Comments for mwh.geek.nz

Comment on Election visualisation by Michael

Michael — Sun, 27 Nov 2011 06:07:00 +0000

I have a Python script that parses the data out of (TSV files created from) the files from http://electionresults.org.nz/electionresults_2011/pp_stats_index.html and geocodes the addresses, then generates the appropriate data code to go in the page. It takes a few minutes to add each electorate.

There wasn’t anything there because there’s no street data for the Chathams, so geocoding failed. I’ve added it manually. It’s tiny, but there.

Comment on Election visualisation by Walter

Walter — Sun, 27 Nov 2011 05:43:39 +0000

Why are there no polling places on Chatham Island in Rongotai?

Also, do you manually add in each polling place?

Comment on Election visualisation by Michael

Michael — Sun, 27 Nov 2011 01:54:22 +0000

And since that last comment I have updated it to include the top five parties at each place as well.

I tried reversing the circles; it definitely isn’t clearer, for reasons that should have been obvious. I’ve also tried including an “all other parties” circle, but that muddles things more than anything.

Comment on Election visualisation by Michael

Michael — Sun, 27 Nov 2011 01:52:54 +0000

I’ve updated the generation to include (almost) all the places – I had tried to add the Northland ones manually, but they clearly didn’t take.

On circles, yes, they are suboptimal. I really wanted a pie chart, but I’m limited by what the API provides. Polygons require lat-long coordinates as axes, which aren’t really much good for graphics, and drawing my own circles for a pie has the same issue. I do think that the circles are acceptably understandable, at least, though I am wondering if they should be in the opposite order.

The code is there to be edited if you want, though…

Comment on Election visualisation by Walter

Walter — Sun, 27 Nov 2011 01:00:03 +0000

This is nice. However, you seem to be missing some polling places.
For example, there are two in Northland, neither of which show up on your graphic.
Also, the graphic for “Newtown, St James’ Presbyterian Church Lounge, 235 Adelaide Road” is a bit misleading perhaps, as four parties are tied in third with one vote, yet it only shows one of them.

I think it’s hard to judge relative areas of circular shells, perhaps a bar graph would be easier?

Comment on LCA video by Michael

Michael — Sat, 02 Jul 2011 03:27:37 +0000

It was recorded, but they lost it somewhere along the line – one of the videos from the Distro Summit is missing, and it’s mine.

Some of the video archives have/had one labelled as it, but it’s actually the unstructured discussion in the following slot from when the speaker didn’t turn up. So it seems to be missing permanently.

Comment on LCA video by trans

trans — Sat, 02 Jul 2011 01:25:50 +0000

/System/Aliens talk please!

Comment on Python Damerau-Levenshtein distance implementation by Lucid Imagination » Creating a spellchecker with Solr

Lucid Imagination » Creating a spellchecker with Solr — Fri, 15 Apr 2011 22:27:09 +0000

[...] the previous listing I have to make some remarks. In line 2 and 3 we use third party libraries for Levenshtein distance and metaphone algorithms. In line 8 we are collecting a list of 70 candidates. This particular [...]

Comment on Python Damerau-Levenshtein distance implementation by emmaespina

emmaespina — Mon, 31 Jan 2011 15:59:08 +0000

[...] the previus listing I have to make some remarks. In line 2 and 3 we use third party libraries for levenshtain distance and metaphone algorithms. In line 8 we are collecting a list of 70 candidates. This particular [...]

Comment on Python Damerau-Levenshtein distance implementation by Michael

Michael — Thu, 20 Jan 2011 21:47:23 +0000

Thanks, OneEyedMan. Those changes should be all that’s required, and they’re what 2to3 will give you if you’re converting automatically. The Python 3 version should run fine on Python 2 as well, with just a little waste around the range() calls.