Adapting to the existing examples makes a bit more sense to me. If they're a contrib, I think they'd be better served as an externally maintained contrib. If we accept them into the project as an internal contrib, that kind of reflects a willingness and an obligation to maintain them, and that's less likely as an internal contrib than as part of the build examples or as an externally linked contrib, I think (from observing the state of the existing internal contrib repos).
-- Christopher L Tubbs II http://gravatar.com/ctubbsii On Wed, Nov 12, 2014 at 11:18 AM, Donald Miner <[email protected]> wrote: > Sean, > > My original thought here was that we could adapt these to fit the mold of > the Accumulo examples that are shipped with core. Does that make any sense? > > Either way, the contrib approach seems reasonable as well, just not what I > first thought. > > -d > > On Tue, Nov 11, 2014 at 12:18 PM, Sean Busbey <[email protected]> wrote: > > > I think these would be an excellent addition, given the improvements > David > > suggested. Since they span a range of Accumulo uses, they are probably > best > > served as a contrib repository[1] rather than an Accumulo > version-specific > > example. > > > > In any case this will involve adopting a non-trivial code base developed > > outside of the project; we'll need to go through the ASF Incubator[2]. > > Before that can happen we'll need to call a vote[3]. > > > > [1]: http://accumulo.apache.org/contrib.html > > [2]: http://incubator.apache.org/faq.html#proposed_new_codebase > > http://incubator.apache.org/ip-clearance/index.html > > [3]: http://accumulo.apache.org/bylaws.html#actions > > > > On Tue, Nov 11, 2014 at 10:38 AM, Josh Elser <[email protected]> > wrote: > > > > > I've given a quick glance over them -- they look like they'd be a great > > > addition! > > > > > > We'd have to figure out some mechanism to distribute the exercises (as > we > > > can't compile them), but that's a manageable problem. > > > > > > If you want to open an issue on JIRA, that'd be the first step to get > > > these into the codebase. Some things to think about meanwhile: > > > > > > * Check out the coding practices and code formatting guidelines - > > > http://accumulo.apache.org/source.html#coding-practices > > > * Add ASL headers to the files > > > * Figure out where might be a good place to include these in the > Accumulo > > > tree - maybe examples/training? > > > * Consider what documentation would be needed for someone to self-guide > > > themselves through these examples > > > * Look into redistribution rights on the included twitter.json file. > I'm > > > not sure what Twitter's terms of service are. It may be easier to > write a > > > script that will generate some example tweets. It keeps us from being > > > liable for what those tweets contain and also prevents us from having > to > > > distribute a big blob. > > > > > > Thanks again! > > > > > > > > > David Medinets wrote: > > > > > >> Can you add descriptions of the exercises to the README file? Many > > >> people finding that page would move to the next one with details to > > >> catch their interest. > > >> > > >> On Tue, Nov 11, 2014 at 9:49 AM, Chris Rigano< > [email protected]> > > >> wrote: > > >> > > >>> I believe they would be of benefit. > > >>> > > >>> On Mon, Nov 10, 2014 at 12:43 PM, Adam J. Shook<[email protected] > > > > >>> wrote: > > >>> > > >>> I had just finished finalizing the training materials for a basic > and > > >>>> advanced Accumulo class my company ClearEdge IT offers. I think > some > > of > > >>>> the more advanced code tutorials would be valuable to contribute to > > the > > >>>> Accumulo examples library. > > >>>> > > >>>> The examples all work with status updates from Twitter and include > > >>>> topics > > >>>> such as: > > >>>> > > >>>> - Basic Reading/Writing > > >>>> - Indexing tweets and creating a program to retrieve tweets > based > > on > > >>>> given search terms > > >>>> - Bulk ingestion of the tweets > > >>>> - Using MapReduce to building a geo-index table for the tweets > > with > > >>>> latitude/longitude information via z-points > > >>>> - Levering the geo-index to retrieve tweets from a given > lat/long > > >>>> bounding box > > >>>> - Custom iterators such as filters and combiners > > >>>> > > >>>> If this is something the community would be interested in, please > take > > >>>> the > > >>>> time to review them at > > https://github.com/adamjshook/accumulo-training > > >>>> and > > >>>> let me know if there are any you think would be worth contributing. > > >>>> I'd be > > >>>> happy to take the time to massage them to meet the standards. > > >>>> > > >>>> Cheers, > > >>>> --Adam > > >>>> > > >>>> > > > > > > -- > > Sean > > >
