5 predictions Geo for 2010 and 5 things that won’t happen
December 31, 2009 59 Comments
Here are 5 predictions for Twenty Ten.
- The shapefile dies: SpatiaLite + ESRI’s File Geodatabase API finally put a dagger in the shapefile.
- GIS on iPhone/iSlate (Apple Tablet) and Android/Chrome OS: With Apple and Google owning the mobile space, we’ll see more proprietary and open source projects being ported to these platforms. Microsoft Tablet PCs and Windows Mobile/CE begin to die off.
- 64-bit: There will be some holdouts (*cough* ESRI), but most of us will be running native 64-bit code on our desktops and servers. Now to just get more RAM in this laptop.
- Mobile: If you aren’t running on the iPhone/Android/Blackberry you aren’t relevant. Web mapping apps become mobile browser aware. Those that aren’t were probably irrelevant anyway.
- Google: Google’s APIs continue to push the envelope and they continue to be the standard for everyone mapping on the interweb. Google is able to throw so much money and manpower at “problems” and their solutions are coming faster than anyone else can match.
Here are 5 things that won’t happen:
- Augmented Reality: Much like the Nintendo Virtual Boy, it sounds great until you try and use it.
- OpenStreetMap Dominates: Between Google’s quick improving of their database and continued licensing issues OSM plateaus. Companies will continue to try and figure out how to monetize OSM, but fail.
- ESRI + Microsoft: This was on the top 10 lists for many people in 2009, but I don’t think we’ll be seeing deeper integration. ESRI will continue to support multiple platforms (Google, OSM, Microsoft, “other”) and not become a Microsoft shop. As Google continues to erode away at SharePoint and Bing Maps, ESRI will make sure that they don’t get caught in Microsoft’s blind spot.
- Geolocation other than Twitter, Apple and Google (TAG): Foursquare, Brightkite, and others will fade as TAG rolls out new APIs and ensure their mobile devices are tagging everything you do.
- MySQL falls apart: Despite the dire predictions of Oracle or Monty destroying the project, too many people have too much invested in the project to let it fail. MySQL will be fine and LAMP will continue to power Badgers.
Hey, don’t worry… It’s gonna be a bright, sun-shiny day!
About James Fee
Hey, what about Manifold?????
Mani-what?
The shapefile will never die.
It already is. ESRI users don’t use it anymore. They use FGDB. The only issue is sharing those files with others. With a FGDB API, the problem goes moot and ESRI users don’t have to convert.
ESRI users don’t use it anymore.
Oh how I wish this were the case among other ESRI users I generally interact with. I send out plenty of FGDBs, but get plenty of .shp, .shx, and .dbf in return, prompting a good deal of facepalm on my part.
Ummm, hell yeah, ESRI users are still using the shapefile. Heck, sometimes it’s more useful to generate it and send the DBF portion to non-GIS users that need to read results in Excel.
Until there’s a universal file out there that any program can read and edit location based information, I can’t see the shapefile disappearing. The personal geodatabase worked great for sharing info but a lot of non-GIS colleagues didn’t have MS Access, so more often than not, I ended up converting tables to an Excel spreadsheet. Or a shapefile if they were familiar with the basics of ArcMap or only had older version of ArcView.
My 2 cents.
Taking a geodatabase and converting to SHP can cause problems because DBF is so bad. Fields get screwed up.
A better solution would be to convert to XLS and with the FGDB API you’ll be able to do that very easily from what I’ve heard.
The problem only goes away if ESRI produces an open FGDB API that can be used to build support in other applications, including open source GIS…
The problem with FGDB is that it is an additional step away from my data. My prediction for 2010 is that the distance between users and data will be reduced (not increased as in FGDB). Users (and apps) want instant access to data, not to be told that first you must create the “container” and then you can create/store data in it.
That step doesn’t exist in shapefiles and is why I think users are hanging on to them. I can save data immediately to SHP and not stop in the middle of my workflow to create a FGDB just to store data.
AA
The step is one created by ESRI, not a limitation of FGDB. You could easily create a FGDB on the fly, but the reality is you want to store multiple data types in one location and you just can’t do that with SHP. No raster support and DBF limitations means SHP is headed to the trash heap.
For those same reasons SpatiaLite (SQLite) will start to take off in the non-ESRI world.
I agree….. for what its worth, I teach a course on spatial data design that uses FGDB as a centerpiece. I just encounter a large number of folks that have never even tried GDB (in any flavor). Whatever is simple (and still works) will win. Einstein said so.
Users for the most part don’t seem to care about the more advanced functionality, and if they do they are more likely to hire someone to do it!
2010 will be a good year indeed.
