James Fee GIS Blog
I could drop a link to weather.gov1 and say living in Phoenix is simply wonderful. I could mention that we’re hiring developers with PostGIS experience2. I could mention we’re working in Node.js3. I could mention we’re working only in Leaflet.js4. Actually I could mention a lot of things but the bottom line is if you want to work with PostGIS/Node.js/Leaflet on projects around the world and enjoy 85 degree days in February, apply here.
I’m looking for entry level developers and those with more experience. Apply away if you want to work in Phoenix with PostGIS/Node.js/Leaflet.
Thanks to Dale and Don of Safe Software for joining me to talk about all the great new features of FME 2014, Safe’s new FME Cloud and the general state of file formats. The archived hangout is below.
Paul Ramsey sums up the situation very well:
Rather than avoiding a lengthy LIDAR format war, we are now entering one. In some respects, this will be healthy: the open LAS community now has to come up to feature parity faster than it might otherwise. But in most ways, it’s unhealthy: users will have data interchange issues, they’ll have to understand and install format translation software, and add extra steps to their processing chains.
Yuck right? LAS is still niche so it isn’t like FGDB where you have to convert it to old shapefiles to make it useful but working outside the community is not good for users. I’m glad I don’t work for a data marketplace anymore, these file formats are springing up like weeds1.
As a user, I don’t leave LIDAR data in LAS but convert it into other formats to use it. But it’s that interchange issue that keeps us stuck with old formats such as the shapefile. Sharing LAS is difficult to to huge file sizes. Binary point clouds with some sort of compression makes complete sense. Now you’ve got multiple file types to deal with. Enjoy…
Hard to keep track of them all↩
Well we’re back at it in 2014. The first hangout is with Dale and Don of Safe Software. FME 2014 is out and Dale and Don will talk about FME, formats, APIs, file types and anything else that interests us. We go live Thursday at 10am PST right here. You can also RSVP for updates.
If there is one area “professional GIS” has failed it is in the mobile arena. Crazy Windows CE, Java and other solutions just confuse and frustrate users. Heck after coming back into the GIS consulting world I’ve picked up these handheld ArcPad GPS units and failed to be able to get them to work1. There are some great Smartphone/Tablet solutions such as my favorite Fulcrum but they really fail on battery life2.
I’m always on the look out for better solutions to solve mobile GIS and the latest seems to be Windows Surface devices running Windows 8. I’ve been getting a lot of requests internally to test the devices for data collection. Most of it comes from the wish that users can run ArcGIS Desktop in the field. We’ve been fighting this on the mobile side for years, but maybe we should just sit back and let them have their day with a hacked up Surface Pro 2 with USB GPS and a checked out ArcGIS Desktop Basic3 and be done with.
Then again, what about standardizing on a PostGIS/QGIS field tool? This solves a couple of issues for me including the licensing implications of having floating licenses in the field for days at a time. I’m personally trying to reduce software licensing costs to a practical level and the unknown of who will check out extensions throws a wrench in it. The beauty of a PostGIS/QGIS solution is in the freedom to send people in the field for data collection and not have licensing bite us in the rear. I’m going to try to secure a Surface Pro 2 test-bed and see what such a PostGIS/QGIS field collection tool can do.
Plus once they get back into the office, sync the PostGIS data up and the GIS analysts can use it with their ArcGIS Desktop projects. Win/win, right?
So antiquated, so limiting, so dangerous…
Yet so important to our daily workflows. It’s the pinch-point on every project. Every year we make statements saying this is the year the shapefile goes away. Yet here in 2014 I’m dealing with the limitations of DBF 1 yet again. Shapefile, you drive me nuts yet I can’t quit you just yet. Here is to another “wonderful” year of .shp, .shx, .shp, .prj, .sbn, .sbx, .fbn, .fbx, .ain, .aih, .ixs, .mxs, .atx, .cpg and of course .shp.xml.
I know so much about DBF reserve words, sadly…↩
- Programming ArcGIS 10.1 with Python Cookbook
- Learning Geospatial Analysis with Python
- Learning QGIS 2.0
- Python Geospatial Development – Second Edition
- GeoServer Beginner’s Guide
The sale ends January 3rd so you have some time.
Update: Well here you go…
Sorry to disappoint everyone but contrary to current reports @BroadMap has not been purchased by Apple.
— BroadMap (@BroadMap) December 23, 2013
We had heard months ago that some BroadMap staff had moved on to Apple but today things have “exploded”.
Based on evidence and chatter from sources, Apple seemingly acquired mapping firm BroadMap in the first half of this year…
It’s complicated what has actually happened. Many people are claiming different things, but from what I’ve been told:
- Apple bought the intellectual property of BroadMap and licensed it back to them.
