I was wandering around testing various options for my new personal project, ranging from PHP, to node.js, to Haskell. I feel most comfortable with Python, though, so I thought I'd go back to it in the end.
I've taken a second look at frameworks like Django, but given I mostly enjoy SQL and I don't want an ORM, I thought I'd use Pylons (as ORM is optional in Pylons). Now, the problem is there's Pyramid, and it would seem Pylons would be replaced by it soon.
Long story short, is Pyramid ready for prime time? Also, given that Pylons itself was a very successful platform (judging from its high-profile deployments), do you recommend Pyramid as a good replacement? (Obviously, I'm asking people who actually use Pylons and/or Pyramid.)
To clarify, the application will probaby use up to 4 database tables, and will be heavy on AJAX calls. JSONRPC with all-JS frontend is also an option. One of the tables is expected to hold tens of thousands of records consumed by different users at all times. Queries on the large table are not limited to select, as they will most likely come from some form of AJAX data grid.
BFG, which is the precursor has been used in "primetime" for a while now. There are in Pyramid some additions to BFG to make it more "Pylonic", and these may not be ready, but I would say that they will be soon.
So if you are worried, you can use BFG for now. The migration to Pyramid should be simple once you want to migrate: migration tutorial.
Update: Pyramid 1.0 has now been released. It is now definitely ready for prime-time.