Sorry to dig this rabbit hole deeper but why can I not create a new FGDB directly from the export window or other GP so I can put my data into it? ESRI forces me to have the forethought to have already created a FGDB before I start me work. As Aaron points out, this is a barrier to using a FGDB.
Shapefile, despite it’s limitations, is alive and well. Anyone depending on FGDB (and an ESRI FGDB API *eye roll*) over shapefile format within 2010 is bending over in front of a cannon.
2010 will see Google taking more control over their map data as they have done for the US in 2009. MapMapker will likely become more important in their policy of owning geodata instead if licensing it. It will be interesting to see how the crowdsourced mapping dichotomy (Google vs. OSM) will shape up in 2010. Google is doing a lot to win over mappers and provides something OSM lacks: really easy to use tools. But OSM provides something Google can nor will: a free and open geodata platform.
OpenStreetMap sure will not dominate crowdsourced mapping in 2010, but it will for sure continue to gain momentum, and it will be an interesting year. We will see how much value people put on retaining ownership of their contributions to crowdsourced information repositories. Yelp declining a 1/2 billion acquisition offer from Google might tell us something about that.
With the next version of MS Office offering 64-bit support it’s rather shocking to think that ESRI still seems to show limited interest in 64-bit architecture (based on the FAQ from the last UC). The datasets we are working with are getting larger by the year, but the ESRI tools to handle those datasets are not keeping pace. Half a year ago Applied Imagery released a 64-bit version of their LiDAR software package (QT Modeler). I spent ~$2k to up the memory of my Dell T7500 from 4GB to 48GB and was amazed at what the software would allow me to do. Conceptually simple tasks, such as generating a seamless raster from millions of LiDAR points went from taking days in ArcGIS to less than an hour in 64-bit QT Modeler.
ESRI’s inability to address the 32/64 issue should be listed in the “things that won’t happen” section above.
This is a *real problem* for ESRI. There are many, many snippets of old 32-bit code rattling around in the bolus. These 3rd-party pockets are really dragging them down.
Scott can’t toss them because so much of the total picture relies on the calls back to these products; things like just about every one of the extensions, any of the backward-compatible items, and all their users with mixed-version environments.
The mixed-use and backward-compatible connections to old versions is [i think] the big issue that’s stopping the change. I wonder how many of ESRI’s total user base are *not* on something in the 9-dot-whatever sequence?
What would happen to ESRI’s share internationally if they forced an upgrade? How many users would re-evaluate the whole platform?
Can you imagine the carnage if they forked the code and started deprecating the 32-bit version over time? It’s what they need to do, but boy oh boy it’s going to make ‘their community’ really cranky.
Augmented reality will probably have to wait for a more major breakthrough in human interfaces to really catch on – probably in this direction.
@James Fee,
“Mani-what”
Modern, that is what. Your naive retort seems to be one of inverted schadenfreude.
Lets put your retort in perspective…
It is now early January 2010.
x86-64 CPUs first became available April 2003 (first of AMD’s Opteron series), a 64-bit version of Windows (Windows XP x64) was available soon after, initially in beta-release.
Users of Manifold System, were the first to be able to enjoy the greater stability, computational power and vastly greater amount of addressable memory (RAM and virtual) that a 64-bit Windows brings over 32-bit Windows, since the release of Manifold System 7.x, (if i recall correctly, it was that long ago).
More recently, 64-bit versions of TNTMips and of MapInfo have also been released.
Meanwhile, ESRI and ESRI product users are still using and are thus limited by, 1990′s hardware and OS technology, at the end of the first decade of the 21st century. This shows how utterly technologically obsolete ESRI products have now become. To put this into perspective, it is quite possible that Microsoft will not be releasing Windows 8, in a 32-bit edition and certainly will soon end support for 32-bit versions of Windows.
Instead, you should be asking the question why it is that ESRI has not updated its products. Then since they have not, why are ESRI product users continuing to put up with this and use such obsolete software.
My prediction for 2010: ESRI will continue with its strategy of managed obsolescence for its software products, whilst becoming increasingly marginalised and obsolete.
Obsolete software with over a million users and a $2-Billion [according to Forbes] corporate valuation.
VHS put Beta out of business quite handily, thank you very much, even though Beta was demonstrably better technology.
It’s not the quality of the technology, it’s marketing, business acumen and ultimately market share that wins the race.
I have an old car with a hole in its Manifold. It makes flatus-like sounds whenever it starts – perhaps you might assist with some of your 64 bits?
[Don't get me wrong...I am no apologist or defender of ESRI's moribund code-bolus. But the answer isn't in Manifold, either.]
It almost sounds like you’re saying ESRI is too big to fail? What a perilous philosophy that has proven!
