Lessons learned and legacy of Kalusa

Hell-oh! :)

I have been revisiting the Kalusa website and the Kalusa Yahoogroup
recently. General interest in Kalusa seems to have faded for most
early contributors and lurkers. However, Kalusa still has a lot to
contribute to the field of conlanging. In this context, I would like
to discuss about the legacy of Kalusa to the conlanging endeavour.
I believe that the main legacy is the set of sentences that were
submitted to the Kalusa corpus (http://kalusa.fiziwig.com/corpus.php).
As Kalusa has been a collective project, it is likely that the
corresponding sentences in English would be remembered by someone
(specially another conlanger) who intends to test the "completeness"
of a language. This is true even for the "weird" (uncommon) sentences.
So a conlanger would have to pay attention to them.
I would like to know about other aspects of legacy and lessons learned
in that project. What have you learnt about collaborative conlanging
on the Web? Does it work? Which technological support (tools) is
needed?
How requirements affect the success of a collaborative project?
Which elements (syntactical, grammatical, etc.) have you reused from
Kalusa into your personal language projects?
Which lessons could you extract from that project? They can be about
technology, language development/evolution process, language quality
assessment, language testing, language requirements engineering,
groupwork, politics, ...

Antonielly Garcia Rodrigues