Hi peeps. I want to try to debug a bug report on . All my computers run Debian Linux (mostly testing). What's the laziest way to get a test environment for OpenBSD up and running? I have a working libvirt/kvm install, and if I must I can fire up virtualbox.

Cool trick of the day. Use org-babel for the code samples on Monday's quiz. Not only can I edit them with indentation and highlighting, but I can combine them all ("tangle" them) into a single file to distribute to the students.

OK, so my threshhold for cool is low at this point of the semester, but still, it's a nice use of so-called "literate programming".

covid, introversion 

I secretly like keeping 2m away from everyone.

90 minutes bottom-crawling to find this amazing subject! Worth every minute. Sea-horses (this is a long-snouted seahorse, Hippocampus guttulatus) are very hard to photograph because they're so shy: to get these photos, besides shooting with a long lens, I had to hold my breath and inch forward little by little, snap, then backfin slowly away. Otherwise the seahorse will turn away from the camera and blanch in colour.

@christianp I bet at least part of the problem is the bare k n the range. If I write 1..{k}, then it stops accepting expressions, which fixes one of the issues.

btw, if your not christian, and wondering what the heck, this is

and numbas is pretty cool.

Show thread

@christianp I used the following JME formula as the minimum and maximum allowed value in a number entry field.
I had two students get answers rejected for k=3.

1. 4.6666
2. 1/3*(1+4+9)

I guess it's all just floating point error, but (2) is a bit more puzzling
since (1+4+9)/3 is fine.

Do I need to specify some slop in my upper and lower bounds?


Wasabi peas, and dark chocolate chips. It's better than you might expect (TM).

Journal Shaming: Bioactive Materials 

I tried to be more civil this time, and try to explain the risk they are taking.

I am not the David Bremner you are looking for. Please update your database, every time I get an unsolicited email about reviewing in areas that have nothing to with my research, I name and shame the journal on the internet.

Show thread

Journal Shaming: Bioactive Materials 

Elsevier journal editors continue to rely on a corrupted / incomplete database of reviewers to send me papers to review in materials science and related disciplines. At least in this case, the editors apparently have zero understanding of the expertise of the people they are selecting as reviewers. Because clearly if I have the same name, I must have the same expertise.

This weeks "winner" is Bioactive materials.

Working out how to explain randomized median finding, I was reminded that $∀|x|<1$, \[ \sum_{k=0}^{\infty} k x^k =\frac{x}{(1-x)^2}\]
It's still a bit surprising after all these years that adding the extra factor of $k$ doesn't really change anything, it still converges to a constant.

Also, Appendix A of CLRS is a great place to find the formulae you need 😜

Me in the 1990s: OK, I really can't just run LaTeX every time I make a change, it takes too long.

Me in the naughty's: Lalala computer so fast don't care just pound keys

Me in 2020: I OK, I really can't produce render the video from this LaTeX file every time I make a change, it takes too long.


I think I'm over maniacally following US politics. I'm not sure if this is sanity or despair.

Hot Take on "Learning Management Systems" 

I wonder if I will actually hate Desire2Learn less after it is taken over by Microsoft.

when i hear weakly typed vs strongly typed i cannot stop thinking about these images

Real commit message: "master theorem slides, plague version"

To be fair, not that many people came to my office hours in the before times either.

Show thread

My first online office hour has been either an utter failure or complete success, depending on how you view having a quiet hour to myself to catch up on email and make part of a quiz.

New Combinatorics journal to watch

Intended to be a successor to JCTA. I hope they sort out a TLS certificate soon.

Show more

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!