Geeky things and other research from a work in progress


"Extensibility and type safety in formatting: the design of xformat" at the Dutch HUG

I'm finally getting around to posting the slides for my talk at the Dutch HUG on 11 September. After that, I was preparing my talk for IFL on incrementalization of datatype-centric programs. I arrived back in Utrecht yesterday morning and crashed mid-day.

You can find the slides on Scribd and on Github. Enjoy!

Update: Thanks to Tom Lokhorst, my talk is also on video.


  1. I've tried the video, but the sounds is very badly broken up, to the point of making it unwatchable. I'm pretty sure it's not a problem at my end. Has anybody else had the same problem?

  2. It's definitely a problem with the recording, not the system you're playing it on. It's unfortunate that the sound got mangled. Hopefully, Tom can figure that out the next time he records and prevent it.

  3. So, I have some questions about xformat:

    1) Are you still working on this?
    2) Are you going to provide facilities to print numbers in different bases?
    3) Are you going to provide wrapper functions to use I/O for these formats?
    4) Is there an all-inclusive wrapper module to do both input and output with one import?

  4. 1) I certainly still want to work on it. I haven't had a lot of time recently, but I had an idea or two to try.

    2) It is possibly, though I hadn't thought about it. But in fact, I do have another ongoing project to get the digits of numbers in an arbitrary base.

    3) IO wrappers would be a simple addition.

    4) Opinions vary on how this should be done. Right now, you can do qualified imports. Do you have a different thought?