Menu
▾
▴
Re: [JPP-Devel] Michaels aggregation plugin
|
From: Michaël M. <mic...@fr...> - 2009-03-11 19:06:16
|
Hi, > I think it may be better to include it into OJ 1.3.1 / 1.4 - because I > would like to avoid to add new translation strings to the core now. > That's OK. Let's see after 1.3 and let's focus on bugs and translation (you already did a great work in this direction!). Michaël > Only bug fixing should be done now. However, I don't even find now bugs > that are fast to do for me (except the selection one). So I personally > would wait for the translations only and do the OJ 1.3 release as soon > as possible. > > my 2cents - so if you think differently pls. tell me > > stefan > > Michaël Michaud wrote: > >> Hi, >> >> I made some improvements in v 0.2 >> (http://geo.michaelm.free.fr/OpenJUMP/resources/) >> I added minority, majority, median, and tried to improve the user >> interface by adding some icons and examples. >> I also changed its place, putting it in the tools menu, along with the >> other. >> I'm still not in hurry to include it, but if you say it's time to do >> it, I can help. I saw you took much time to fix bugs add plugins, >> translations and so many improvements. Nice you could add undo to some >> plugins, I'm sure many users will thank you for this addition (I never >> used undo capabilities of OJ code, but I have to, I add it on my todo >> list ;-) >> >> Michaël >> >> >> >> >>> Hei Michael, >>> >>> a long answer >>> >>> I) Minority/Majority/Median: >>> they are only applicable to integer or string values - end return the >>> value that occurs the most = majority, or the least = minority* (i.e. by >>> counting the occurence). The median is the value right in the middle if >>> we build an ordered set of all values (makes sense for integer values, >>> less sense for double). >>> >>> *) mmhm.. I should check what I actually do when there are several >>> values that are only 1 times in >>> >>> II) "query" message >>> I think it is best to schedule it for a new release... >>> so no need to hury >>> >>> III) "do you think that i) count features with non null values, ii) >>> concatenate all values, iii) concatenate distinct values - would be >>> clearer for aggregation function names" >>> >>> not sure what you refer to by "function names"... ??? >>> >>> IV) Lecture available >>> Unfortunately not - I have all the material but it is build for a >>> complete course, i.e. 13x 1.5h hours. So definitively not working for a >>> 2-3h tutorial >>> >>> V) giving a tutorial >>> well.. It would require quite some time to prepare the stuff. However, >>> I/we should look into Ravis Mock-Excersise. Which he is using for >>> Geologist training. Maybe that would be an option? >>> >>> >>> I am actually still pondering if I submit something. For instance: >>> >>> a) an article or abstract on OpenJUMP (an evaluation) ... Btw. do you >>> know what actually the "Show Case" category means? Would fit a general >>> presentation on OpenJUMP into that? >>> >>> b) or: an abstract for a talk on landscape pattern detection using >>> buffer ops instead of raster morphology (but ... this is just an idea, >>> and I am not sure if it makes sense. Because in the end I need to >>> compare buffer and raster ops - which is not sooo easy) >>> >>> c) or: an abstract for a talk on "analysing the requirements to obtain >>> geographical data from image segmentation objects" >>> >>> ... the latter two I was thinking about, since I hope we can push the >>> conference a bit towards science, i.e. away from FOSS4G which is a >>> mixture of business and science. >>> But the problem is - for all talks I need time to prepare and I rarely >>> have any research done on this (well on the latter I already have a >>> presentation - but I should show some nice things). >>> Nacho plans actually to submit his evaluation of JUMP forking. He wrote >>> about this last year for a meeting, but it is all in spannish... so I >>> don't really know by now what his conclusions are. >>> >>> stefan >>> >>> Michaël Michaud schrieb: >>> >>> >>>> Stefan Steiniger a écrit : >>>> >>>> >>>>> Hei Michael and others, >>>>> >>>>> I had a look on your aggregation plugin. It is a bit more advanced than >>>>> my one. If we would add Minority/Majority/Median and try to get the >>>>> "query" message fixed