i write the best practices column for software test and performance magazine (http://stpmag.com). currently i'm looking at the issue of post-deployment tuning. i'm wondering if anyone here would be willing to share a few anecdotes about tuning after deployment? (and we can define deployment broadly here; say, anything from beta release on.) yes, yes, i know the general rule of thumb is to avoid tuning downstream, but it's an unavoidable fact of life given system complexity. so, what works best if you have tune after deployment?
i'm particularly interested in post deployment tuning in the new age of application-in-a-browser computing. everything seems increasingly componentized and fast (in terms of release cycles). so i often think maybe the best route is just to get code out there as soon as possible and let users given feedback or even tweak if via open source (rather than waiting years to release something that's supposedly airtight). anyway, i hope to hear from folks here. my deadline is saturday.