@mflatt @robby It looks like this commit https://github.com/racket/racket/commit/69d7636770acc968fc448bc896595e66d8ff570b broke this test http://drdr.racket-lang.org/38650/racket/share/pkgs/drracket-tool-test/tests/check-syntax/syncheck-direct.rkt, but my guess from the commit message is that the new behavior is right. Should I just change the test?

@mflatt also, it looks like https://github.com/racket/racket/commit/572b96a6efc4dcb6c4e00fd4abeaedcae5eae868 broke this test: http://drdr.racket-lang.org/38641/pkgs/compiler-test/tests/compiler/embed/test.rkt

@samth: yes, thanks

@mflatt Should we make the change we talked about related to PLT_DELAY_FROM_ZO?

Do you mean turn in on by default? No, I don’t think we should do that.

@mflatt no, I mean patch read.c so that it stops crashing

ie, is that change the right patch?

I forgot that we didn’t merge; please do

ok, will do

@lexi.lambda Not sure if you still need the info, but on my mac that snippet yield a “certificate verify failed” error

Guys, I post this in #beginners, but since that channel has way fewer people in it, I figured I might as well post it here and see if I get a response faster: > Guys, because I get more and more confused the more I read about it, continuations are just functions which pass their state as arguments to other functions, right?

you can think of the continuation of a computation as what’s left to do after you execute the next step of the computation

so for example, in the computation (– 4 (+ 1 1)), the “next step” is to add 1 and 1, and the continuation is to subtract whatever the result of that is from 4

(see the Racket evaluation model in the guide http://docs.racket-lang.org/reference/eval-model.html)

(there are also more academic/PL theory references I could find for you)

things like call/cc let you capture a continuation (like “subtract the result from 4”) at a particular point in the program and use them like you might a function

ahh, I think I got it @thinkmoore. Thanks a lot. I would welcome the more PL Theory references though, if it’s not too much of a hassle.

if you have access I think there is a good treatment in the PLT Redex book

@thinkmoore Thanks. Noted.

i think we could pay for longer history

@thinkmoore sure, but $

… but $7 per month per user is perhaps gonna dissuade us :slightly_smiling_face:

oh wow

yeah…

At least I suspect the API is open enough to take out the data when you decide you want to leave, with a bit of elbow grease.

I once migrated a team from Slack to a self-hosted Mattermost installation. It was a breeze.

If we have the resources to run Mattermost (servers and system administrators), I believe it’s better than Slack: 1. Less expensive (when compared to paid Slack, of course). 2. Unlimited history. 3. More customizable front-end (e.g., change fonts, colors for whole UI). 4. Easier to sign up. 5. Better Markdown support (e.g., links).

leafac: We’ve actually been trying to move away from having our own servers and system administrators. :)

As the system administrator of a few services myself, I understand :slightly_smiling_face: