[sword-devel] Status of OSIS?

Mon May 4 21:24:11 MST 2020

1. What I mean by 'encoding' a copy (for my own use originally, but after
trying to find the list of corrections... i'm on this thread thinking the
work needs to be shared.)

Problem:
The PDF has issues when you try to copy tabular information.. it was not
generated to be reused digitally at all.. random text sequencing is
present. (at least for me using both Adobe PDF REader and Ubuntu default
PDF reader on Ubuntu 18.04, but I remember this from my windows days 12
years ago.

My Personal Use:
I need those tables to become data tables for programming.  I 'code' in
spreadsheets as much as possible. So, this morning I started trying to
squeeze the lemon. And just like everyone, I run into text integrity
issues, and thought this isn't working well enough, I need to find the
source text. Query 1 was to Patrick Durasau, but the email in the spec is
non-functional.  I traced him til 2013 when he was let go as the document
manager of  OASIS (of OpenDocument/OpenOffice.) But while he crafted the
text, he wasn't the owner. American Bible Society holds the copyright, so
the logical first step is to query the guy managing USX and USFM about it.
But even before that, the only live document on the web seems to be from
crosswire (and ebible too... ) so that led me to this thread asking
a. Does crosswire have ownership of the OSIS spec now? (can we initiate
changes, maintain it?)
b. does a project to at least bugfix the OSIS spec already exist?

I think I've seen enough to understand that niether is true, so the logical
next step (unless someone who was involved directly in 2006 can correct me)
is to ask Jeff Klassen at UBS if there is source available from UBS, and
what would we need to be able to become the maintainers or primary
contributors to a project if they want to continue owning it.  However,
sending that email when I've tried to draft it, I come across speaking for
crosswire, which again prompted me to create this thread.

On Mon, May 4, 2020 at 2:49 PM DM Smith <dmsmith at crosswire.org> wrote:

> Can you clarify what you mean by “I’m encoding the entire spec for storage
> in a source form….”?
>
> What has happened so far is that we’ve been responding with the 2006 2.1.1
> version of the OSIS Spec. We host the file at
> https://crosswire.org/osis/OSIS%202.1.1%20User%20Manual%2006March2006.pdf.
> If I remember correctly, the PDF is generated from a different document to
> which we don’t have access. If we wish to fix and improve that document, I
> think we should see if we can formally take responsibility for it by
> contacting Patrick Duruasau.
>
> Short of that we’ve been documenting shortcomings in the wiki.
>
> We do require valid OSIS. There have been bugs and shortcomings in the
> schema. I’m the pumpkin keeper of that and have made a few changes that are
> agreeable to this mailing list. We’ve attempted to document that in the
> wiki. One of the considerations is whether the suggested change works with
> how the SWORD and JSword engines understand the spec.
>
> You are right that the wiki is not issue tracker and gets harder to
> understand as more gets added to it. David and I have tried to have the
> wiki on OSIS be an addendum to the spec. And guidance on how to build a
> SWORD module using it.
>
> Do you have a suggestion on to get from where we are to where you think we
> should be?
>
> In Him,
> DM
>
> On May 4, 2020, at 3:12 PM, Michael H <cmahte at gmail.com> wrote:
>
> David,
>
> That page you refer to is the problem that created this email.
>
> The page you refer to shouldn't exist like it does.. that information
> should be going onto a problem ticket system. (Think the 'issue tracker' on
> the USFM 3 list.  It's searchable and the status of most items is clear,
> and anything already acted on is already on the official documentation
> pages.)
>
> That collection of pages in its current form provides little information.
> it's not sorted by status of investigation/implementation, nor by the
> spec's organization, but rather by the impression of the author as to it's
> nature.  It seems to be a collection of writings, some of which describe
> real problems that have been acted on already, some that describe
> misspellings (but can safely be ignored for module creation.), and some
> that fit into "wishlist" meaning even if they're in the 'bug' category and
> actually bugs, they aren't affecting what happens today. Each writing in
> the wiki will have to be processed before I can code. I can't see any clear
> status marker present so I can sort the already dones from the wishlist.
>
> THATS what I'm suggesting/working toward.  I'm encoding the entire spec
> for storage in a source form, so that implemented bugfixes can be updated
> into the spec. We should not have to go through megabytes of text to find 3
> misleading characters in the spec that will break every module someone
> trying to follow the spec will run into. I've seen enough in the wiki that
> I'm pretty sure there's at least on issue listed there that is likely to be
> in that class, but I'm going to have to sort through each and every
> sentence on each and every page to find them all.
>
>
>
> On Mon, May 4, 2020 at 1:25 PM David Haslam <dfhdfh at protonmail.com> wrote:
>
>> Have you looked at our Wiki page?
>>
>> OSIS 211 CR
>>
>> It was even edited again today!
>>
>> The Bible Technologies Group has not met for years & the original website
>> went AWOL.
>>
>> It may well be the case that CrossWire is the only remaining de facto
>> maintainer of OSIS.
>>
>> Best regards,
>>
>> David
>>
>> Sent from ProtonMail Mobile
>>
>>
>> On Mon, May 4, 2020 at 19:07, Michael H <cmahte at gmail.com> wrote:
>>
>> I've got 40 works and growing that I've been meaning to look at creating
>> Sword Modules. All of these are genbooks. Almost 100% currently are works
>> by Andrew Murray (but the list is much bigger.)
>>
>> But, as I try to make sense of the OSIS spec, I'm facing a 2006 spec in
>> not very well done PDF, and another one with comments, and an xslt file,
>> and a mountain of comments on the wiki that span from outright errors, to
>> support gaps, to wishlist.
>>
>> What is the status of OSIS? Is there a draft or official source, or even
>> Crosswire source that we can at least fix typos to? I've started one, just
>> to turn Appendix F into a real table... but as I read through the wiki, now
>> it seems I'm going to have to process everything to be able to trust what
>> I'm reading, and it makes sense that I should be dropping the result
>> somewhere more official than my google drive.
>>
>> If we have permission to host the OSIS spec, do we have permission to
>> bugfix it (at least the spelling gaps, and fixing the tables of information
>> to be tabular?)
>>
>>
>>
>> _______________________________________________
>> sword-devel mailing list: sword-devel at 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 at 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 at crosswire.org
> http://www.crosswire.org/mailman/listinfo/sword-devel
> Instructions to unsubscribe/change your settings at above page
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.crosswire.org/pipermail/sword-devel/attachments/20200504/1d8a52f9/attachment-0001.html>