[sword-devel] OSIS Users Manual considered confusing

Karl Kleinpaste karl at kleinpaste.org
Sat Mar 3 18:37:46 MST 2012

DM Smith <dmsmith at crosswire.org> writes:
> I'm not sure you're wanting a reply or just wanting to rip on OSIS as
> a "standard."

Not at all.  Consider: I have a long history in IP communications, whose
RFCs are The Definition Of Interoperability.  The idea of ambiguity in
such a field is virtually anathema.  In the long-touted OSI
communications stack -- back in the late '80s when there was genuine and
serious debate as to whether IP or OSI would win the global battle[*];
cf. _Elements of Networking Style_ -- the existence of mutually-
incompatible subsets of available features is a substantial part what
made OSI lose against IP, because IP successfully presents what is much
more of a strict superset/subset between "MUST" and "MAY" (as those
appear in RFCs).

So what disturbs me is the idea that The Standard, in whatever field and
from wherever it emanates, could have left operational ambiguity in its
wake, which is then abandoned to individual implementations to make
concrete.  This would be a Problem for me.

With that said...

> I think the ambiguity is in SWORD, not in the standard. I guess I
> should have been clearer.

Well, that's fine.  Er, not in current implementation; rather, meaning
insofar as The Standard had something to say about it.  What I think you
express is that the implementation has not fully implemented The
Standard, and addressing that over time toward eventual resolution to
The Standard is dandy.

I am admittedly a little frightened by the fact that the implementation
is still not up to par after N years.  We have, after all, discussed a
number of different failures of how The Standard doesn't cope well in
some area or how Sword's implementation continues in a buggy state; I
recall particularly a brief discussion about spaces in genbook keys from
about a year ago.  I won't sugarcoat that kind of thing; I think that's
a Problem all its own.  And it's why I don't personally put OSIS to use.


[*] After I wrote that paragraph, it occurred to me: I wonder how many
people today are not even _aware_ that there was a battle over some
years, over what kind of worldwide internetworking should be put to use?

More information about the sword-devel mailing list