|
|
We’ve never been a company which sits on its hands and wonders what is around the corner. Sure we have some key partners, but they don’t limit our reach and exploration. Our goal is to provide the most appropriate solution to our clients. That might be an ESRI solution, Google, MapQuest, technology combination, open source. We are continually working to expand our skills and add more tools to our geospatial toolbox. The more tools we have available, the more effective we are at picking the right tool for the job. (we all know using pliers as a hammer is never ideal.)
In the past we have leaned on the likes of ESRI’s ArcGIS Server (and their various web mapping APIs) as well as some of the more advanced open-source options like GeoServer, OpenLayers, OpenScales, etc. But things are changing. Attend any GIS focused conference and you will notice two things. First, that ESRI now talk about “non GIS users”, and not just in passing; all the time. And second that Google are usually there in one form or other. After chatting with one senior Google geo person we decided to look at their offering in greater depth.
Google are considered by many professionals to be on the lighter side of geospatial. Their flagship products like Google Maps and Google Earth shook up the geospatial world in the mid 2000’s. Recently Google has been working hard to expand and deepen their roots into the geospatial world, particularly on the mobile side.
Rory (me) went out and did his own Google geo investigation.
Google Maps Geospatial Scope
Google obviously have many software products, services, and APIs related to geospatial so for my purposes I’ve limited my investigation to those areas applicable to mobile platforms. By no means is this intended to be an in-depth evaluation of all of their geospatial products, but more of a light overview of what’s available (or that we know they’re experimenting with), focusing on the potential for mobile applications in particular. Within the mobile realm, further separation must initially be made between native and web-based/cross-platform development. Google now has native Google Maps APIs available for both Android (MapView) and iOS (MapKit) for building custom mapping tools for those individual platforms. While single-platform (native) development can offer hardware specific advantages, the increased efforts to support multiple platforms tends to push us as a company more towards cross-platform, hybrid approaches such as Flex (via Adobe AIR) and increasingly HTML5/Javascript applications. Google Maps has now deprecated their Flash API in favor of their Javascript API in an effort to push the HTML5 advantages of plugin-less tools that work cross platform. Currently on its third version, it appears to be heavily oriented towards supporting mobile platforms.
Google API Buffet
The Google Maps Javascript API v3 is now the primary API for developing google-based maps in both the web and mobile environments. In addition to this main API, there are several sub/add-on APIs providing additional web services such as directions, distance, elevation, geocoding, streetview, latitude, and places. Some of these even have their own sub APIs as well such as the Places Autocomplete API which allows developers to not only leverage the power of Google’s places database, but also incorporate it as autocomplete suggestions to users typing a search, something that has almost come to be expected nowadays thanks to Google incorporating it into their primary search engine page. The power of these location-based searching and geo-marketing capabilities is undoubtedly leaps and bounds above their competitors in this area. Additionally, for consumer check-in/check-out applications the Latitude API can be used to access and update locations programmatically.
Google GIS lite – Static, Standard, Styled, and Custom Maps
For low bandwidth applications or mobile browsers that don’t fully support the power of these advanced javascript APIs, there is also a Static Maps API. A simple URL call can quickly return a custom static map image to the mobile application.
This may suit the user’s needs perfectly well in certain applications. For users that need full dynamic “slippy maps” the plethora of javascript APIs described above provides a variety of tools and standard Google map types such as terrain, streets, satellite imagery (with or without annotations), and even off-nadir 45-degree aerial imagery. However, lesser known options now available include adding your own styling to standard Google map types (e.g. different colored roads, parks, or backgrounds) or including your own custom basemaps (ie: your own imagery), tiled overlays (such as your own annotation layers) on top of basemaps, or even your own custom streetview panoramas.
Additionally, while Google uses the webmap-standard Mercator projection (EPSG 3857) for all of its maps, you are allowed to implement your own custom projections as well. Needless to say, Google Maps has worked hard to provide options beyond the basic streetmap.
