What "schemaless" seems to mean to me (at least in the context of nominally
structured data, where similar records still share some kind of structure)
is that you deal with "schema versioning" at the application layer rather
than the database layer; if you started using a property after deployment,
you need to make sure in your model / etc that the data has the structure
and content you expect. In other cases, where the data really is freeform
(just a hash, say), "schemalessness" could be a great boon; otherwise I'd
say it probably just shifts your headaches upstream ...

I tried mongo, redis and a couple of the mappers for both. I spent Australia
day ripping mongomapper (and about 5 fairly simple, related models) out of
my app and replacing it w/ ActiveRecord, and feel much better for it. My
take is that most of the k/v / document stores probably aren't powerful
enough to support a decent relational object mapper; I reckon mongo is, but
the mappers aren't mature enough for me to want to use on anything intended
for deployment. Too much time wasted on quirks of the O(R)M I could have
been writing code.

I've said this before: I like maturity in my persistence layer(s).

Disclaimer: these are loosely held opinions based on a limited set of
experiences.

-- 
You received this message because you are subscribed to the Google Groups "Ruby 
or Rails Oceania" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/rails-oceania?hl=en.

Reply via email to