Fork me on GitHub

@tilak, @barton, @jstokes: Hi, just joined clojurians on slack and learned about your course. Would like to here about any comments, because I'm the one behind sparkling.


@chrisbetz: just started using it. thanks for great library. I'm using this opportunity to learn both Spark & Clojure. My comments/questions so far - parallelize fn does not work with set and map that makes sense as they are not List, they need to be explicitly seq'ed, is there a reason why you choose not to have that conversion logic in parallelize fn? It's easy for beginner to think that all Clojure data structures can be parallelized. And same for parallelize-pairs, does not work with map. Some trivial or not so trivial wrapper fns are missing and I’m planning to add them for Clojure practice -, let’s see if I can manage to try all the exercies & labs with Sparkling.


@chrisbetz: i’ve only played around with a repl in sparkling, but first impressions were great. i’d love to see this carried forward with support for dataframes/mllib that were a part of 1.3/1.4 - then i would be able to make a compelling argument for use at work. i’d be willing to contribute to that cause, i just need to get more up to speed on spark/clojure first


@tilak: Good idea, to spread parallelize to other collections. Will have to check.


@jstokes: Yes, dataframes are the next thing. But I'm a little busy bringing my Spark ETL process to production right now. Just follow me on github, I'll keep you posted.


great, let me know if i can help out simple_smile


@jstokes: good to know. I'll come back to you simple_smile


@chrisbetz: @jstokes great, not an clojure expert, but if there is something at beginner level to contribute, i’ll be happy to contribute.


@tilak: thanks for offering. simple_smile I need to check, I'm pretty sure there are things...


wouldn’t implicitly calling seq defeat the performance gains of parallelizing? You still have to visit every element in order to create the list


how would you describe leiningen in one sentence?


A Clojure build tool and dependency management system with all the dexterity of a 50x50x50' piece of granite.


that’s 3.68e6 grams or 4.06 tons


Proof that the Clojure community comes together to help.


hey I’m waiting for hadoop.. may as well help out right?


Heck yeah.

fogus (Clojure Team)20:06:18

How does one invite or request to invite someone?


Oh I guess that’s only admins.


@fogus just send the user to if they enter their e-mail they will get an invite right away


[takes credit anyway]


@akiva: that link is only shown for admins though 😉


Yeah, I see that now.

fogus (Clojure Team)20:06:29

BTW, the Call for Presentations for CUFP 2015 is ending in 2 days. More Clojurians needed!

fogus (Clojure Team)20:06:40

Otherwise they'll need to change the conference to "The Galois Show"


hmm 2 days to come up with a talk though!

fogus (Clojure Team)20:06:07

You don't have a talk available on a moment's notice?!


Well I have an idea “The Immutable Law firm - changing the legal landscape with Clojure and Datomic”


Ooh, Tony’s on the committee… tempting to submit something very provocative about dynamic languages being better than statically typed languages… 😄