#jython IRC Log

Index

IRC Log for 2015-06-10

Timestamps are in GMT/BST.

[0:35] * fwierzbicki1 (~Adium@99-106-169-5.lightspeed.sntcca.sbcglobal.net) Quit (Ping timeout: 244 seconds)
[0:36] * enebo (~enebo@c-75-73-8-169.hsd1.mn.comcast.net) Quit (Quit: enebo)
[0:50] * fwierzbicki (~Adium@37.sub-70-197-11.myvzw.com) has joined #jython
[1:30] * fwierzbicki (~Adium@37.sub-70-197-11.myvzw.com) Quit (Quit: Leaving.)
[1:41] * lheuer1 (~Adium@f048016139.adsl.alicedsl.de) Quit (Quit: Leaving.)
[2:35] * gopar (~gopar@2601:9:1100:4c50:a6db:30ff:fea2:873) has joined #jython
[2:52] * koo6 (~koo5@236.152.broadband3.iol.cz) Quit (Ping timeout: 276 seconds)
[6:24] * gopar (~gopar@2601:9:1100:4c50:a6db:30ff:fea2:873) Quit (Quit: Leaving)
[6:27] * lheuer1 (~Adium@f048016139.adsl.alicedsl.de) has joined #jython
[7:31] * lheuer1 (~Adium@f048016139.adsl.alicedsl.de) Quit (Quit: Leaving.)
[7:36] * mbooth (~mbooth@redhat/mbooth) Quit (Ping timeout: 250 seconds)
[9:03] * mbooth (~mbooth@redhat/mbooth) has joined #jython
[10:33] * mbooth (~mbooth@redhat/mbooth) Quit (Ping timeout: 245 seconds)
[10:54] * mbooth (~mbooth@redhat/mbooth) has joined #jython
[11:27] * koo6 (~koo5@236.152.broadband3.iol.cz) has joined #jython
[13:19] <gthank> So I finally got started with jimbaker's pycon talk, and I have a question: Isn't Jython dead?
[13:32] * xemdetia (~xemdetia@129.42.208.176) has joined #jython
[13:55] <pdurbin> gthank: nope. 2.7 was recently released
[13:56] <gthank> pdurbin: I know :-) Jim just covers the topic??? extensively??? at the beginning of the talk.
[13:56] <gthank> Also, congratulations on the release, everybody.
[13:57] <gthank> I'm looking into doing some "online" machine-learning stuff in Apache Spark-Streaming, so I may be getting back to using Jython more as part of my job, so YAY
[14:11] <pdurbin> cool, some folks at work are using spark
[14:13] <gthank> It seems pretty flipping awesome
[14:14] <gthank> At least from what I can tell, Spark-Streaming is exactly what I've been trying to design for myself off-and-on for the last year or so
[14:20] <xemdetia> is spark just a distriubted cluster thing
[14:20] <xemdetia> or am I missing the ML angle
[14:21] * enebo (~enebo@c-75-73-8-169.hsd1.mn.comcast.net) has joined #jython
[14:45] * koo6 (~koo5@236.152.broadband3.iol.cz) Quit (Ping timeout: 256 seconds)
[14:52] <gthank> Yeah, it does distributed cluster things, but distributed cluster things are really important to modern ML stuff
[14:53] <gthank> Since none of them work very well (you might even call it not working at all) without a LOT of data
[14:53] <gthank> And I'm fairly sure there are some machine learning libs baked right in?
[14:54] <gthank> I could be wrong, since I haven't looked in a few days (knee-deep in something unrelated, right now)
[15:47] <xemdetia> yeah they have the usual simple decision tree stuff
[15:48] <xemdetia> I mean I usually use weka
[15:48] * mbooth (~mbooth@redhat/mbooth) Quit (Ping timeout: 276 seconds)
[16:00] * ELFrederich (~ELFrederi@12.54.94.108) Quit (Remote host closed the connection)
[16:11] * ohumbel (5390f784@gateway/web/freenode/ip.83.144.247.132) Quit (Ping timeout: 246 seconds)
[16:13] * fwierzbicki (~Adium@99-106-169-5.lightspeed.sntcca.sbcglobal.net) has joined #jython
[16:15] * mbooth (~mbooth@redhat/mbooth) has joined #jython
[16:25] * ohumbel (5390f784@gateway/web/freenode/ip.83.144.247.132) has joined #jython
[17:51] <gthank> Yeah, I don't think Weka would work for me. I'm talking about a backlog of billions of records, and streaming 100+ (or, depending on how the granularity plans out, 1000s) more through the pipeline every second.
[17:52] <gthank> Weka seems to want everything in a single flat file?
[17:55] * xemdetia (~xemdetia@129.42.208.176) Quit (Ping timeout: 276 seconds)
[18:32] * koo6 (~koo5@236.152.broadband3.iol.cz) has joined #jython
[18:55] * koo6 (~koo5@236.152.broadband3.iol.cz) Quit (Ping timeout: 276 seconds)
[19:51] * xemdetia (xemdetia@nat/ibm/x-usrmkyyegliqdvti) has joined #jython
[21:53] * xemdetia (xemdetia@nat/ibm/x-usrmkyyegliqdvti) Quit (Quit: Leaving)
[22:18] <jimbaker> so i have worked with spark streaming. one downside is that spark currently uses py4j for its python support, which means that everything in spark needs to be wrapped to be used by python code
[22:21] <jimbaker> also py4j uses tcp sockets (not unix domain sockets), with no security other than firewalling, so it basically allows for remote execution of any code that can be run by the host jvm
[22:21] <jimbaker> other than that, it's great :)
[22:23] <jimbaker> gthank, fwiw - most of my work on spark has been in scala. but i do hope to see jython support at some point - should improve spark's python performance, security model, etc
[22:39] * enebo (~enebo@c-75-73-8-169.hsd1.mn.comcast.net) Quit (Quit: enebo)
[22:42] <jimbaker> https://cwiki.apache.org/confluence/display/SPARK/PySpark+Internals
[22:43] <jimbaker> and http://py4j.sourceforge.net/faq.html#are-there-any-security-concerns-with-py4j
[22:46] <jimbaker> someone should probably respond to http://www.quora.com/Is-Jython-alive ;), but there's this problem: https://xkcd.com/386/
[22:59] <pjenvey> i think someone added pypy support at some point, FYI

Index

These logs were automatically created by JythonLogBot on irc.freenode.net using a slightly modified version of the Java IRC LogBot.