Secondary indexing with Redis
redis.ioRedis is great but I think this is a bit too complex for that specific use case.
Forgive this shameless plug but have been working on a distributed transactional key value store that has native secondary indexes. Feedback welcomed: www.quasardb.net
For what it's worth, I think your website headlines need a lot of work. In the first few seconds of visiting your site I should know exactly what quasardb is and why I should care (i.e., whether or not the solution/service/product/whatever relates to me and my problems). Your current headlines tell me nothing.
I have some suggestions for a better marketing message, so shoot me an email (address in profile) if you want to discuss further.
Thanks a lot we know we have a lot of work ahead of us marketing wise.
If you start doing this, think long and hard about your architecture choices.
This is such a banal and not-constructive comment. Of course he would think about architectural choices. Your comment suggests one of the following:
1. The proposed feature is harder than other proposed features (which ones and why? edit: it turns out there isn't even a proposed feature here...)
2. You see pitfalls or good choices and bad choices (which ones and why?)
Either way, your comment as it stands isn't helpful in the slightest, and I'm calling shenanigans. If you are in either camp 1 or camp 2, why not actually say something useful instead of speaking from this glass pedestal of falsified authority.
edit: formatting.
you are aware, by the way, that the antirez' blog post does not discuss (as far as i can tell) implementation of a new feature, but instead just explains a few community best practices that have coalesced over time regarding implementation of multiple indices using one of the redis data structures (zsets)?
but even if he were, what risk would this simple feature pose to the redis architecture such that "long and hard" thought is required to avoid some apparently grim consequence?