Google Geospatial – Capabilities and Functionality
When you think of Google Maps, you think of fast basemaps, streetviews, routing, kml’s, etc… Very GIS-lite kinda stuff, but can it do more? Google Maps doesn’t try to be a full GIS, but to paraphrase them…many of their solutions come pretty close.
One exciting new area Google is experimenting with is Fusion Table Layers. To quote them, “Google Fusion Tables lets you store, share, query and visualize data tables. It offers a REST API to run SQL-like queries to manage tables […] The results of queries can be a .CSV or used in the Google Maps API or Google Chart Tools.” Any data that can be mapped in a fusion table can apparently be used as a layer in a webmap via the Maps API v3 now. Basic data such as points with field attributes can be put into a Google webmap or mobile application and even made clickable with info window popups. But what about more complex geometries? Can a fusion table layer contain lines and polygons? Could we take something like an ESRI shapefile and turn it into a fusion table layer for use in our maps? Quite possibly….yes! Further experimentation is needed on this idea, but in creating a fusion table for use as a map layer, you need at least one location column which can be lat/lon, geocoded addresses, OR “KML geometric data”. Could you import the geometry from a shapefile into a KML geometry and use it as your fusion table layer location field along with the other field attributes of your original shapefile? We haven’t tried it yet, but we think so. If you beat us to the punch and getting this tested before we do, be sure to let us know!
Aside from fusion table layers there are Overlay objects that can represent spatial data in the form of markers(points), polylines, circles, rectangles, or polygons. Any overlay can be made editable and combined with the Drawing library for allowing users to edit the overlay object. This editing is done on the client-side (basically as a graphics layer) and once the application is loaded, does not need to be online to edit. Thus offline editing could be possible using Google Maps, however the syncing of those edits back to a server appears to be lacking right now. Enter ODK (Open Data Kit), created by University of Washington’s Computer Science and Engineering. The Google Earth Outreach team seems to be currently recommending the use of ODK for offline field data collection. ODK currently works only on android devices (or simulated on a PC) and seems to be geared towards field data to replace paper forms rather than more complex GIS operations such as heads-up digitizing in the field. Although it should be noted that the obvious potential for combing the ODK system with overlays or fusion table layers is certainly something to be excited about and we’ll be looking into further. Additionally, the data collection part of ODK is a native android application and thus has the ability to tap into the hardware sensors of the device to collect location via the GPS chip, images or videos via the camera, and audio via the microphone. The real power of ODK though is the ability to collect data offline and then sync it to a server later on when the user has a connection again. The Google Earth team outlines a process by which to do this and eventually import the data collected into a fusion table for viewing as a layer on Google Maps. It’d be nice to see this process more streamlined and built into the abilities of the Maps API since offline editing requests are continuing to explode in the mobile geospatial realm, but at least there is a solution available, even if it’s not ideal.
Last to explore is the capabilities of hosting your own spatial data on a GIS server (or equivalent) for disseminating out to webmaps and mobile applications. Currently ESRI’s ArcGIS Server and the open-source GeoServer lead the pack here. Google’s answer to this is their new Earth Builder product. Google Earth Builder is designed to let users publish geospatial data into Google’s secure cloud. Google Earth Builder supports ESRI JP2, MrSID, GeoTiff, JPEG file formats for raster and ESRI Shapefiles, MapInfo Tabfiles, CSV, KML, KMZ for vector. Once published, the data can be accessed via the Maps or Earth APIs and viewed on webmaps, desktop apps, or android apps. Sharing and access control relies on Google Accounts and OAuth2. According to their site, “Google Earth Builder pricing is based on the amount of storage quota and pageviews for the account”, but no rates were listed.
Google Mobile Location Solutions – Licensing and Limitations
Google Maps data, APIs, and services have licensing and usage constraints. The Maps API is free to use as long as your site or application is free and publicly accessible. If usage of the Maps API by a for-profit site or application consistently exceeds the free daily usage limits, you’ll need to purchase additional map loads or purchase a Google Maps API for Business license. For-profit sites are permitted to generate up to 25,000 map loads per day for free, other services’ limits vary.
