[pypy-dev] chat log past 8:00 pm GMT

[Christian Tismer]

This is a cut'n paste protocol, since I had no logging on :-)

*** Topic is 'Python in Python!  Strange?  Maybe, but powerful - and can 
be done!'
*** Set by ChanServ on Fri Apr 11 16:01:08
*** ReportLabRobin has left #pypy
<lac> hi christian
<Stackless> howdy laura, although I'm late, I'm here (hear) :-)
<hpk> hi christian (this is holger)
<lac> you just missed Guido
<Stackless> Have you settled a date already?
<Stackless> Hi Timmy!
<hpk> two :-)
<lac> not for May, but looks liek the EuroPythonj sprint is on
<hpk> oh, but the end-of-may sprint hasn't died, has it?
<lac> it most definitely hasn't.  But the people who said they could 
come or could probably come are the same ones as weren't able to make 
this irc chat
<Stackless> Is there a log of the recent messages, so I can keep up with it?
<hpk> guido sent it to pypy-dev
<lac> yes.  Guido made it.  he is sendiung it to pypy-dev
<Stackless> so this is the time to read pypy-dev again. :-)   (and I'm 
ready for it)
<Stackless> ...loading the archives...
<hpk> Stackless: how is it going?
<lac> well, nbot much heppened after you went away
<lac> Armin will be done moving on Monday
<Stackless> Well, I shrunk myself so much that I became focused again, 
saved my US support, and I'm almost ready to push SLP 3.0 out of the door.
<hpk> cool. would be a good time to write that article (ducks)
<Stackless> Yes, a couple of days before I checked the subversive files, 
and there was not much new.
*** sp has quit IRC (Read error: 110 (Connection timed out))
<hpk> Stackless: yes, just ten new revisions
<Stackless> hpk: when I looked a fe days before, there were only three.
<Stackless> hpk: What, you didn't write the article yet? (har har)
<hpk> whatever. we should try to discuss  pypy-development on the list 
in april
<Stackless> hpk: Still fighting with the Trillke environment issues?
<hpk> sure. we are about to get the city principals into problems
<Stackless> -off for 2 mins-
<sp2> whois lac
<hpk> laura creighton
<sp2> oops, I know (thanks)
<Stackless> who is sp2, then ?:-)
<lac> I think we have a great chance to get Commission funding for Pypy
<lac> Sameule Pedroni
<hpk> lac: sounds good.
<lac> (I think -- he was sp at any rate)
<Stackless> Oh, that's great. You bot has a bug: The name is "Samuele"
<lac> I am such a lousy typer
<sp2> yup, that's me (sp2 because the computer at one point crashed on me)
<Stackless> Laura, I missed all of that discussion. How do you think to 
get the funding? This is incredible!
<lac> I went to the commission's page and found out what they want to 
spend money one
<lac> er on
<Stackless> Did you ride the horse of "European answer to Java"?
<lac> Yes, They want this.  I haven't written a proposal yet.  I am only 
at this one week
<lac> and what i have now is an understanding that what they want to do
<lac> is pay money for
<lac> a) open source alternatives to corporate things
<lac> b) things that are European and not American
<lac> c) people who count as scholars who want to visit each other to 
collaborate
<lac> d) networking and handheld devices
<lac> d) is where Armin is already getting his funding
<Stackless> Whow, I'm very impressed. So I should stop all my current 
projects yesterday, right?
<lac> No, no
<Stackless> :-)
<lac> these thigns take time.  And indeed we have a small problem in this
<lac> we need to get a proposal out the door by April 23, or then its 6 
months before they approve the next lot.
<lac> I don't know if we can do this in time.
<lac> We can make the next deadline easily, of course.
<lac> look at http://fp6.cordis.lu/fp6/call_details.cfm?CALL_ID=1
<lac> as the place to start for where they want to hand out money
<lac> The idea is to see what is it that we can/will be able to do that 
they want funded
<lac> Then we write a proposal, saying, give us money so we can do this 
thing you want
<lac> we are a little more polite than that, of course.
<lac> Object Spaces are a cool idea for the EU because they have lots of 
problems with people who never want to upgrade their software
<lac> We could, in principal, say 'this is Oct 2003 Python Object Space'
<lac> and then let all the people who never want to change it again, run 
in that
<lac> while the rest of us move on to something more interesting.  All 
at the same time.
<Stackless> Music in my eyes.
<lac> This will save a dramatic amount of money in upgrading software 
that never wanted to be upgraded anyhow
<Stackless> As I just skimmed the session log, BDFL and Tim might not 
make it to Godersburg, but to EuroPython?
<lac> its Gothenburg, byt he way
<lac> of Gvteborg
<Stackless> stack overflow, sorry.
<lac> oops, they don't do o with an umlaut here
<sp2> the trick work as long as 'Oct 2003 Python Obj Sp' does not depend 
on something that can change under its feet
<Stackless> Non comprehende?
<lac> yes indeed. and here we get to the part where I don't understand 
how that works
<lac> This is what Armin was explaining to me before my brain overflowed 
with new information and I couldn't absorb any more
