#haskell - Thu 22 Mar 2007 between 00:00 and 00:18

NY Lost Funds



heatsinkAre there ISA differences from 586?
pentium
kc5tjafantasma: I can check...
fantasmakc5tja, if you let me know where, I can check (I've tried googling but it's hard to find)
kc5tjafantasma: processor : 0
vendor_id : AuthenticAMD
cpu family : 6
model : 2
model name : AMD Athlon(tm) Processor
stepping : 1
It is indeed a 6th revision, from what I can see -- unless I'm interpreting the fields incorrectly.
(and to think! I thought I was the last of the slot A users!)
I did cat /proc/cpuinfo to get the above information in Linux.
donssjanssen: oh?
you mean the protocol?
fantasmakc5tja, I was so surprised to look inside two of my machines the other day....they were both slot A's
soreardons: is this wrt the 64 bit issues?
hpastesjanssen pasted "xcb and binary" at http://hpaste.org/1081
soreardons: if so: the haskell X11-bindings are borked with *extensive* Int/CInt confusion, ie Not Our Problem
donsi know.
sjanssen: ah ha
sjanssensorear: not exactly. We want to ditch -package X11 in favor of a new library based on XCB (which comes with spiffy XML specs)
SgeoWould prolog be more suited to math stuff like derivatives than Haskell?
sorearSgeo: a function is opaque
Sgeo: the abstraction is one of the defining features of haskell functions
Sgeo: this makes it possible to do arcane tricks, like compile them to machine code
sjanssendons: xhsb uses a handrolled state/reader monad
donsurgh
sorearnot xhb anymore?
sjanssenwell, not exactly handrolled. They're ReaderT and StateT
sorear: xhsb is the very rough beginnings of a binding that somebody over at X.org started
sorearso x.org gave up?
do they know we've taken over?
should they?
sjanssenby very rough, I mean it supports only two functions
sorear: nobody is actively working on it. It's more of a proof of concept, I think
if I actually start writing a replacement I'll post about it on their mailing list
dons: the neat thing is that the XML spec describes both their C API and the actual bits-on-the-wire in the X protocol. So you could start with their C connection layer early on, and graduate to a pure Haskell implementation later
allbery_bhah. the CMU SCS folks are discussing how to do debug printing nicely in SML. it looks remarkably primitive compared to Haskell
sjanssenthe connection layer could even be a type class, so you can choose which to use at runtime

Page: 2 9 16 23 30 37 44 51 58 65 72 79 86 93 100 

IrcArchive

NY Lost Funds