Camelia, the Perl 6 bug

IRC log for #crimsonfu, 2013-04-10

crimsonfu - sysadmins who code

| Channels | #crimsonfu index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:02 melodie good night
01:18 codex joined #crimsonfu
01:24 whorka joined #crimsonfu
05:13 codex joined #crimsonfu
06:07 Itkovian joined #crimsonfu
06:30 jimi_c joined #crimsonfu
07:33 chasmo joined #crimsonfu
07:47 Itkovian joined #crimsonfu
09:54 melodie_ joined #crimsonfu
09:54 melodie_ joined #crimsonfu
12:09 chasmo joined #crimsonfu
12:53 Whoop joined #crimsonfu
12:53 Whoop joined #crimsonfu
13:00 pdurbin ben_e: you around?
13:04 jimi_c joined #crimsonfu
13:10 melodie_ hi pdurbin I have sent the mail earlier today about the cs50 appliance. (it was uploaded during last night and I could provide a link to it in the mail)
13:11 ben_e yes
13:12 ben_e tastypie talk is today
13:12 pdurbin heh. yep! ABCD WWW » Eric Mattison – Easy APIs to make your life easier - http://w3.abcd.harvard.edu/ai1ec_event/eric-m​attison-django-and-tastypie/?instance_id=198
13:12 pdurbin had the link ready :)
13:14 ben_e the 3 year old brought home some daycare plague last week which i managed to avoid until yesterday
13:15 ben_e so i'm not entirely sure i'm going. we'll see what 3 PM feels like
13:17 pdurbin something is definitely going around. 103 degree fevers in my group
15:18 jimi_c_ joined #crimsonfu
15:19 perun0 joined #crimsonfu
16:09 jimi_c__ joined #crimsonfu
16:42 jimi_c_ joined #crimsonfu
17:17 jimi_c__ joined #crimsonfu
19:51 Itkovian joined #crimsonfu
19:52 JoeJulian Argh... I spent half the day yesterday re-writing the authentication for a web site I manage, creating a much more robust password storage and tokenized authentication system for it. Then as I'm falling asleep I realized I should have just used keystone.
20:01 semiosis code is a liability
20:02 JoeJulian I was really pleased with what I wrote too, 'till I realized that what I did was basically what keystone already does, with the advantage that a keystone server is an api abstraction from the database that someone would most likely want to break in to.
20:02 JoeJulian Well, second most likely.
22:22 pdurbin cool that keystone is that pluggable
22:40 JoeJulian Well, I don't know about pluggable, but it's an api. I should just use curl to authenticate, get a token and verify that token on subsequent access (along with validating and changing the session id, and validating whatever else one thinks shouldn't change - like maybe the user agent string).
22:54 e1mer joined #crimsonfu
23:09 pdurbin does keystone talk SAML? I need SAML :)
23:11 pdurbin holy shit. "It has recently been rearchitected to allow for expansion to support proxying external services and AuthN/AuthZ mechanisms such as oAuth, SAML and openID in future versions." -- https://wiki.openstack.org/wiki/Keystone ... maybe some day! :)
23:12 JoeJulian Doesn't seem like it should be all that hard to add...
23:32 pdurbin westmaas: any idea of the timeline? when keystone will support SAML?
23:32 westmaas sorry not really :/
23:33 westmaas hey did any of you guys have any openstack commits?
23:33 westmaas cause if yes... http://iopenedthecloud.com/
23:33 westmaas 500/mo free cloud hosting at rs for 2 years
23:34 pdurbin JoeJulian: you've made commits
23:35 pdurbin I'm still not comfortable enough with SAML to understand how I'd use it with an API provided by a service like keystone
23:35 pdurbin I was planning on trying to get SAML from OpenAM
23:36 pdurbin but just today in #openam I was hearing that OpenAM and Shibboleth have different interpretations of the SAML spec :(
23:36 pdurbin and I'll need to have my app (a Service Provider or SP) auth against a Shibboleth-based Identity Provider (IdP)
23:39 pdurbin of course, I haven't been working on all the auth stuff lately. it's been taking a back search to me adding faceted search to our app. I'd love some feedback if anyone wants to try it out: http://dvn-3.hmdc.harvard.edu/dvn/
23:42 pdurbin it's a little slow going. today I wrote up a bit of a roadmap for myself to talk through the issues: http://dvn.github.io/devguide/f​eatures/search/faceted/roadmap
23:44 St34LTH joined #crimsonfu
23:44 pdurbin JoeJulian: if you're still around... something else I spent a bit of time on was Shapado. I'm a little worried it's dead. no commits since August 2012. no one responding in IRC. can't find where bugs are being actively tracked
23:44 JoeJulian I don't think they develop it as a community project. It's an internal project they just happen to open-source.
23:45 JoeJulian At least that's the way it appears to me.
23:48 JoeJulian Ok, I read your roadmap as: My business logic is too closely tied to my data source. I should put an api abstraction layer between the business logic and the data so that the data store is simply a re-implementation of that api.
23:52 pdurbin JoeJulian: yeah, it is tied quite closely to Lucene. lots of passing Lucene BooleanQuery objects around. I'm thinking I'll try to encapsulate that somehow. So we can switch to Solr or ElasticSearch some day if we want to. I need to re-write a bunch of methods anyway. I'd might as well re-think it.
23:53 pdurbin JoeJulian: thanks for reading it :)
23:53 JoeJulian I'm in the process of doing the same thing here.
23:54 JoeJulian Luckily, I can point the blame at my predecessor.
23:56 pdurbin :)
23:56 pdurbin at the time Lucene didn't have features we now want, like faceting
23:57 pdurbin but as with your keystone thing there's also this though of switching to a web service like solr of elasticsearch. something with a web api
23:57 pdurbin but that's a whole can of worms, re-implementing all of our business logic, like that roadmap says
23:58 pdurbin first I need to understand the business logic :)

| Channels | #crimsonfu index | Today | | Search | Google Search | Plain-Text | summary

crimsonfu - sysadmins who code