The true innovators in the geo-sphere are not selling desktop GIS software products. Manifold (warts and all), OSM, Google APIs-galore, spatial SaaS, spatial DBs are all examples of products that are gaining popularity because the spatial community is tired of being held hostage by bloated GIS packages that have legacy code older than the latest college grad.
I think GIS, into the foreseeable future, is going to have a hard time defining itself. There is so much functional overlap in other ‘free’ products whose main goal isn’t to replace GIS. ESRI has been *reacting* to these types of changes ever since 8.0.
What surprises me the most is the protective stance ESRI users have for their products. What’s wrong with a little competition? Do you really think you’re getting $13,000/license + 20%/year functionality?
I agree with the you on the cost, except in Australia ArcGIS Server is AU$13,400 per core for Standard and AU$26,600 for Advanced last time I checked about 12 months ago. The problem is ArcGIS desktop is so entrenched, and resourcing is difficult to justify for someone to build WF and WM services for another product, when it is very easy to publish an mxd to ArcGIS Server
The protective stance is two-fold.
One is training. Not so much of the people you see mounting the defenses, but of the people under them. Finding 20 people who can competently generate spatial data from a legal description using ArcMap for $30k per year is difficult enough. Now try to find 20 people who can do that same task for $30k per year using one of the cutting edge GIS products. Or watch how many of your existing staff take their pensions and go elsewhere when you announce that you are switching to the opengeo stack and they will all train on the new software.
And that leads to the second part, budget. When you start moving to “cheaper” products like Manifold or the Microsoft or Google APIs, you make up for the cost in software with the cost in support and man hours. Software is pretty cheap in comparison. If that ESRI stack means employing one less person or paying $30k less per year to consultants/support companies, then we can buy an awful lot of ESRI software. And purchasing software is easy (maintenance, not so much), but hiring is difficult. So when the choice is between buying more expensive less capable software that meets the primary workflow requirements that requires less staff to maintain and is more familiar to more staff with a wide range of skill sets, versus buying less expensive more capable software that exceeds the primary workflow requirements but requires more staff to maintain and is less familiar to the broader staff that has a wide range of skill sets; the more expensive software is going to win. (It’s not just GIS. Emergency management software is big business, and the high capability cutting edge packages that require more staff maintenance hours are rapidly losing to more expensive less capable turnkey packages that require little to no staff maintenance hours.)
You’re right, people are hard, and “turn-key” is good, because it (theoretically) takes less people. But of course, providing solid turn-key products requires a tight functionality box (this platform *only*, this data model *only*) and the pain point hits when the organization decides to step outside the box. That’s when the duct tape comes out and the value proposition quickly inverts. The tragedy is organizations rarely know in advance which category they fit in: off-the-shelfer or customizer?
I’ve found that organizations just plain do not step outside the box in that scenario.
They just build a new box, and then hire someone one-time to make the old box and the new box talk
How is Manifold 9 working out for you guys?
How is Manifold 9 working out for you guys?
Apparently, not too well….
http://forum.manifold.net/forum/t92551.87
looks like our little Nicky has a few problems with it:
All I have received from you so far, then and now, is a lecture and patronising, dismissive statements, that boil down to “that there cannot possibly be a problem, because, “hey this is Manifold” “. You pretend this, if you like, but it does not change the reality that your product is not functioning as we are led to believe it should.
Hi there, I think ESRI and Microsoft are in bed together, guess who has to move off Solaris
. I am waiting for the sale announcement and ensuing federal anti monopoly investigation
Spot on for your forth won’t, I think. Was at Plazes.com for years (literally) which is now dead after Nokia bought it. Brightkite came and went. Am now dabbling with foursquare a bit. This needs to be obiquitous and needs to work (transparantly) for large numbers of people to actually use it. Could be a while…
“Continued licensing issues” aren’t the biggest challenge for OSM. I’m more concerned about the community’s head disappearing up its own fundament, Wikipedia-style, caught between “Rules! Rules! Ve Must Haf Rules” and mailing list keyboard heroes.
Too many other apps support shape file imports and viewing…give it a couple more years, then, it might go away…
I agree – there are too many other applications that use shape files, and it is still very widely used to move data between applications. It will be a while before it dies.
One must wonder though when and if ESRI will move to 64 bit on the desktop. This is not an either/or question. Many vendors in the 3D market moved to 64 bit years ago, but still support 32 bit (each new version comes in 32 and 64 bit flavors). Well designed code can be compliled to both. More GIS vendors are going to 64 bit as noted above (recently Global Mappper went 64 bit), even open source applications like MapWindow and Grass are now 64 bit. The 64 bit ship has set sail, ESRI better jump on board before it leaves port, or they will miss the boat, and have to hire a speed boat to try to catch up and get on board.
