Next day: 2017-10-23 →
#dlab logs for 2017-10-22
- rawles (Simon Rawles) (new topic)
A distributed laboratory. Everything which appears here is logged.
- adamfc (Adam Forsythe-Cheasley)
I've invited Tim here so too.
Today I'm implementing a basic dependency manager in Java, which allows you to define expressions and variables, then when one variable is changed, so are the ones it depends on, automatically. Like in a spreadsheet.
so the dependencies are anything?
i.e. are they objects?
So a dependency is 'a is b + c'. The variable a is dependent on b and c, and takes the value b + c. When either b or c change, a changes too.
I could have just put '=', but that suggests assignment in programming to me.
I've got a big plan for this, but this is the ground work.
Do you have a project right now?
I was reading a bit more about homotopy type theory
I guess that was the closest thing to a project recently
it was interesting
esp. in the fact that you could build up everything from it
and it's essentially spaces
I keep meaning to read the PDF book that Adam linked to and put it on my Kindle -
which reminded me of an idea you were talking about where you were representing logical statements as spaces
oh I forgot he linked to a book
from his post at
oh this is good, you can just read the book for free
There's some order theory in there, so I should really take the time to look at it! What do you find interesting about it?
Is there anything you could do with it, like a programming project?
I guess at the moment I find it interesting that it's *not* set theory
but I guess also that it's easier to encode
so you can have a computer check a proof
that's quite interesting
I think people are working on a framework and a language with homotopy type theory as the basis
The book I was reading ages ago was
Like a programming language?
'a formal proof management system'
I've heard of Coq but never looked very deep into it. This seems interesting.
Right, that's on my reading list then.
Oh, it's in OCaml too.
yeah, I was just seeing that
have you done much OCaml?
No, but there are lots of fans of it among the programmers we know in Bristol.
I don't know anything about it really
I've not looked at it, no
Luke, a friend of Rick's, posted this:
Let f(n) denote the number of digits of n (in base 10) that are greater than or equal to 5 (so f(128) = 1, f(1024) = 0 and f(1048576) = 4). What is the sum of f(2^k)/(2^k) (from k = 0 to infinity)?
All timestamps are in UTC.
If you want to join in, channel information is available.
This page is generated from raw logfile dlab-2017-10-22-000000.txt.