Conclusion
Overall Google’s buffet of geospatial tools and development APIs is actually pretty impressive! When compared to the expensive, but well-polished products of ESRI, they may seem a bit rough in some areas, but they are also considerably cheaper or free to use as well. Everything is a trade-off. The hardcore GIS capabilities are not up to the same level as others such as GeoServer or ESRI, but then again that’s not what Google is shooting for either. They seem to be happy helping to expand the lighter, much larger-scoped public side of geospatial (which do cover a lot of use-case scenarios). And they’ve done an excellent job on that side with the Maps and Earth products and associated APIs. The heavy emphasis on HTML5/javascript as well as tight cohesion amongst their APIs is really nice to see as well. However, the geospatial side of Google seems to be continuing to advance. The direction is not towards traditional GIS work, but more into expanding the geospatial capability and usability on mobile devices. Personally, we think this is a brilliant move.
The time they certainly are a changing. If you’ve hitched yourselves to ESRI, some exciting things are in the works with ArcGIS online. But the real action in the geospatial world is mobile and no one company solution will prevail. Many location based API’s have now been released, more are planned. Spatial problems we could not solve before, now have potential solutions. A wider, deeper approach is required. Google kicked the old GIS world in the pants in 2005. They have plenty of expertise and money on their side to do it again.
As a company we look widely for the best solutions, and Google is part of that solution mix.
Let us know your thoughts email us
We’ve mentioned before that the mobile market is in a very fluid state. New hardware releases, differing approaches to app development and new tools to help improve locational accuracy are just a few of these areas of change. The lack of maturity is one of the reasons why the market remains consumer driven. Business’ fear any investment today will need replacing tomorrow. But enterprise adoption of mobile is beginning to ramp up. Companies have started to realize the potential improvements in efficiency and operation that mobile can bring. Let’s review the mobile landscape from a corporate perspective.
Mobile Hardware
A survey of corporate users in the USA by Good Technology found that Apple’s iPhone 4S is the most popular smartphone with 37 percent of all mobile activations for the first quarter of 2012 (four times that of any other device). The iPad 2 claimed the second spot overall, with 17.7 percent of activations for the quarter, with the newly released iPad claiming 4.3 percent of all activations for the quarter, and an impressive 12.1 percent of activations in March alone. Android smartphones represented 26.1 percent of all activations for the quarter, while Android tablets came in at 2.7 percent.
Apple mobile products continue to be the most popular mobile choice within corporations. Yet, Android is now challenging iOS in the consumer market. And with Windows increasing popularity, competition in the mobile hardware space will continue to grow.
Mobile Apps
Today most mobile apps are targeted at the Apple iOS and Android platforms. We’ve mentioned in previous columns, there are two possible approaches to mobile software; web or installed apps, let’s explain that in practical terms. Thinking about the gas pipeline business, much activity is conducted on site. Operations is a key department. Many of those working in operations carry mobile devices, though they are not all tech savvy, the use of mobile apps providing them with simplified workflows could dramatically improve how they complete their daily tasks. Increases in efficiency, thus reduces company costs. Let’s imagine a potential scenario. An operations manager needs to record safety inspections data on site. Location, time, text and image data needs collecting. His smartphone has all the tools needed to do this work; a camera, mobile app designed to provide a workflow targeted at the task in hand, local and network storage of the data. So will a web based mobile app or installed mobile app best serve his purpose? In this case an installed app. There are two reasons why this is the case. First, working with imagery and local storage is not possible through a browser based app. As is the case on PC’s, largely for security reasons, browsers have limited access to local resources. Second, it is quite possible that the operations manager will be in areas without wi-fi access. No wi-fi access eliminates the possibility of using web based apps. In this case, any operations data required, needs storing and accessing locally. So interactive maps, base maps and pipeline relevant layers, will need to be stored as tile packages on the device.
Locational Accuracy
Mention need be made of accuracy. At present most mobile devices have GPS accuracy to around 5m. Conversations we have recently had with the US Forest Service and Chevron, have raised major concerns over this level of accuracy. They need sub metre accuracy. Efforts are now underway to improve these ground readings. The US based Gas Technology Institute recently announced a new high accuracy GPS integration with tablet devices used to document new as-built pipeline installations.
