Happy Thanksgiving!

Time to reflect on what we are thankful for. I count my blessings each and every day to be working for such a great company, in a great industry with great people. Enjoy your Turkey folks!

GeoThanksgiving

From our family to yours – Happy Thanksgiving!

Clockwise from the lower left; Steve Coast (stevec), Peter Batty (@pmbatty), Learon Dalbey (@learondalby), Thea Clay (@theaclay), Andrew Turner (@ajturner), Mikel Maron (@mikel), Kate Chapman (@wonderchoock), Paul Bissett (@pbissett), James Fee (@cageyjames).

HT: Norman Rockwell

What Steve Coast’s Move to Bing Really Means

OK, sure – there is a rush to blog the news. Steve Coast joined Bing. But leaving it at just that news misses the whole point of what Bing/Microsoft wants and why Steve would go join the Bing Maps team.

Microsoft and OpenStreetMap

Remember this news a couple months ago? It was cute right? Nothing earth shattering, just another layer to do your Bing Maps mashups over (assuming you are one of the 5 people who have ever done that). But deeper something else was going on. See Google broke the detente that the vendors had with the data providers. (can you believe that was barely a year ago?) In doing so it put a ton of pressure on the other mapping API vendors to change how they did business. Despite the problems with Google Maps, the one thing it had going for it was that it could be updated. Microsoft was stuck with the old model. Navteq is authoritative (By authoritative I mean busted and old business model) and that means it takes months to update the data. Google’s navigation might have been totally borked, but at least they controlled it. Microsoft and MapQuest didn’t have the luxury.

Time to Bail!

So MapQuest went first and in a big way. Not only did they announce they were breaking with their data vendor, but they were also going to donate money and time back to OSM to help further the project along. Already MapQuest’s Open website is running and those at least in the UK are happy as clams. But despite all those German OSM mappers doing great work in the middle of Europe, for OSM to be viable it needs to have a workable/routable map in the good old USA. This was the whole concept behind those Cloudmade ambassadors. The idea was they could have succeeded if they had the map right. Well the reality of the scope of the project killed Cloudmade’s efforts and the USA has lagged behind. What the OSM project needed was some company that could do the heavy lifting and move the project forward in the USA and other parts of the world where it needed help.

Enter the 500 lb Gorilla

So here we are. Microsoft needs to get involved with OpenStreetMap to continue to be relevant in the web mapping space and OSM needs Microsoft, their aerial images, their big pocketbook and their need to dominate all spaces they exist to join up. Simple as pie, right? Oh, but wait – the minute Microsoft enters the community, they’ll be a huge backlash. I mean the goal of Microsoft is to put a proprietary OS on everyone’s desktop (at least that is the feeling I got from an OSMer at WhereCamp5280). Ah, but remember this wonderful news? Who better to represent Microsoft with OSM than one of the founder’s of OSM!

The Way Forward

So what does this mean, really? Short term, not too much. The OSM data in the USA is still not good enough to route with ( Flame away guys, but you’d be wrong to think otherwise!). But the better aerials will improve the USA greatly and quickly. If there is one thing that Microsoft has done well, it is the aerial imagery in the greatest democracy south of Canada, north of Mexico; OSM-US . Plus rather than investing in their technology to route with Navteq, they’ll push those efforts toward OSM routing and BOOM; we’ll have a working, routable OSM map that everyone can use.

Makes you wonder about Google’s choice to roll their own now doesn’t it? How quickly Google’s map looks like it will be a drag on their innovation. Between the OSM mappers, MapQuest, Microsoft and all the others who are part of the open project; I see no way OSM doesn’t dominate the market as the choice for mapping data moving forward. And you know who wins, everyone who wants free and open data. That’s the take away – congratulations to Steve Coast because he sure deserves success, but the real winners here are you and I.

WhereCamp5280 – The Take-away

So yea, great time had by all. As far as an unconference goes, it was a nice improvement over last year. These things take time to get right and WhereCamp5280 is looking like its going to be a very successful gathering. So what did I see that I think I need to pay attention to in the next year.