so that it doesn't exceed the box, >>>>> >>>>> >>>> Hey, I did not notice that. I'll have a look to see if it can be fixed >>>> easily. >>>> I understand median, but I'm not sure what minority and majority means. >>>> Should we authorize some of these functions for strings ? >>>> >>>> >>>>> then I think we >>>>> rather should use your plugin. Or we add both, calling my plugin "Simple >>>>> Attribute Aggregation..." and your "Advanced Attribute Aggregation...". >>>>> Of course some of the options need to be explained a bit (i.e. what is >>>>> Count(non-null) and what is Concatenate-All and Concatenate-Distinct). >>>>> >>>>> >>>>> >>>> I think we should try to keep only one plugin, except if the one with >>>> more capabilities is much more difficult to use. >>>> It seems that my aggregation plugin needs some enhancements to be really >>>> user-friendly (usable without the need of a user-guide). >>>> One of my ideas was to build a dynamic illustration of the result based >>>> on micro generic dataset (but it would need some more work...) >>>> Any idea is welcome to improve use experience >>>> Do you think that >>>> - count features with non null values >>>> - concatenate all values >>>> - concatenate distinct values >>>> would be clearer for aggregation function names ? (a bit long, but long >>>> is better than not understandable) >>>> >>>> >>>>> However, I have been looking for the source code and it is not included >>>>> in the downloadable jar. >>>>> >>>>> >>>>> >>>> Peppe answered for me, thanks Peppe >>>> >>>> >>>>> What is your opinion shall we integrate it or not? If yes - do you think >>>>> we should add that for 1.3 or into a future 1.3.X / 1.4 release? >>>>> >>>>> >>>>> >>>> I think there is no hurry. >>>> I'll have a look next week end to see how much work it would need to >>>> improve it. >>>> If we are not fully satisfied, we can let it as a plugin and wait for a >>>> future release. >>>> >>>> >>>>> cheers from Calgary >>>>> stefan >>>>> >>>>> PS: I did a bit of experimenting with Sextante (the current version that >>>>> works only with the Pirol Image lib). I used Sextante to vectorize a >>>>> binary raster to obtain polygons. After applying on the resulting >>>>> vectors OJ's Planar Graph function I got a pretty nice polygon cover. :) >>>>> I now need to check if one could do with OJ the complete lecture we did >>>>> in Zurich with ArcGIS. >>>>> >>>>> >>>>> >>>> Would be great (is this lecture available ?) >>>> Erwan asked me if we considered preparing a lab on OpenJUMP for OGRS >>>> spatial 2009. What do you think ? >>>> >>>> Michaël >>>> >>>> >>> ------------------------------------------------------------------------------ >>> Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA >>> -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise >>> -Strategies to boost innovation and cut costs with open source participation >>> -Receive a $600 discount off the registration fee with the source code: SFAD >>> http://p.sf.net/sfu/XcvMzF8H >>> _______________________________________________ >>> Jump-pilot-devel mailing list >>> Jum...@li... >>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >>> >>> >>> >>> >> ------------------------------------------------------------------------------ >> Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA >> -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise >> -Strategies to boost innovation and cut costs with open source participation >> -Receive a $600 discount off the registration fee with the source code: SFAD >> http://p.sf.net/sfu/XcvMzF8H >> _______________________________________________ >> Jump-pilot-devel mailing list >> Jum...@li... >> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >> >> >> > > ------------------------------------------------------------------------------ > Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are > powering Web 2.0 with engaging, cross-platform capabilities. Quickly and > easily build your RIAs with Flex Builder, the Eclipse(TM)based development > software that enables intelligent coding and step-through debugging. > Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com > _______________________________________________ > Jump-pilot-devel mailing list > Jum...@li... > https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel > > > |