Feels Like Just Yesterday That I Was in NYC

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

Oh wait, it was. Anyway, my less than 24 hours at home were totally worth it. While we were delayed by about 45 minutes taking off [1] yesterday, I got quite a bit done on the plane. And speaking of – if you’re waiting for an email reply from me, you might want to resend b/c AFAIK I’m caught up on my “ToReply” list. In any event, the combination of excellent Denver Mexican food, Sox game on TV, and some quality time with my resident troublemaker beat the hell out of yet another night in a hotel. The only bad news was the word on my car, which does indeed need yet another transmission. Fortunately, it’s covered by warranty but they don’t get it until Friday. I’m beginning to think we’ll have completely run out of oil by the time I get the poor thing back from the dealer.

Anyway, trip out here was good, and the roundtable/dinner I just came from was thought provoking on a number of levels. Not sure how much I can talk about that, but I have a question out of the dinner. When I stated that audio is mostly an opaque format – with the exception of workarounds like tagging – another analyst contradicted that, contending that there do exist credible, viable audio speech to text software packages. This surprised me, because while I know of research that’s been done in the area, I’m not aware of anything I’d call viable and more critically production ready. If there are, why are they not in wide distribution? I can’t believe that something like IT Conversations would not be significantly transformed by such a technology.

Case in point of this was my trip out here, when I finally caught up with some of the items in my “To Listen” folder, with this IT Conversations interview of Active Grid’s Peter Yared among them. I was a bit surprised to hear my name taken in vain, though the interviewer mistranslated RedMonk into Webmonk. Although it was in reference to this post of mine, this was the first I’d heard of it because there’s no Technorati for audio that I’m aware of. Nor even close, from what I can tell. But what do you guys think? Is there solid audio crawling technology that I’m missing? Because if so I’ve got a ton of backed up podcasts to turn it loose on.

Speaking of Technorati, I’m guessing that everyone by now has seen Google’s verion of blog search – I’m curious as to your reactions. Just plugging some vanity queries and quick link checks, the results have not been great for me. While the mediocre quality of a beta release of new technology should probably be expected, Google’s set the bar a bit higher for me and I have to say that at least on my first pass I was disappointed.

Back on the podcasts thought, I also had the opportunity to catch Adam Bosworth’s talk from the MySQL conference, and while I’ve linked to the deck before I highly recommend the audio. Besides the insight he provides into the simple, sloppy, standards and scalable model that’s driven the web, he comes to some very interesting conclusions about the opportunities in front of MySQL. When I chatted with Marten Mickos on Monday, we shared a laugh about the analogy some try to use in dismissing MySQL as a real competitor, calling Oracle a 747 to MySQL’s Toyota. To me, that’s a selling point for MySQL, rather than an attack, because there are a hell of a lot more Toyotas out there than 747’s. Adam seemed to agree, because he expressed skepticism at the desire of some of the MySQL conference attendees for “enterprise” features such as stored procedures, triggers and views. His reasoning was that this pushed processing logic onto one computer, which is the opposite of the web approach. In short, the message was “don’t become Oracle because Oracle can’t do what we need it to do.” Very interesting listen, I promise.

It wasn’t the points made vis a vis MySQL, Oracle and simplicity that resonated with me most, however: it was his aside on conference start times. Adam noted, as I have *many* times before, that it seems slightly ridiculous that developer oriented conferences begin at an hour when many developers are not awake yet, let alone functional. Maybe it’s just because I’ve got to get up for the Impact conference tomorrow at some ungodly hour given that it’s downtown at NYU, but I can’t understand why conferences like OSCON and such continue to kick things off at 8 or 8:30. Is it a logistics question? Just like Sun made a lot of friends by keeping their Monday session surprisingly short, I think other conference planners would do well to start things at 9 or 9:30. And for those that object on the grounds that there’s just “too much to cover,” that to me is a sign that you’re trying to do too much. 9:30’s far more reasonable. We’re technologists, not dairy farmers.

At any rate, I have little idea what to expect from tomorrow given that I only found out I’d be attending a day or two ago, but I’ll be sure to bring you anything I think would be of interest. I’ll be back in the office Fri AM, and meantime hit my cell if you need me: 617.320.9757.

Update: More on the Google Blog Search – how am I the #1 “steve o’grady” on regular Google, but not even in the 1st page of results on Google Blog Search? Weird. It’s really, really fast though – I’ll give them that.

[1] When your place in the takeoff line is double digits, you know it’s going to be a while.


  1. Stop calling yourself sogrady on your own blog.

    Anyway, NOW you're the first result. This post actually.

  2. It makes sense that a blog search would more heavily bias timeliness than the regular search does — I'm not sure how clear that is from the puff and marketcrap I've seen on BlogSearch, but if I want Authority I'll go to the main websearch and if I want What's Being Said Right Now I'll go to the blogsearch.

    So rankings are going to fluctuate wildly on the blog search, I bet.

  3. Danno: given the way that Google appears to be returning not results from my blog, but only the mention of my name inline, i guess i will. seems weird tho. i would have expected that the ranking algorithm would have some mechanism between determining a blog in the name/title/etc and mentions in an entry.

    Fraxas: agreed, but timeliness didn't appear to be the issue as some of the postings that did show up were not at all recent. i think their search seems to be fast, but the ranking is completely chaotic from what i can tell. i apparently need to be mentioned in the post itself to be credited.

Leave a Reply

Your email address will not be published. Required fields are marked *