Request for Help

Repost of my request on LtU:

Any help with good starter papers in these areas would be appreciated! Prior knowledge would probably be around TAPL level. I'm hoping for papers that are a) primary sources b) clean/modern formalizations or c) provide perspective (or a retrospective). A mixture of b) and c) would probably be ideal :) Realistically, book chapters are less useful than papers, unless they're from a relatively common text (CTM, etc.) or freely available. Talks/slides might be good too.

I have guesses for a couple, but the community here can probably do significantly better and help suggest which makes the most sense for the goal.

1. Dependent Types
- "Simply Easy! (An Implementation of Dependently Typed Lambda Calculus)"
- "Why Dependent Types Matter"
- TAPL 23,25

2. Concatenative languages (pure & typed? impure? optimized?)

3. Concurrency/distribution/communication calculi
- "A calculus of mobile processes"
4. Linear naming / typing

5,6. Modal logic (with an eye to distributed comp.), separation logic (distrib. comp or heap sep.)
- likely papers from CMU :)

Again, any help would be appreciated!

This is for the good of grad. students today, and hopefully for generations to come :)

how to write a paper

So clean! I rarely understand type theory papers, so these were a welcome change :)

• Gradual Typing for Functional Languages
• Gradual Typing for Objects
• Gradual Typing with Unification-based Inference

Lurking behind the scenes is parametricity/polymorphism. That has already started to show up (go Arjun! hadn't thought of your paper in that light) and we'll probably see more soon :)

Now I might actually have a shot at formalizing gradually typed transparency! This could be a good chance to learn Coq/Isabelle/?. Too bad our resident expert is now at Harvard :(

There were no open revolts at my bayfp talk tonight, so I'm counting it as a success :) I think it was recorded, and the slides are available on my site for now (ppt 2008, pdf). Not sure how to embed it into blogger (phopkins, if you're out there, help!)... that I could even think of doing that makes me impressed at how far the web has gotten in the past 3 years.

Thanks to everybody who came, Mike and Keith of Skydeck for organizing, and the folks at Heysan! for hosting - tonight was fun and it was good to meet s'more people. One thought is to go out for a post-talk drink next time, being on Thursdays and all :)

Now go and register something aweso.me . Even better, if you're a language person in the area, sign up to talk! I might give you a hug if it's a crash course on modal logic or linear types...

Presents

I'll be subjecting BayFP to my experiments on using FRP for the web (flapjax tutorial, rundown of my adobe project, maybe some future adobe or parlab directions) next week. If anybody wants to come and field the questions about monads, co-data, functors, and arrows, I'll be the one not answering them :) Not sure about the merit of discussing glitches and conditionals either - perhaps a a choose-your-own-adventure format would be more appropriate.

To prepare, I whipped up a fun little Flapjax* repl last night. Firefox only and made for large adjustable fonts.

More amusing is my experiment in latency (explanation forthcoming).

But first... the fire festival!

*Slightly modified Flapjax lib with a few untested fixes & conveniences

Legible Row Coloring

While working on some latency demos, I came to the following thought on tabular representations:

Bad:

a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c

Better:

a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c
a	b	c	a	b	c

Edit: removed flawed analysis. It's interesting to think about this with respect to bits of information conveyed to support quickly scanning a row horizontally.