Given your personal take on everything relating to your talk, I decided to do this from a throw-away, just to simplify matters on my end and avoid complications

Given your personal take on everything relating to your talk, I decided to do this from a throw-away, just to simplify matters on my end and avoid complications on m-o. Your talk was not bad; it did, as claimed from the get-go, cover a fair bit of ground at a moderate level in very little time. You pointed out the relevant documents to dig deeper, and several people I talked to found you struck the perfect balance there. Now, I disagree with them there: I was after a more technical talk - a token implementation of semantic hashing and discussion of issues run into while making/scaling it would have made this talk amazing for me. But that's a personal preference thing, I was aware of the techniques mentioned before entering so I was not really your target audience. Beyond that, everyone I talked to was unimpressed by your begging for coverage as per the "tweet this" slides and matching vocal appeals for such. This isn't a blog article where you make the appeal clear and loud to get more followers; this was a presentation at an industry conference, and that behavior was particularly out of place and ill-received. In addition, you are the only conference speaker I saw appealing to attendees to rate their talk (at least twice on the #strataconf tag) so i would take the ratings with a grain of salt. I know the above seems very negative, so I should repeat: it was a fairly good talk, and would have been great (even though less technical than I was hoping from the 'practitioner' track) except for the primadonna attitude surrounding it. I have enjoyed many of your postings on HN, I've found substantial value in metaoptimize/qa, and I really wanted to love your talk - so perhaps i'm being unduly harsh. I sincerely hope you will tone down the self-promoting aspect of your talks in future; if you do I will be happy to catch your talk at Strata 2012.