Mobile Location Solutions
The GeoInformatics sector is gradually undergoing change. The niche that was GIS is beginning to coalesce with a plethora of other location focused services. A number of traditional GIS companies have widened their focus and mission to location technologists. The advance of the mobile revolution is exposing a wider audience to spatial tools and solutions. Attend any ESRI presentation and you will hear increasing reference to satisfying non-GIS users. Google have started moving into areas traditionally served by GIS. Attending a recent oil and gas workshop, we listened to Google directly encouraging ArcGIS application developers to start looking at their solutions. Consumer focused location based service providers such as Foursquare are using location technology for marketing and advertising purposes. Indoor Positioning System (IPS) is a navigation system which works indoors. Though still in the development stages, this system will allow for indoor geoinformatics.
The geospatial world is changing; those focused on geoinformatics are being presented with a huge opportunity to move from the periphery, to the centre of the information technology stage.
Feel free to contact us with your thoughts
Just back from CalGIS in Sacramento. An interesting two day conference. Our presentation; ‘Is mobile the Future of GIS?” was one of the last sessions on the Friday. Its a shame since we gave an overview of the mobile market space; in hindsight it would have been better positioned early in the conference. Even so, we still managed a good sized audience.
The conference was made up of a cross section of GIS focused organisations across the state; both public and private. We were interested to see which themes were most discussed, and gauge where mobile fits into a traditionally web/desktop GIS focused conference.
GIS WorkFlows
The discussion of simple workflows dominated many discussions. Both in casual conversations and as part of presentations. How do we provide users an easy way to get their tasks done, was often repeated. Avoiding overhead and unnecessary steps. We heard criticism of the training needed to work with tools like ArcPad. Hide the complexity. Make apps intuitive, design them with as few steps as possible to arrive at an outcome or solution.
Open Source GIS
Open source GIS is alive and ever more popular. The keynote mentioned the work the State of California is doing to build open source GIS solutions. Conversations with LA County and 34 North showed both how popular and robust open source GIS tools have become.
Google GIS
Another GIS conference with a Google presence. Josh Livni gave a very interesting presentation on geospatial Google solutions. To paraphase Josh; we aren’t GIS solutions providers, but many of our solutions come pretty close.
Mobile GIS
Apart from Farallon Geographics, we were the only company to focus our presentation wholly on mobile GIS. Most presenters mentioned mobile in passing. Data collection was the most common mobile use case. Mobile adoption in the GIS sector has proved slow. In part due to mobiles relative newness, which has created various levels of confusion. This was confirmed from all we saw and heard at the conference.
GIS or Location Based Solutions?
It was hard to resist; repeating our thoughts on retiring the term GIS. We raised the question a number of times. Not to irritate, simply to gauge opinion. Others are beginning to agree that the term GIS does not capture the true core of what we all do; provide location based solutions!
Thoughts
Overall, this was a well attended, most interesting conference. It reinforced the path we have followed as a company. To build low cost custom location focused mobile apps which provide simplified workflows. We see mobile presenting huge opportunities to the location focused (GIS) sector. As we discussed in the final slide of our presentation. Mobile brings:
1) Context – i can now see live, while in the field my location/GIS data
2) New users have started using location tools. This user base will continue to expand.
3) New tools can now be built which take advantage of mobility and context.
4) The markets for location apps will expand.
Widespread adoption of mobile location apps remains consumer focused with the likes of FourSquare. This is beginning to change. Companies like ours will continue to expand as clients come to us for an increasing range of location based mobile app solutions.
Let us know your thoughts
It seems the blog post we wrote recently entitled “Should we retire the term GIS?” resonated with the geo-community. We thought is worth a follow up post.
First the responses. Below are a few on the comments we received:
“No. I saw this a few months ago (maybe from the same group; maybe not). This came from a location based services firm – so that’s why they want to change it. GIS is still the main term to use, for what I do at least; and folks can still use lots of sub terms if they want.”
