IIRC part of the motivation for using a "gettinstarted" collection in the
tutorial was so that after people ran the tutorial they would have a
collection (with configs) they might actaully want to use for their own
experimentation -- w/o any of the "weird" config specifics of the
techproducts collection -- which had configs and data tailored to show off
specific features of Solr, that may not make sense in an "out of the box"
_default configset.
I think the really nice thing about examples in the ref guide that use
the gettingstarted collection is that it's pretty clear they should be
applicabe to *any* collection, while (in theory) the examples that refer
to techproducts tend to assume specific documents are indexed so they can
show of exactly how that feature works with those documents (ex: the JSON
Facet api docs)
Personally: I think the techproducts example data is sooooo stale it
stoped being funy, and then circled back to being funny again.
I would love if we had a more "timeless" sample dataset, with it's own
tailored config, that could be used for all of the "complex" examples in
the ref guide (including nested docs, complex faceting, custom synonyms,
etc...) -- but i also really like the idea that "simple" standalone API
examples can/should refer to a "gettingstarted" or "my_collection" when
there is literally no expectations or preconditions about what collection
you run it against other then the _default configset (eg: create
collection, add/remove replica, add/remove field, etc...)
-Hoss
http://www.lucidworks.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