- Management of BroadMap and many key staff have joined Apple. I’ve gotten many different explanations as to “the how” of this but they’re at Apple
- BroadMap is still servicing existing clients. They may be doing this with “legacy staff” or have been using a 3rd party for support.
BroadMap has tweeted that they’re still around as well.
— BroadMap (@BroadMap) December 23, 2013
Given the uncertantity of it all, I’m guessing there isn’t much marketing staff left at BroadMap to deal with all these questions. Thus the ambiguity of it all.
Now what I’m interested in is Apple now has many old GDT1 staff on board. I’m assuming they’re going to be working to replace TomTom with their own mapping data. That’s more interesting that who owns what shell company anymore.
Which was acquired by TeleAtlas and then TomTom↩
With Amazon Web Services, I had a backed-up and replication-ready database up and running in under an hour, and OpenGeo Suite tied into it in another hour. With the knowledge I’ve gained, next time will take only a few minutes.
Think about that for a moment, a few minutes and you can deploy an enterprise GIS solution. Life is easier when you don’t have licensing constraints.
So over coffee I read this:
Apple’s “Interactive Map” patent filing details a mapping program that enables users to dynamically adjust and view different “layers” of content pulled from the Internet. Examples include commuting, tourism and weather map layers, among others.
Layers of content, brilliant! Why did we not think of this before? The patent filing is a road map for GIS. But I see huge problems with their method, look at this:
Where is the north arrow? Not only that Apple’s map data for Washington DC is cartoonish. No wonder people hate Apple Maps. I can tell you the maps of Phoenix look much better. Looking over the other supporting images is depressing. Breaking down what I do into flow charts, gawd we suck.
I look forward to licensing Apple’s patent to do my work.
Look, I got suckered into thinking BIM was the future. The painful fact is BIM is only useful to talk about, not actual use. But that won’t stop CAD companies from trying to convince you BIM is the future.
Let’s just say that BIM is the new GIS. In speaking with Mike DeLacey, CEO of Microdesk, a long time Autodesk partner and reseller, the terms are interchangeable in an Autodesk environment.
Will all do respect Mike, Neogeography (are we spelling it with our without CamelCase these days) is the new GIS. If BIM is interchangeable with GIS in Autodesk, I fear for Autodesk users. Given DWG is all about 0,0 (seriously though, how about require a projection Autodesk?) and unless we’re mapping null island, 0,0 is about as far from GIS as you can get.
- What a glorious time we live in. Track Santa “officially”!
- If academics put dots on a map, it better be for pool locations.
- Nokia’s HERE is Honda’s choice for in car mapping. ”Eyes free”? That that’s the thrill out of texting while driving.
- Google Maps API vs OpenLayers 3: Is OL as awesome as Leaflet.js?
- PostGIS 2.1 documentation as ePub.
- The Wolfram Language and Mathematica on Raspberry Pi, for free. Title says it all…
Mapmeter (despite not being CamelCase) is a pretty awesome way to get analytics for your maps. We’ve talked on my hangouts about how it is not free, nor is it open source. But you can get a copy free until the end of the year:
…we’re including a free annual Mapmeter subscription with every OpenGeo Suite purchase made before the end of the year. While a Mapmeter subscription is already included at the Platform level and above, this is the perfect opportunity for prospective Plus and Professional purchasers to save nearly $2400.
I’ve been keeping this archive for years. Each one of these has been created with that version of ArcGIS. If you’re ever in doubt, go with one of these native versions.
Learn more about the awesome of the punt flag.
Looks like GeoJSON got a new home on the Internet.’
— Sean Gillies (@sgillies) December 2, 2013
Not only that, GeoJSON has a new GitHub home.
— tschaub (@tschaub) November 26, 2013
Update 3: It appears this is a total FS effort to sell PDFs using Avenza. I talked with Avenza and they said that the FS is just a client of theirs using the marketplace.
Update 2: There is a free method here:
— joe larson (@oeon) November 27, 2013
— joe larson (@oeon) November 27, 2013
You’ll need to scroll to the bottom of that raster page because the retro image map doesn’t work with touch devices. The links are at the bottom.
Update: Well this is interesting, read this statement on the PDF marketplace.
A current printed copy of the Forest MVUM is available FREE from local Forest Service offices, and the current downloadable copy is available FREE from www.fs.fed.us/recreation/programs/ohv/ohv_maps.shtml
So the only thing here is the FS is promoting a 3rd party marketplace (well besides not bothering to create hyperlinks). The whole thing is completely weird. Why would the FS put out a press release promoting awesome PDF mobile mapping when it’s a 3rd party effort? Why not just point to the data for free online and be done with it? I’m completely confused here as to why the FS would even bother. Clue me in folks!
