 
performance problems that BibleDesktop has are related to Sun's anemic 
handling and rendering of HTML. This is why we are moving away from it.

>--When installed, is able to use advanced word processing features, as in Open Office.
>  
>
How are you seeing this be used? Would you send a passage to OpenOffice 
as a new document? Would you embed word processing capabilities into 
BibleDesktop (or what ever the UI becomes) for writing personal commentary?

>--Is fully compatible with on-board word processing, so that verses can be pasted easily into documents, in fact, so the whole interchange is as seamless as possible
>  
>
While there is more work to do here, BibleDesktop provides a variety of 
formats for copy and paste. Different people have different purposes for 
the documents they create and they have different needs for the 
presentation of the text.

They paste easily into MS Word but have to be inserted with "Paste 
Special" in ThunderBird (my mail program) and in OpenOffice. This loses 
most of the formatting but the text is retained. I would like a way to 
make the normal paste recognized by these programs.

>--is simple to use without being bare-bones, or, as feature-rich as e-Sword, but as clean-lined as JSword or in some ways like the new Office that Microsoft is launching.
>
>--uses both buttons and expanding menus to access features and texts
>
>--has menus and other frontend text in as many languages as it has Bibles
>  
>
BibleDesktop and JSword are ready for translation. All it takes is the 
translation of the "properties" files.

>--is available as a freely copiable CD for a reasonable production cost
>  
>
BibleDesktop and JSword are truly free. Anyone can create a CD and copy 
it to their heart's content. They just need to include the source on the 
CD along with the GNU license.

The goal, of course, is to create a useful "live" CD.

>--Incorporates first and foremost as many foreign-language Bibles as possible
>
>--Includes the major public domain dictionaries and commentaries in English
>
>--Has, as well, any and all translations of key study texts, in at least one other language per CD (for example, a CD with Spanish Bible, menus, and reference/devotional texts)
>  
>
A CD could hold all the modules.

>--Can be run in a form that leaves as little data on the computer history as possible, to minimize danger for Christians in uncertain circumstances
>  
>
 From my tests, embedding IE into BibleDesktop would not leave any trace 
in IE's cache, cookies or history. Running it in IE would. While we can 
minimize its trace in the cache, cookies and history, if it is run in a 
browser, the user would be responsible for removing all traces from the 
computer.

The versions of the browser that run entirely from a USB flash drive get 
around this. Their traces are not on the local computer. I have not 
played with it as mine is way too slow and too small.

>--may be modified by the user to assemble a package of Bibles and other works that are appropriate to a target user audience
>
>--may even be run from a CDRW, to incorporate document/highlighting/commentary storage ability
>
>--pipe dream:  run a GNU word processor off a CD--mobile tract design!
>
>--Has the ability to include devotional and reference works in XML that may or may not currently be in the Sword module library (for example, books by the leader of a missions organization that hands the CD's out to its teen missionaries)
>  
>
One of our TODO's is to allow the direct use of OSIS Bibles. Currently 
OSIS does not have good markup for commentaries or dictionaries. Since 
OSIS is evolving, I see this is something that is coming fairly soon.

>That's a long laundry list, I know, and extremely disorganized, but the implementation is looking fairly simple to me.  I'm just not a developer, unfortunately.
>
>There are 2 ways to go on this.  One is using the JSword resources to develop a technical package that will run a suite of features anywhere.  The other is basing everything on HTML texts, and finding a way to adapt existing display platforms simply to what we want to do.  I tend to prefer the second way, but can't do it myself without more education.
>  
>
We have had the HTML discussion before. The ultimate problem with HTML 
is that it is not useful in marking content. Another major problem is 
that it is too easy to embed display commands which presume a particular 
platform. The third problem is that there is no standard use of HTML to 
markup text. So doing lookups for specific references becomes difficult.

JSword's architecture is to convert everything (that it knows hot to) to 
OSIS and then use XSLT to convert it from there to HTML or whatever in a 
form that is appropriate for the target platform. The net result is that 
you can end up with HTML.

So my preference is to keep this existing architecture and build upon it.

>JSword is the closest thing I've found yet to my dream of a FreeLight Bible Run CD, since it has the goal of being able to implement through applets (servlets?), and is an interface that is close to what I'm looking for.
>  
>
BibleDesktop/JSword consists of a blend of what we as developers want 
for our own Bible Study tool and what others have asked. The requests 
that get worked on the soonest are those that fit well with where 
BibleDesktop/JSword currently is or are clearly explained.

>Feedback would be appreciated, especially on how I myself could do most of the work to get this thing running.  I don't program, but I know what it entails, and I'm the one at the office always trying to tweak MSWord to jump through hoops for me.
>
>Anyone who would like to help move toward this goal, and I know the JSword developers have responded to my initial questions favorably, and have put most of them on the TODO list, could keep me posted if they have anything to report.
>
>Thanks,
>
>Coby
>