>>No documentation, almost no javadoc, extremely sparse tests, quite a few bugs, some funky code / left over unused code. Some things are now better, but I don't think great. Part of the nature of open source as well, but irksome non the less. =====
That is true, and I do accept responsibility for it, I have had to step away from open source over the past year to focus on significantly more important personal things. The most important goal I feel isn't perfection, but a basis that can be improved up, I will be honest that if spatial lucene had not been in lucene/contrib, I doubt that it would be alive today. It exists because others have helped. (Which is why I'm as irked as I am about the original subject of this thread) Consolidation of documentation I found to always be a problem, which is why I put www.gissearch.com together a few years ago to assist with it while it was in Sourceforge, again I haven't had opportunities to update and move that over to Apache's Wiki. To me open source is not run as a corporate waterfall delivery, where the initial delivery is the end goal, but a mere start, that other engineers can contribute significantly to. If I wasn't a firm believer in it, I would not be pushing as hard as I am to ensure it is added to Solr On Thu, Dec 3, 2009 at 11:30 AM, Mark Miller <markrmil...@gmail.com> wrote: > patrick o'leary wrote: > > Think we crossed lines somewhere on the first part of the discussion. > > > > > > > >>> But your doing that yourself at source forge? Hasn't there been a lot > of > >>> > > work on an external LocalLucene, even after it was put into contrib? > > While the contrib version was left in a fairly hairy state? > > Thats just the nature of the license - but putting LocalLucene into > > contrib hasn't appeared to help much. > > ==== > > > > I disagree Mark, locallucene hasn't been updated in 8 month on source > forge > > http://locallucene.svn.sourceforge.net/viewvc/locallucene/trunk/ > > locallucene/< > http://locallucene.svn.sourceforge.net/viewvc/locallucene/trunk/locallucene/ > > > > *168*< > http://locallucene.svn.sourceforge.net/viewvc/locallucene/trunk/locallucene/?view=log > > > > 8 > > months pjaol Added getQuery(Query) method to convert distance filter to > a > > query allowing loca... > > Only localsolr has had work performed on it,while waiting to get > something > > in Solr, spatial lucene has been slowly updated over time, by more than > just > > I > > which is what open source and iteration is all about. > > If you want to wait for perfection, you have to wait. > > > Okay, fair enough - I was going by the front page stats that mention a > number of recent commits and hearsay that was told to me by others (or > other ;) ). > To be fair, I haven't looked at the code, so I'm happy to take your word > for it. My apologies. > > > As for leaving spatial contribution in a hairy state, you care to > clarify? > > > No documentation, almost no javadoc, extremely sparse tests, quite a few > bugs, some funky code / left over unused code. > > There is a history of it on the Lucene-dev list - not that its that big > of a deal, but kind of irksome that those that put it in haven't > responded to any of the issues, and others that are > less familiar have had to somewhat pick up the torch. A couple people > were made contrib committers with the hope that they would help with the > upkeep and fixes - otherwise an existing > committer could have just dumped it all in. I don't really mean that as > negative as it prob comes off though. > > Some things are now better, but I don't think great. Part of the nature > of open source as well, but irksome non the less. > > > > On Thu, Dec 3, 2009 at 10:53 AM, Mark Miller <markrmil...@gmail.com> > wrote: > > > > > >> patrick o'leary wrote: > >> > >>> Someone pulling an Al Gore (inventing the internet) on this isn't my > >>> concern, heck you can just google for some of the class names of > >>> locallucene and see how far spread it is, > >>> > >> Then whats this about: > >> > >> " > >> > >> but it's giving significant, 'impression of ownership' of a lot of work > >> that's been completed > >> by other folks." > >> > >> > >>> what I am more concerned about > >>> > >>> "Future versions of these patches may include support for search with > >>> regular polygons, and the introduction of distance facets, allowing > Solr > >>> users to be able to filter their results based on the calculated > >>> > >> distances." > >> > >>> They're now 'flogging' recent and current work I and others are doing? > >>> > >>> ... not encouraging, and certainly not healthy for open source. > >>> > >>> > >> Doesn't sound that way to me. > >> > >>> I'm going to be brash and request that there is commitment to adding a > >>> > >> basic > >> > >>> Spatial feature set for distance searching (restricted by distance) & > >>> sorting > >>> to Solr's trunk by the end of December. Iterate and refactor as needed > >>> > >> after > >> > >>> that. > >>> > >>> There should not be any more excuses to having this code out in the > cold > >>> > >> as > >> > >>> patches and external projects. > >>> > >>> > >> But your doing that yourself at source forge? Hasn't there been a lot of > >> work on an external LocalLucene, even after it was put into contrib? > >> While the > >> contrib version was left in a fairly hairy state? > >> > >> Thats just the nature of the license - but putting LocalLucene into > >> contrib hasn't appeared to help much. > >> > >>> On Thu, Dec 3, 2009 at 8:48 AM, Mark Miller <markrmil...@gmail.com> > >>> > >> wrote: > >> > >>> > >>>> Yonik Seeley wrote: > >>>> > >>>> > >>>>> On Thu, Dec 3, 2009 at 11:22 AM, patrick o'leary <pj...@pjaol.com> > >>>>> > >>>>> > >>>> wrote: > >>>> > >>>> > >>>>>> What spatial contributions have been contributed from this? > >>>>>> I'm only seeing some query parsing / multi-threading extensions, no > >>>>>> > >>>>>> > >>>> shapes / > >>>> > >>>> > >>>>>> SRID's etc > >>>>>> but it's giving significant, 'impression of ownership' of a lot of > >>>>>> > >> work > >> > >>>>>> that's been completed > >>>>>> by other folks. > >>>>>> > >>>>>> > >>>>>> > >>>>> Looks like they acknowledge building on local solr and local lucene > to > >>>>> > >>>>> > >>>> me: > >>>> > >>>> > >>>>> """SSP started out its life as a patch for Solr Spatial Search > >>>>> (Solr-773) and Spatial Lucene (Lucene-1732) and extends Solr and > >>>>> Lucene with hereunto missing geodetic search functions (bounding > boxes > >>>>> etc) while improving on the speed of the result and performance when > >>>>> dealing with a large data set through better query parsing and > >>>>> multi-threaded filtering. Also included are improved extensibility > and > >>>>> documentation.""" > >>>>> > >>>>> And in a way, they do "own" their plugin - their customizations, > >>>>> packaging, etc (note: I haven't looked at it). And they offer > support > >>>>> for it - which might be attractive to some companies that need > >>>>> supported geosearch now. > >>>>> > >>>>> It's also open source under the Apache license, so presumably we > could > >>>>> borrow anything we want from it. > >>>>> > >>>>> -Yonik > >>>>> http://www.lucidimagination.com > >>>>> > >>>>> > >>>>> > >>>> I think Patrick is obviously referring to: However, in the last 6 > months > >>>> support for spatial search has begun to be added to Apache Lucene and > >>>> Solr, much of which has been developed here at JTeam. > >>>> > >>>> "Much of which" is obviously a bit of an overstatement (to a great > >>>> degree or extent) when you look at all the work thats been done. > >>>> > >>>> Oh well though. So it goes. Its Apache - they could package it all up, > >>>> hide the code under the covers, put a notice saying some work was > >>>> derived from Solr, call it Solr: geo search edition, and essentially > >>>> take even more credit while adding little to nothing. I wouldn't sweat > >>>> > >> it. > >> > >>>> > >>> > >> > > > > > > > -- > - Mark > > http://www.lucidimagination.com > > > >