“I’ve also heard geospatial technologies (geospatial being duplicative). But GIS includes the concept of analysis, where location does not. Location Based Solutions are applications, while GIS is analysis – how about “Geographic and Location Based Solutions?”)”
“True, I completely agree. GIS is not only a niche term, it is a discipline which occupies the time of people aged 55+ dealing with sub millimeter accuracy, INSPIRE (I prefer to call it EXPIRE), land management and maybe utilities. It has nothing to do with crowd sourcing (OSM), modern technology (mobile apps, SOA, Cloud) or “new” markets (Business GIS, LBS etc.). GIS people are still looking for that “killer app”, but cannot find it. Call it location services, Spatial Business Intelligence or whatever: the people in the industry need to change. We do still need the “55+ sub millimeter” people. Mainly for accuracy and standards.”
“I agree with most of what you have to say except I disagree that the term GIS should be retired. What I’d like to see is the use of specific terms where appropriate (like LBS as you’ve suggested) instead of using GIS as a catch-all for all things spatial. Just my two cents.”
Very interesting. Actually (counter to the first comment above) we have historically been an ESRI focused company. We wrote our first Web based ESRI map viewer (for the US Forest Service) back in 1997. Long, long before the term location based services (LBS) had been coined. Maybe this is at the crux of the discussion. Mobile has made us rethink our mission. We are a company started by geographers. Ok, we have a more diverse group together now, but geography remains at the heart of our work. Our focus is understanding better, and visualizing the space around us.
Geography, space, location, place. With my mobile in hand, I want information about what or who is near me. I want to run analysis based on my current lat/long. Is that handled by a GIS, LBS or other geo-backend? I care little, I just know that my current location is the centre of my focus. Whether I am a public utility worker trying to visualize an underground pipe (layer) and run a network analysis to find the valves which feed the pipe beneath my feet. Or I’m simply trying to see which friends are within a certain distance of my current location. With mobile everything is about location.
We are at the beginning of a geo-sector boom. Its a boom which is driven by mobile, centred on location. Nobody cares about the tags attached to the technology which provide the solutions. Users just want the solutions. And they are location based solutions.
Feel free to let us know your thoughts.
The future of mobile is location! ………. The future of computing is mobile!
Two bold statements. We don’t necessarily believe them yet. But one would be foolish to ignore them offhand. What do these have to do with our question:
“Should we retire the term GIS”?
GIS deals with location. Why not simply use this universally understood term when we sell our geo-technology solutions .. and drop GIS altogether? As the need for location technology grows, lets begin to use the language all can understand. GIS is a niche term understood by geo-nerds, often in the public sector (nothing like a good generalization).
I heard a story the other day from a geo-friend of a geo-friend. She works for, as she describes, a fantastic GIS company. The story was about the company meeting last year. One of her colleagues asked the Head of Sales “should we advertise ourselves as a GIS or location focused company?”. The Head of Sales and CEO both chipped in with pleasant answers. But there was a depth to the question noted by my friends friend. Today, a year later, there has begun a discussion within the company on the absence of the term GIS at geo-focused conferences. My friends friend says it makes her smile. She has quietly patted her colleague, who asked the question, on the back.
Why tell that story? Like it or not, the term GIS is being edged out. New geo-sectors are emerging; see Location Based Services (LBS) for example. Is it GIS? No, well yes. It deals with location. The term is in the title. Okay, its consumer focused, a tool for marketeers. But who cares?
Mobile is changing our perception of terms like location. Let’s try that silly elevator game, where you are pitching new company ideas, in one sentence:
“My new company will provide GIS services.”
“Our company will employ geographers and provide information about the space around us.”
“The focus of our company will be to provide location(s) based information and services.”
I know in which of these three companies a layman would invest his money. And which one potential clients would gravitate towards.
As mobile becomes more popular, so too will the need for solutions which use location data. We are entering a boom period for companies in the geo-sector. Its time to drop the term GIS. Let’s all call ourselves what we are: providers of location based solutions.
Feel free to let us know your thoughts.
|
Follow Us!