<Stackless> (how I love that known feeling...)
<lac> I miss it.
<Stackless> Please, explain, where is a trick, here?
<lac> I think what Samuele meant is that it is nice to think of object 
spaces as being separate, but are they really? if down at the bottowm 
something becoems new and influences the old object spaces, then things 
will nto work
<lac> correct?
<Stackless> Talking about an Oct 2003 Py O-SP before April 23, without 
really knwoing what we're talking about?
<hpk> ok , just updated http://codespeak.net/pypy/ to reflect the 
current situation
<jacob> I have done a bit of thinking about the interaction between 
objectspaces. I think we need defined translations from on ObS to another.
<Stackless> I think that I think that objectspaces *are* separate, 
especially they are not related to the language you want to implement.
<sp2> if you want to achieve the goal the ObjSp as repository of some 
semantics should be isolated/captured indipendently from os / execution 
substrate, and in particular if you want to run it in 3000 there should 
be a way to interpose emulations for older OSs/substrates
<lac> looks good to me holger
<lac> somebody should log this, and I don't know how
<sp2> I'm logging it
<lac>  great
<lac> does this imply translations between one ObSpace and another, or 
are these separate problems?
*** moshez has joined #pypy
<Stackless> Hi moshez
<moshez> hi, christian
<Stackless> long time no see.
<lac> hi moshez
<moshez> christian; come to #twisted, I'm always there :)
<moshez> I just got here to talk to lac about logs :)
<moshez> lac: it's fairly easy if you install Twisted
<moshez> it's in doc/examples/ircLog.py
<moshez> you just have to change some hard-coded variables :)
<moshez> and make sure the file you log to is web-site-accessible
<lac> ok, I have Twisted installed, thank you.
<moshez> lac: via Debian?
<lac> yes
<lac> but I have a recent source tree as well
<moshez> lac: how about the twisted-doc package?
<lac> that i have as well, at least on my laptop. not sure about this 
machine
<moshez> lac: now, IIRC, it's in 
/usr/share/doc/twisted-doc/examples/ircLog.py
<moshez> or something like that
* moshez checks
<moshez> ircLogBot.py
<moshez> same path :)
<moshez> cp it somewhere, change the nickname attribute to something 
appropriately descriptive (I used moshezbot) and run it as ircLogBot.py 
#pypy <name of file>
* moshez pokes lac
<lac> hi moshez.  I am on the phone,
<lac> but I will do that in the future. thank you.
<moshez> np
<Stackless> moshez, you want em on twisted? Will come, soon.
<jacob> j #twisted
*** chrchr_ has joined #pypy
<Stackless> Is this channel doomed, now, so I should send the extra log 
to pypy-dev, or do we continue?
<chrchr_> Stackless: Why would it be doomed?
<Stackless> Just because the mesage frequency got down below 1/min :-)
<chrchr_> Ah.
<lac> well I am on the phone. But i am very interested in discussion 
about object spaces if anybody wants to continute that so i CAN READ IT
<lac> oops sorry for capslock
<hpk> i thought you switched to italian style :-)
<Stackless> : /that/ *I* _really_ "don't" --understand--  (hand waving)
<Stackless> hypertonically y'rs -- chris
* hpk loughs
* lac smiles at Christian
* Stackless has to filter the log before posting...
* Looking up chrchr_ user info...
<Stackless> Ok, maybe I missed it: Did you come up with a final date for 
the next sprint?
<lac> no. but last week in May.  what is best for you?
<hpk> have to leave now. see you soon.
<lac> ok, take care holger
*** hpk has left #pypy
<Stackless> bye, hpk
<Stackless> I planned to do one week of holidays before may, 7th. I 
would have turned that into
<Stackless> a sprint. On the other hand, having both is just fine. What 
are you heading for: Yet another week, or shorter?
<lac> everybody who wants to come couldn't make tonights meeting.
<lac> so a week is fine with us.
<lac> accomodations are reasonable, and the airfaire to Gothenburg from 
Frankfurt, Hamburg and London Stanstead is cheap
<lac> we have access to a huge machien room at Chalmers university, so 
lots of machines and bandwidth
<Stackless> Ok, so I'm keeping the last 12 days of May open for 
everything -- count on me for PI(T|ZZ)A
<lac> though they are old machines -- our lapstops will be faster
<lac> cheer
<Stackless> Going off for today -- should I post my logs?
<lac> yes please
<lac> I should go as well.
<lac> take care Christian, all
<Stackless> See 'ya, and thanks for the work with the funding -- I want 
to help, definately!!
<lac> terrific
<lac> we discuss this some other night.

-- 
Christian Tismer             :^)   <mailto:tismer at tismer.com>
Mission Impossible 5oftware  :     Have a break! Take a ride on Python's
Johannes-Niemeyer-Weg 9a     :    *Starship* http://starship.python.net/
14109 Berlin                 :     PGP key -> http://wwwkeys.pgp.net/
work +49 30 89 09 53 34  home +49 30 802 86 56  pager +49 173 24 18 776
PGP 0x57F3BF04       9064 F4E1 D754 C2FF 1619  305B C09C 5A3B 57F3 BF04
      whom do you want to sponsor today?   http://www.stackless.com/