[sword-devel] WBTI/BL modules in CrossWire Beta

[Greg Hellings]

Beta more or less is our "separate repository."  In reality our end
users shouldn't be in Beta.  In the past modules have been known to
get stuck in Beta for uncertain reasons.  I'd take a guess the reasons
involve some mixture of the following:
1) Very few people knowing exactly why a module was Beta - was it new?
Was a bug fixed? Were new features added?  Therefore there was little
direction for the testers to take.
2) Most of our list, who are largely either developers or module
creators to begin with, got in the habit of using the
Beta repository while waiting, so we were not very vocal about goading
people to release updated versions.  And since we are most of the
voices here and the voices of users tended not to know about the Beta,
there was little incentive to move forward.
3) Often modules come in large dumps, like you're mentioning, and
frankly I don't have the time or talents to learn 43 languages AND
read through the Bible 43 times to test.  And if we don't know anyone
who does know that language, what are we to do while waiting for a
"Hey, I read through that and it seemed well put together, let's
publish it."
4) For a long time only Chris Little had rights to move from one
repository to the other.  This might still be the case.  If so, only
one person to process 109 modules, plus be responsible for creating
them from source materials and probably fix the reported bugs is a
massive undertaking.  However, due to the exceedingly high standards
for modules in CrossWire's repositories, very few other people would
have Chris' attention to detail.

If I go much farther I probably will end up getting into the numerous
objections where some people prefer the "release early, release often"
method and others prefer the "wait until it's perfect" method.  CW
repos tend to wait longer until release while Karl tends to do the
quick iteration approach.  As a result, using "Beta" became mainstream
as a user would report "My module is broken in such-and-such a
passage" and they would be told to try the version in Beta and they
would say "All better, thanks."  Thus a bad habit of pointing users to
testing material led to "Beta" becoming used in production - observe
the KJV 2.4 in Beta which nearly everyone ueses while 2.3 continues on
in the "released" stream complete with all of its quirks and invalid
OSIS from the old importer.

--Greg
On Thu, Dec 9, 2010 at 3:16 PM, David Haslam <dfhmch at googlemail.com> wrote:
>
> The 43  http://crosswire.org/wiki/WBTI_Bible_discussion modules  we were
> given a long time ago do rather swamp the module manager when
> http://crosswire.org/wiki/Modules_in_the_beta_repository CrossWire Beta  is
> selected.
>
> This makes it hard to see the wood for the trees. Ploughing past all those
> unusual language names is quite a chore.
>
> 43  / 109 is almost 40% of the beta repo.
>
> I suspect that few users would wish to install any of these, unless they
> happen to be involved in ministry in Central America or the Caribbean
> region. Thus they have a highly specialized focus.
>
> Would it not be a good idea to have these in a separate repository?
>
> What do others think?
>
> David
> --
> View this message in context: http://sword-dev.350566.n4.nabble.com/WBTI-BL-modules-in-CrossWire-Beta-tp3081010p3081010.html
> Sent from the SWORD Dev mailing list archive at Nabble.com.
>
> _______________________________________________
> 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
>