Nodejsconf
9 - 11 Nov 2012 at Nodejsconf

It's very interesting to see how someone has used the various techs in a production website. But there was too less Node in it ;)

Very interesting

I don't like a lot the sponsor-driven talks

I think it was too low-level for the conf ;)
Anyway: well done

I appreciated that Glen gave it to us straight: this is a product and this is how you can use it.
I value a talk that shows me how a "paid-for" technology can answer for me questions like "ok, I have this amazing app - where/how would I host it".

Plus he was never boring or "slow": the pace was great and the speaker entertaining. Even in the "cheese" "Buon Giorno!!!" over use - that's skill.

Plus, and that's important, he highlighted multiple time the free-tier you can get for this product, and that's of great value too - great way to get started before getting serious.

Useless talk not backed by any fact.
Plus, very biased on personal taste.

To comparison, the talk by Andrew Nesbitt at the same conference was a MUCH better way where you should use NodeJS, and where maybe is not such a great fit.
Facts, experience and consistence.

All things that instead lacked in this talk.

Many issues with this talk:
- No clue exactly of what the topic was: a mix of bad english + an overload of text and bullet points killed my attention
- ... that was later woken by the disgraceful slide that highlighted "backends" with a picture of Girl's bottoms (just the kind of thing Italy needs to get out of the sexist "bunga bunga" stereotype)
- The Microsoft guy, while spoke a far better english, pushed the marketing too much, for no real gain: the previous talk from the other Microsoft guy, Glenn, covered all there was to highlight the awesome work Microsoft is doing

kudos to Golo to prepare a presentation in such short period :)

it was interesting although not sure how much we need another testing framework.

apart from the nice abstraction of in/out arguments that sometimes lack in some test framework it looked more like a simple test fwk for functions only. (abstracting away only the "call" and the tooling code to check/assert output values). but then again it is such a young project that it's maybe too early to say.