Smathermather's Weblog

Remote Sensing, GIS, Ecology, and Oddball Techniques

Archive for October, 2013

Favorite Slide, #cfasummit

Posted by smathermather on October 22, 2013

So, if I had to choose one iconic slide from the CfASummit (and I do as I really need to catch up on sleep, so long posts will wait until the end of the week) this would be it presented by our illustrious MC (David Eaves) for the event:


The gist of the slide is that CfA has gone from the simple website/front end design overhaul within governance to addressing rules, processes, and culture, and is aiming for the root of the government stack, processes of procurement and their impact on tech in government, and establishing and “open by default” culture for governance.

There are three things I like about this diagram:  the analogy of government and software stack, the idea that these changes are part of a culture of “play” (which by the way is a wonderful encapsulation for the process of experimentation and learning taking place), and finally that the plan is to address through thought leadership and practical solutions underlying issues of procurement and culture that are at the root of the disconnect between government and technology.

Speaking of government, technology, and procurement, it’s refreshing how on-point the media that I have caught on is.  Nice that this has moved from a niche issue to something that is being commonly discussed.

Finally, a quick link to my only (and humble and nacent) post on open government:

Posted in CfASummit, Conferences, OpenGov | Tagged: , , | Leave a Comment »

Building something really useful #CfASummit

Posted by smathermather on October 20, 2013

I just came back from the CfASummit, 2013 and thoroughly enjoyed it.  I met a whole bunch of really dedicated, intelligent, and idealistic people who have been blessed with a position to change the world, the presentations made me laugh and cry (really).  CfA represents many the changes I have been hoping for in the public sector since I entered it 5 years ago, and many I hadn’t thought about.  And the Kool-Aid was amazing.

I have a few really good posts regarding these experiences in the queue that I haven’t finished yet, but just saw a post by Steve Citron-Pousty called “Hey Civic Hackers! How about leaving the ninja skills at home and building really useful applications?“.  Steve had 4 core points:

  1. At the very least, the only programming languages you should use on the server are (in decreasing order of priority) PHP, Java, or Python.
  2. Next up, most civic hacks should use a popular framework.
  3. Document your project more than you think you need to document it.
  4. Finally, in truth, most civic hacks should really make a plugin for a platform like WordPress, Drupal, Django, or Magnolia.

The core idea being civic hackers should be providing tools that can be changed and adopted by the municipality.  Responses on twitter have varied, with a nuanced response from @spara:

Screen shot 2013-10-20 at 10.54.20 AM

Screen shot 2013-10-20 at 10.50.20 AM


So, I want to be nuanced too.  I have a lot of affection and respect for the CfA Fellows who serve our neighbors and partners in Summit County.  They have built a simple and beautiful interface for finding trails in the park systema.  I celebrate what they built, and have enjoyed watching this product develop.

But, one frustration I have had with the process was the balkanization that resulted from the project–here all partners (I am not employed by a partner, but I’m a trouble-maker with the Brigade) involved were hoping for some regional product from the work of the Fellows.  There was already locally commissioned work complete built on (gasp) PHP/Codeignitor application with similar intents.

The Fellows application is compatible with the existing application in no single way– not employing a similar back end, not sharing any front end code, not even sharing API compatibilities.

So let’s return to Sophia’s comment on twitter above: “We are there to start process of changing everything rather than bolting on to entrenched process and infrastructure.”  This, I am afraid, right or wrong, is the language of colonization.


So where does this leave us regionally?  Now we have two related applications, which share no heritage, front end, back end code, nor API, and an expectation from local leadership to avoid the mistakes of the past of balkanization, those very same entrenched processes that CfA aims to undo.

So, when civic hackers enter a space and see change for the good happening, is it better to build that rails race car that Steve’s post points to, or should we be, in addition to changing the existing civic tech space also be celebrating and enhancing good things that are in progress?


From a practical standpoint, how do we put the pieces back together?  I want to emphasize again, the great and meaningful work work the Fellows did.  But they will leave us in a few weeks with a real challenge.  Thoughts?