Interesting thoughts in the comments about PDF mobile maps being less than awesome:
I think there is some confusion here about the difference between data and maps. …… The maps built from this data (and other data) are a different product; maps are designed with a certain audience in mind and serve a specific purpose.
I’ve been chewing on this for a while since that comment was posted. Ignoring that I feel charging for digital data is (despite quoting some USDA regulation) offensive to me, I can’t really disagree with this statement. Does providing the data for free absolve any of implications of charging citizens for data they have already paid for?
If I put my old data marketplace hat on, the other part of this all that bothers me is the federal government is picking Avenza (or whatever they are called) to sell their maps for them. Quote whatever regulation you want, this stinks. You can have our maps as long as you pay some 3rd party for them? Tell me how that isn’t bad for the consumer?
Nowadays Python is probably the programming language of choice (besides R) for data scientists for prototyping, visualization, and running data analyses on small and medium sized data sets. And rightly so, I think, given the large number of available tools (just look at the list at the top of this article).
R you say?
While R has traditionally been the programming language of choice for data scientists, it is quickly ceding ground to Python.
While there are several reasons for the shift, perhaps the biggest one is that Python is general purpose and comparatively easy to learn whereas R remains a somewhat complex programming environment to master.
Common theme huh? Python is easy to learn. If you ever find yourself reminiscing the days when you used ARC/INFO at the command line to do all your processing and are tired of GUI tools constricting your creativity and productivity, embrace Python.
I’m a vector guy, I love nodes and even lines that connect nodes. Don’t even get me started on polygons! But rasters are still very much part of my life. I’ve used FME quite a bit to manipulate them because of it’s speed and lack of “legacy” COM madness. But even it has overhead that I don’t want all the time. I’ve written quite a bit about Fiona and how it really simplifies vector workflows so you better believe I sat up when I saw Sean’s announcement of Rasterio.
Ever since I wrote Fiona, I’ve been asked if I have plans to do something similar for geospatial raster data. Having been out of the raster business for a few years, I always said “no serious plans, just blue-sky ideas.” Today, I’m back in satellite image processing and very much want and need something like Fiona-for-rasters. Rasterio is my attempt to write such a Python package.
You’ll remember that Sean joined MapBox a couple months ago and clearly he’s been working on their cloudless satellite efforts. I’m sure the first thing he needed was better tools for manipulating raster images. If you use GDAL you’ll be really excited about Rasterio. Like OGR is to Fiona, GDAL is to Rasterio. Plus if your a NumPy user (if you aren’t, you should be) you can read rasters right into your NumPy arrays. Let that sink in for a moment. I’m so looking forward to the 4 day holiday weekend to start playing with Rasterio.
Ever since I picked up my iPad Air LTE (AT&T) earlier this month I’ve been testing the AT&T network to see how much faster it is than Verizon (my work iPhone is Verizon LTE). In my neighborhood AT&T LTE blows Verizon away to the point I attach my iPhone to my iPad hotspot to get faster speeds. But of course what happens in Tempe, AZ doesn’t translate across the country. We’re mostly left with commercials reminding us who the faster network everywhere else. And heaven forbid you ask the question on Twitter.
I’ve used 3rd party tools for this testing but you have to wonder how consistent they are across the network. Well now we’ve got the F.C.C. releasing an Android app to trying answer the question as to real networks speeds across the country.
The Federal Communications Commission has released a mobile speed test app for Android to help the agency crowdsource data about wireless performance across the country. The app, simply named the FCC Speed Test, doesn’t have the best looking design out there, but it doesn’t necessarily need to: once installed, it’ll automatically check a phone’s connection speed in the background when the device is not in use. While that’ll allow individual users to clearly see how well their own data provider is performing, it’ll more importantly allow the FCC to gather a wide amount of data on cellular carriers nationwide — that is, if its app gets enough users.
Now the kicker…
Early next year, the commission intends to release an interactive map online that’ll detail, in general, how well each data provider does in different locations. As the year goes on and it gathers more data, the FCC intends to make the map more and more detailed, adding in more local results, speed variations, and packet loss reports. It’s also promising infographics.
I’m going to go out on a limb here and say that the F.C.C. won’t be releasing PDF maps. I’m totally looking forward to the results. Now being iOS only I’ll have to wait until January to take part and it appears Windows Phone and Blackberry (I guess for all those Fed users) apps will be released as well.