<beneroth>
[OT] joke of the day: how to prevent HTTP remote exploits - by Cisco (one of the market leaders in network and firewall equipment, and NSA backdoors) - just blacklist user-agent string "curl": https://twitter.com/RedTeamPT/status/1110843396657238016
<Regenaxer>
Clever trick!
<Regenaxer>
Nobody will use the --user-agent option ;)
<beneroth>
exactly ;)
<beneroth>
I use it even when I have no reason to hide that it I'm using curl - just to propagate "picolisp" as a term :P
<Regenaxer>
ha, cool! :)
<rick42>
beneroth: right on!
<rick42>
hello everyone
<Regenaxer>
Hi rick42!
<rick42>
i hope you are all well
<rick42>
hi Regenaxer! :)
<Regenaxer>
yes, fine, you too?
<rick42>
yes, thank you for asking
<rick42>
i saw the latest ML convo about (local)
<rick42>
interesting
<rick42>
looks a bit strange to be since i'm used to seeing (and thinking about things) like: (local ...)
<rick42>
to me
<rick42>
but hey if it works to overcome any issues, im for it!
<rick42>
aw will have to change his lib code. heh
<Regenaxer>
yes, local cannot take args
<Regenaxer>
What is confusing perhaps is that namespaces are a pure reader issue
<rick42>
ah!
<rick42>
that explains it
<Regenaxer>
So (local sym ..) will have read 'sym' already
<Regenaxer>
(read) runs in the new namespace *only*
<rick42>
ok, yes it is very clear now
<Regenaxer>
so it creates these symbols if not there already
<Regenaxer>
:)
<Regenaxer>
Often the "R" in "REPL" is harder to understand than the "E" :)
<rick42>
So:
<rick42>
(local) (x1 x2 x3)
<rick42>
has the same effect as:
<rick42>
(local) (x4 x5 x6)
<rick42>
(local) (x1 x2 x3 x4 x5 x6)
<rick42>
?
<Regenaxer>
yes, exactly
<Regenaxer>
or (local) x1 (local) x2 ...
<rick42>
ok. yes R is harder to understand. for me, it is because i tend not to think about the reader too much
<Regenaxer>
me too
<Regenaxer>
we concentrate a lot on the "E"
<rick42>
more is going on in E anyway :)
<Regenaxer>
true
<rick42>
the interesting part
<Regenaxer>
yes, and it can be anything, while the reader is quite pre-defined
<Regenaxer>
(except for read macros)
<beneroth>
hi rick42 :D
<beneroth>
nice talk :)
<beneroth>
I guess the E is intuitively easier to grasp (or at least SEEMS so) than the R step, because the E step is similar to what people know from other stacks, while the R part is pretty Lisp-specific (it is there in other interpreted languages like python, but it is not as powerful and not as diverse as with lisps)
<Regenaxer>
Very well described beneroth!
<beneroth>
even more so when people, being sceptical about their ability to understand such a thing, assume magic and think it to be more complicated then it is instead of seriously attempting to go through the algorithm steps in detail just once.
<Regenaxer>
yeah
<Regenaxer>
It is so simple that it gets hard to explain (at least for me)
<beneroth>
T
<beneroth>
well sometimes (I could observe this both here in the channel as also in personal talks irl) people are very confident that something must involve some complicated X, and you explain them the thing and the explanation does obviously not contain X, so people think you left something out, sometimes they even get suspicious you left X out on purpose, when there simply is no X anywhere...
<beneroth>
and then they cling so much to this X that they don't really look at the parts of the thing...
<Regenaxer>
indeed
<Regenaxer>
They think there must be one big magic box, instead they put together a few simple pieces by themselves