Posterior samples

Database design is important, especially if someone else has to work with your database. It’s not really something we teach in undergraduate science, perhaps the American model of requiring a certain number of credits from certain fields would help remedy this.

Ever wanted a glimpse of Bayes’ notebook?

Laura McInerney’s comments on the benefits of small conferences are similar to my experience with 8 BNP in Veracruz, Mexico. As long as you’re within the niche field this sort of conference is a great experience. I felt a little like an outsider at 8 BNP because while I was interested in non-parametrics and was working on smoothing, a lot of people were working on things that I had no experience with which are actually the central elements of the field. I got to learn about a lot of neat things, hear some great talks and meet lots of amazing people, but I don’t think I was steeped in NP Bayes enough to really get the most out of the conference. My research went a bit away from NP Bayes these last few years so I didn’t get to put anything together for 9 BNP in Amsterdam. Perhaps ISBA 2014 in Cancún, Mexico will provide a bit more of a chance to get back to that work.

We’re teaching R in SEB113. Perhaps any students reading this might be interested in these 60 R resources.

I use multiple monitors at work but really enjoyed the virtual monitors setup in Gnome when I ran Ubuntu. It turns out that having a large canvas of pixels, rather than multiple monitors, is the key to workplace productivity. My work setup has two widescreen monitors side by side in portrait orientation. This doesn’t work particularly well with programs that assume you’re using a single landscape monitor (such as RStudio) or give you a single window with multiple documents inside that each have focus one at a time (Microsoft Office, why can’t I have a spreadsheet on each monitor?) but it means I don’t have to keep switching back and forth between TeXStudio and RStudio when I’m writing up my analysis.

Flipped classes are an interesting model for education. I remember taking an Honours level mathematical modelling course a few years ago where the three hours of lecture time allocated us were used to discuss concepts and do modelling. We would read a chapter from the textbook in the lead-up to the class and then have a talk about what it meant and then work out a model based on a case study. I don’t know how well a truly flipped class would translate to a group bigger than about 30 students, but Sue Savage (QUT) tells me that the new lecture theatres in P block are designed to facilitate small group discussions within lectures.

Daina Taimiņa explains hyperbolic geometry with crochet. Every once in a while something similar pops up and I can’t help but get excited.

Daniel Hills recalls his memories of working with Richard Feynman on developing a massive parallel computer in the 1980s.


3 thoughts on “Posterior samples

  1. Matt

    I think that the aversion to programming / database design amongst students probably rivals the aversion to maths/stats. Do you know if there’s any equivalent to SEB113 teaching that kind of stuff? Not sure if the existing INB210 & INB340 would be suitable for non-majors…


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s