I’ve already mentioned in many conversations, that this particular conference was probably the least “social” of all the conferences I’ve ever attended, and I am not sure why. I used to be in different positions at different conferences: either a novice, and then everybody, all the big guys want to encourage you, they initiate conversations even if these conversations are to the effect: come on, little girl, what’s your name and how old are you? Or an “experienced” person, and most of the time me being “from industry” actually added value – the computer scientists are often insecure, and not sure whether their theoretical work is of any interest to anybody :). In the latter case I always try to engage the students into the conversations, I always tell people, when I liked their presentations< and what exactly I liked about it…
So, at this conference most of the people didn't really try to meet somebody new, I saw people congregated on the same groups for the whole conference, until the very last day, and often they were with their co-workers or people from their own university all the time.
Because of that I decided to do a thing, which I usually do not do at the conferences – to attend the Demo (poster) sessions. And let me tell you, that was the best decision I made at this conference! Because people with demoes actually have to get engage the visitors into the conversation. So here is a brief summary of my demo session experience.
The demo with Personal Information management was kind of OK, it looked nice, but more like a student work that a conference demo (which is not to say, it was bad).
Then there was quite interesting poster about space management in Postgres. The technique proposed in this paper, allows in-place update , which sounds pretty cool. Ive talked to the authors of this paper, and I actually encouraged
strongly recommended them to submit a talk to PG Open. So now I am waiting for the call for submissions 🙂
Although I do not work with a web search in general, I could share some of my experiences of building the customer search in Aperture. For example, many researches believe, that the end users actually want to write queries in a “natural language”, and/or to write them “in a declarative way”. Which is not always true, as people are 1)lazy 2) are used to google search 3)or are super -efficient with the old search techniques, and do not want to switch to the new ways.
Overall all the personal interactions where very interesting, and way better that some sessions, where there were virtually no questions asked.