Perl 6 - the future is here, just unevenly distributed

IRC log for #crimsonfu, 2014-03-20

crimsonfu - sysadmins who code

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

All times shown according to UTC.

Time Nick Message
12:25 SidGBF joined #crimsonfu
12:25 melodie joined #crimsonfu
12:25 melodie hi
12:25 melodie I have just been told about puphpet, to be used in a vagrant machine
12:25 melodie <SidGBF> melodie: the idea is that you create a VM and all your webserver (or whatever deployable) project you make inside of it. Database, apps, stuff like that, so you don't need to install in your machine. And with puphpet you can easilly deploy too.
12:25 melodie this is at #ubuntu
12:29 melodie https://puphpet.com/
12:33 pdurbin melodie: I use puppet and vagrant to demo our app: https://github.com/dvn/dvn-install-demo
12:34 melodie hi pdurbin !
12:34 pdurbin o/
12:35 melodie I does not look that hard, I should give it a try
12:35 pdurbin melodie: yes! please try it :)
12:35 melodie do you install puphpet in the vagrant machine? Is that the way it works?
12:35 melodie not yet, but at some time perhaps soon
12:35 pdurbin vagrant is already installed in the vagrant "base box"
12:35 pdurbin er
12:36 pdurbin puppet is already installed in the vagrant "base box" (sorry)
12:36 melodie I need to try openerp and have to install postgresql to make it work
12:36 melodie oh so
12:36 melodie puppet or puphpet ?
12:37 melodie and I would also need to have a lamp server independant from my current install, that would be handy
12:37 melodie do you think I could make a vagrant machine out of Bento?
12:37 melodie bento being a 12.04.4 (lts up to date) that could be handy for me
12:44 pdurbin heh. puppet
12:52 melodie joined #crimsonfu
12:52 melodie hi again
12:52 melodie I have rebooted to get the configuration of the usb headset work with alsa : that worked
12:53 melodie someone here is keen with udev rules?
12:54 melodie at boot I get this:
12:54 melodie udevd[625]: failed to execute '/lib/udev/mtp-probe' 'mtp-probe /sys/devices/pci0000:00/0000:00:1d.2/usb4/4-1 4 2': No such file or directory
12:58 melodie rebooting
13:16 chasmo77 joined #crimsonfu
13:20 melodie joined #crimsonfu
14:16 balo left #crimsonfu
14:46 pdurbin help me #crimsonfu! http://irclog.greptilian.com/javaee/2014-03-20#i_64690
15:45 sivoais pdurbin: if it is something they have to restart, why not check if those are done on startup?
15:48 pdurbin well, it's not really about restarting things...
15:48 pdurbin "what's the best way to communicate things to your team such as: 1. run this SQL script after you pull the latest code. 2. drop this new schema.xml file into place after you pull the latest code. 3. drop your database and create a fresh database after pulling the latest code. ?"
15:52 pdurbin does that make sense?
15:54 sivoais pdurbin: is there some sort of versioning stored somewhere in the configuration?
15:55 sivoais because it would make sense to compare against that to know when someone is running an upgrade
15:57 sivoais that combined with a new file (like UPGRADE.2014-03-20) and mailing list / blog post would make the most sense to me
15:59 pdurbin everything is in git
16:29 pdurbin sivoais: that's effectively what we're trying. a new file per breaking change. but we're trying it in Google Docs
16:29 pdurbin "we're experimenting with a folder in Google Drive. made a breaking change? create a doc. everyone has applied the fix locally? delete the doc"
16:33 sivoais ah, I saw that in the logs. Google Docs as a way to get feedback?
16:38 pdurbin no. as a way to communicate when you latest code requires action from the rest of the dev team to get their dev environments working again
16:44 pdurbin because of a breaking change.  need to run an sql script or whatever. or update your solr schema.xml
18:42 skay pdurbin: you could create tasks in your tracker that are subtasks of a containing task, and assign all the devs a task
18:43 skay pdurbin: I'm guessing you need a stopgap until you find some way to automate stuff?
18:44 skay I'm on a team that communicates information via google docs and it is driving me crazy
18:48 skay i used jira before. issues could block other issues
18:50 semiosis i was a big fan of google docs until a couple weeks ago when they abruptly deleted one of our collaborative docs for an unspecified TOS violation
18:51 semiosis our requests for review went unanswered
18:51 semiosis fuck google docs
18:51 semiosis the doc in question was a form with questions we use for QA testing our mobile apps
18:51 semiosis absolutely not a TOS violation
19:19 skay I like the art in the mediagoblin campaign that makes internet drones that remind me of google style elements http://mediagoblin.org/news/do-something-about-censorship.html
19:19 skay I think he could have done that with the big eye too
19:21 pdurbin skay: well, dropping your database completely always works. but yeah, the setup to get workign again is not completely automated yet
19:23 skay I wrote a script for myself called smashdb. hulk smash
19:23 skay dropdb! makedb! psql thingees!
19:23 skay haven't automated all of that yet
19:25 skay ... do you ever get asked to do things that are almost like painting bikesheds more more like being asked to put a screen door inside of a hatch in a submarine?
19:25 pdurbin um
19:25 skay that was apropos nothing
19:25 pdurbin :)
19:28 pdurbin yeah, i have a dropdb script too
19:29 pdurbin maybe this is just part of greenfield development. or agile. lots of things changing under you
19:29 pdurbin trying to build a house while the foundation was just poured and is still wet