Code for each project can be found (naturally) on GitHub.

Posted in CfASummit | Tagged: , | Leave a Comment »

North American Cartographic Information Society (NACIS) Conference (yay #NACIS!)

Posted by smathermather on October 14, 2013

In a short blog post, I won’t be able to do the NACIS conference justice, but if you haven’t gone and you are a map geek, then I recommend you attend next year’s conference in Pittsburg.  First:

The People:

What a collegial and warm group of people.  NACIS was a very welcoming community, an interesting mix of private industry geniuses (ahem, Mapbox, Stamen, Vizzuality etc.), academics, students, National Geographic cartographers, and typically some serious Federal representation (although largely absent this year, a notable exception being Mamata Akella from NPSMaps who is technically not directly employed by the Feds and so was allowed to attend anyway).  As someone who spent the better part of a decade in the academic sector, it was fascinating (and comforting) to be back among (largely) introverts.  I would, however, argue the Nat Geo Cartographers were generally an exception to the introversion, but really pleasant and interesting folks as well.

Also, Andrew Hill from Vizzuality was there, so I picked his brain on features coming down the pike (including better Torque support, including a GUI soon), and bugged him to add pgRouting to the back end of CartoDB… .  He seemed receptive to this, and suggested following up under CartoDB support.


All the presentations were great.  From a very tech-practical standpoint, I enjoyed Carl Sack’s presentation on D3.  It was a great intro to D3js, and really got me over the basic barrier to using D3– getting the data in.  The TL;DR– order matters in the use of the API, queue.js is mighty useful, and all the data manipulation needs to go in the callback function to ensure your D3 goodness isn’t sunk by asynchronous execution.

The Maps:

Oh, boy there were some nice maps in the map gallery.  One trail map was an excellent theft of Swiss cartographic techniques, all the maps were interesting and well polished, and it was fun to look around at the diversity of approaches and topics.  Two maps of note that I really enjoyed– one is a map/infographic of wool exports from New Zeland, knitted in wool on a maker space constructed knitting machine:

Wool map

Also a really nice alternative piece was a bathymetry map by Carolyn Rose, which can be viewed at her blog:


My presentation went well.  There were lots of questions about a project we’ve been working on for our public web interface to help people find trails, parks, picnic areas and other parky amenities.  We just posted the code for the project to GitHub ( and I’ll have my slides posted from the presentation shortly.  It was really great to connect with others working and starting to work in similar spaces of parks and recreation mapping.  I am already enjoying the follow-up and hope this builds in to a large spatial/web community servicing this sector.


Posted in Cartography, Conference, NACIS | Tagged: , | Leave a Comment »

Pirating APIs

Posted by smathermather on October 14, 2013

Took me a while to post this, but Pete Warden DM’d me on twitter after this post: about geocoding with the following link:

It’s funny, I hadn’t given serious thought to whether this use of Google geocoding was a violation of the terms of service, but he makes some great points about why not only does it violate those terms, but why as advocates of open solutions, we shouldn’t use these solutions.  Go ahead, read it for yourself.  It’s a nice post:

Posted in Geocode | Tagged: | Leave a Comment »

Geocoding in QGIS, the easy way.

Posted by smathermather on October 1, 2013

Search the Google for geocoding in QGIS and you find a few nice articles, mostly pointing to third party web sites that will geocode for you, e.g. this post from 2009.  It’s a good post, but there’s an easier way now.

Well, let’s see if a little SEO will help solve this fgeocodeor people– a great tool for geocoding in QGIS is MMQGIS.  Add it as an extension, choose “GeoCode from Google”, and load up your CSV of addresses.  You may have to sit and wait a while, but rest assured, the result will be worth the wait.  So, load up the extensions manager and search for MMGIS.  Install.  Enjoy.

Now, off to find census data during the sequester… .


Posted in Geocode, QGIS | Tagged: , | 8 Comments »