  1. Google Fusion Tables – Are you kidding me? These stuff is “teh awesome”. Fusion tables are going to be more “killer” than Google Maps was. Yup, pay attention.
  2. Polymaps – I have no idea how I’m going to use this thing or even how the darn thing works (voodoo is the general assumpition), but damn it works and it kicks butt. You can’t be serious if you want to actual serve up vectors via WFS. Another game changer…
  3. Raphael – I was playing with this a couple months back and forgot about it. Don’t get me wrong, Google Chart API is sweet, but Raphael is like chocolate syrup (with whipped cream and a cherry on top). Hey look at that two SVG rendering projects in a row.
  4. SEO for Mapping – Brian Timoney talked about SEO and how data needs to be found better. I can’t find a link to his topic anywhere, but just know what is good enough for John Elway should be good enough for the City GIS Department. Seriously though, if you want your data found, you need to think about how Google crawls it.

There was a ton more discussed (see Andrei’s blog). What caught me off guard was the general lack of interest in Microsoft technology. In a huge shift from last year. Silverlight, .NET, SQL Server and Windows were all lacking in representation.

As I said, great time was had and it was good to get me focused on where I want to go in the next year.

ArcGIS Desktop 10 UI Wackiness

So recently I’ve moved my work from the 9.3.x version of ArcGIS to 10. There are some really great things with 10 that make it much easier to use (I love the basemaps being integrated in to the toolbar. So much easier than going to ArcGIS Online and clicking on the LYR file), but there are some other things that I just have to scratch my head and wonder what they heck were they thinking.

Dissolve is Good Honest GIS Work

I had to create a vector output from a Esri Grid last week and I did the good old Desktop equivalent of GRIDPOLY and got my shapefile. But as you’d expect, there was millions of little polygons from the grid. I did what any sane geogeek would do, grabbed the dissolve tool. A couple clicks later I’m ready to hit the run button and watch ArcGIS struggle to dissolve such a large/complex dataset. But when I clicked the run button, away went the dialog and I assumed the process crashed. So bring of the dissolve again and try it all over again. Yup, away goes the dialog and I have no idea what is going on. I go to force quick quit ArcGIS and I see its churning away on the shapefile. Hmph?

ArcGIS 10 Progress Bar

Back to the ArcGIS Desktop window and what do I see in the lower right hand corner? This bizarre feedback?

I’m sure all the ArcGIS 10 users know that thing pretty darn well. All part of the multi-threading that ArcGIS 10 does. But it is so counterintuitive that I wonder how many times I might have force quit ArcGIS while that thing was moving along.

So What am I Looking at Here?

It’s an interesting design choice for sure. I wonder if they were in a conference room and couldn’t agree if it should be a progress bar or percentage complete feedback. I can only imagine the eureka! moment when they came up with this design choice.

"It's like a progress bar, only better" -- Esri UI Design Specialist

No it could be the most annoying Esri status feedback since that “Cylon eye” thing that ArcView 3.x used to do. I admit, I’m not the youngest rooster in the coop and my eyesight is poor, but I feel like I have to struggle to see what the darn thing is saying. I’m not the only person who feels this way either. I’ve had about 5 emails about this in the past two weeks asking me essentially, “WTF?”.

Neither Here nor There

We all know ArcGIS Desktop 10 is not multi-threaded. Before this hack (Now that might be a strong word, it does work pretty darn well. Let us not kid ourselves though, it is just a bandaid solution until they get true multi-threaded ArcGIS out.), we had that dialog with the progress bar that was usually modal in nature. Click dissolve and then go to YouTube to watch Justin Beiber videos until ArcGIS was done. Now you don’t have to worry about keeping yourself busy while running a geoprocessing task. It does slow down ArcGIS (at least heavy processing), but it is surprisingly workable. But this half-empty solution clearly leads to weird design choices.

No I Don’t Want the Processing Dialog Back

What would work? For me, I like the percent complete feedback over a progress bar. But don’t scroll the darn thing. Just leave it up in the corner where is. If you have to have some sort of indication that something is going on, throw a throbber up there like happens when ArcMap redraws the view. That’s such a more elegant solution than this current one and it won’t make me put my nose on my screen trying to read what the darn thing is saying.

Oh and my dissolve is still running/scrolling along?