19:44 skay I don't know how much you are changing going from 3 to 4, but I know at my old job we changed out the travel platfrom from behind classic Orbitz and it was quiet challenging at times
19:44 skay at first we had no good process and it was "death march"
19:45 skay later on people learned and got more sane and started trying to borrow methodologies from here and there
19:45 skay with varying degrees of success
19:46 pdurbin skay: did you know there's a book called "Death March" about software?
19:46 skay anyway, we ended up with multiple environments with different things and having to do some lock step stuff from time to time when making major revs
19:46 skay pdurbin: YES OMG
19:46 skay that's where  know the term from
19:46 skay I haven't read hte entire book yet
19:46 skay I tend to pick up non fiction books and read for a while and then pick up another and so on
19:47 skay as opposed to loosing myself in novels until I am done with them
19:47 pdurbin I discovered this recently on Google+: https://plus.google.com/+PhilipDurbin/posts/KmDWQrbvjm7
19:47 pdurbin this is the book: http://en.wikipedia.org/wiki/Death_march_%28project_management%29
19:47 pdurbin I haven't read it but I did find that wikipedia page interesting
19:48 skay pdurbin: I confess that I have imposter syndrome so horribly that I often day dream that I should just quit and get a different career
19:48 skay Or I day dream that one day I am independently wealthy and can go cocoon and do programming for fun and not at any job
19:49 skay I have imposter syndrome so horribly that I am even afraid of interviewing prospective employees. I was barely working up to that orbitz.
19:50 skay I was interesting in helping to interview people but was so nervous that I avoid it
19:52 skay nice post
19:53 skay pdurbin: more classical books for you: Mythical Man Month.
19:53 skay Peoplesoft
19:53 skay I really hate the Real Programmer meme
19:55 skay I grew up in an economically depressed region and I care fuck all if people want to program to make money versus programming because the get warm fuzzies
19:56 skay it's so incredibly pretentious to devalue the former
19:56 pdurbin I've heard good things about Mythical Man Month. I think that's where the "you can't get a baby in 1 month with 9 women" idea comes from
19:56 skay could be
19:58 semiosis close... i think the analogy would be that it takes *longer* than 9 months to get a baby with more than one woman
19:58 semiosis ;)
19:59 skay pdurbin: from your shared post " if they want to really be good they have to do more than just what it required in class. They need to have a side project. I truly believe that to be the case."
19:59 semiosis MMM thesis is that adding labor to a late project makes it later
19:59 skay here's my college experience -- blue collar parents didn't give me any money, so I went to school part time and worked part time. I spent outside time hanging out in a cognitive psychology lab
20:00 skay one of my cs profs asked me if I would consider working in his lab but I turned him down because I would have been making a lot less than my job in the IT department
20:01 skay real people do not always have the luxory to be real programmers -- maybe they have to spend time paying their way instead
20:01 skay at least the IT job involved me doing flunky stuff in operations and learning perl
20:01 skay I think I learned more useful things from that than from many of my classes
20:03 skay semiosis: it is hilarious when people who know about this book and admit it do still try to add more people to get a lte project going
20:03 skay late
20:05 skay pdurbin: did you see the reproducibility paper that this is in reaction to? http://cs.brown.edu/~sk/Memos/Examining-Reproducibility/
20:12 semiosis i got a lot out of the theory courses... how to reason about security protocols, modeling communications channels, queueing, etc
20:12 semiosis best programming course i took was AI in which I learned lisp
20:13 semiosis really learned lisp from a book
20:13 semiosis but for that course
20:13 semiosis imho just doing the required work for classes isn't enough
20:14 semiosis it's necessary but not sufficient to be a good/great developer
20:25 melodie joined #crimsonfu
20:25 pdurbin semiosis: should I learn clojure?
20:25 pdurbin skay: sorry, doing real work
20:25 skay scala
20:25 semiosis clojure!
20:25 semiosis no scala
20:26 skay erlang!
20:26 semiosis i leared common lisp
20:26 semiosis if clojure had existed 10 years ago i'd have done that instead of CL
20:26 pdurbin I don't know any lisp. but I was thinking about learning some clojure
20:26 semiosis definitely
20:27 semiosis i learned lisp but i never wrote practical programs in lisp... lisp changed how i write code in all languages, esp. java
20:28 semiosis i'm a much better dev (in all languages) for learning lisp
20:28 semiosis it forces you to think bottom up
20:28 semiosis which is quite different from most languages
20:28 pdurbin yeah
20:31 semiosis it's also nice how recursion is more natural than iteration in lisp
20:33 semiosis in any problem you can not solve, there is a simpler problem you also can not solve
20:38 pdurbin sing it
20:48 semiosis ♬ in any problem you can not solve, there is a simpler problem you also can not solve ♬
21:44 pdurbin_m joined #crimsonfu
21:45 pdurbin_m semiosis: lol. that was worth saving: https://plus.google.com/107770072576338242009/posts/N77RpSX3DER :)
21:46 semiosis ha
22:09 pdurbin finally getting some real answers back in ##javaee :)

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

crimsonfu - sysadmins who code