The problem is that ESRI salesmen will tell their users “actually, 64-bit really doesn’t make that much of a difference”. Then, their users will say “actually, 64-bit really doesn’t make that much of a difference”.
This is a real problem, as ESRI in many ways is setting the agenda with their user base, and all their educational materials are designed to support ESRI software. Thats why you don’t really hear about more of the cutting edge things in ESRI training.
How true! I think in other circles this is often often refered to as cult behavior. But regardless of what it is called, you have to admire the ESRI marketing machine. They are very very good at what they do. Of course the cutting edge things you do hear about usually come with a rather heafty price tag.
I don’t know how many times I’ve crashed ArcGIS (map or toolbox) with large datasets or projects. I’m convinced that if ArcGIS was 64 bit, the extra memory headspace would resolve most of these issues.
Pingback: The Shapefile – Reports of My Deprecation have been Greatly Exaggerated – Spatial Galaxy
Pingback: excited about in 2010 :: High Earth Orbit
James – sorry for the slight attack here but you’re posts (for me IMHO) are starting to really drink way too far into the Google kool-aid… its bordering on the Apple fanboi level…
Google killing SharePoint… really? seriously? c’mon James you can’t be serious…
sincerely,
Kevin
Yea I guess you are right. I’ve never run into Sharepoint in the wild myself so I guess it was dead already. To be fair though, I’ve never seen an Android in the wild either so take that as you will.
Google Mail is starting to take away (http://googleenterprise.blogspot.com/2009/12/why-city-of-los-angeles-chose-google.html) from Exchange and if you aren’t running exchange, then there isn’t any incentive to run SharePoint. WeoGeo dumped exchange for Google Apps and I know many others who are doing it as well.
SharePoint is a ponzi scheme by Microsoft and their integrators to get companies to spend money on things they don’t need.
Sharepoint is being used more in my agency now. For whatever reason, the past year has been a “good” year for Sharepoint here. Seems to be in a upward swing.
As for the future. It doesn’t matter what we do and don’t look forward too. The end of the world as we know it happens in 2012.
Makes all this moot!
“if you aren’t running exchange, then there isn’t any incentive to run SharePoint” – you do know that sharepoint is a intranet collaboration system right?
My prediction is that GIS integration with SharePoint will explode in 2010 bringing spatial data to everyone in the organisation and breaking down the barriers that exist today. While GIS professionals seek better tools the real market explosion for GIS is bringing the data to everyone in the organisation on a web based system that can scale to thousands of users and interfaces that are easy to understand for non GIS users.
I’m a huge fan of what is possible with Silverlight, Bing Maps and Sharepoint. Clearly James you should go and check it out sometime this year.
We tried SP and GIS a couple of years ago and I still have nighmares about it. Horrible doesn’t even begin to address what happened. The Forest of Woe awaits anyone trying to work with data outside of the MS model.
I know what SharePoint is, but if you don’t need Exchange, you don’t need Microsoft Servers and then you don’t need SharePoint. That said, since you seem to be a SharePoint integrator I hope you have a successful year.
To me on the outside, Microsoft is a very distracted company and I wouldn’t tell any client of mine to standardize on any of their proprietary systems.
I’m a web developer James, sharepoint is one platform that allows us to integrate web based GIS into an organisation, the interesting part for me is the GIS app rather then what hosts it, MOSS seems to be the dominant intranet plaform out there. I’m not going to reinvent the wheel.
In the real world you have to standardize on propriety systems, there the ones with robust support that management feels safe in deploying. What is important is that the data is in a open format, easily accessed by any system and easily exported to the next system. Sharepoint is pretty good at that. It sounds like what is really lacking is a good public demonstration of what is possible with GIS in MOSS, the unfortuneate side of being an intranet platform.
That is not strictly true, I know of several installations that are running SharePoint as an Internet & Intranet Content Management System (CMS) and a couple of others who are using it as a Document Management system, having Exchange certainly helps as does Office 2007 or later. I am most definitely not an MS fan, but the integration between SharePoint and front end Office products, particularly in Office & SharePoint 2010 is impressive.
Oh, behave…get Grooovy, baby!
Works when you’re not online, then syncs when you reconnect. You’d be stunned at the number of folks that like the idea of Groove.
I wish I could say I’ve never run into it in the wild. It is proliferating wildly across the Federal Government. Just about every new “portal” or intranet seems to be based on SharePoint these days. It’s like a stimulus package for Microsoft and I shudder to think how much tax money is shoveled into licensing, installation, maintenance and customization of it across the government.
James -
SharePoint is only for Exchange? Sure SP can integrate with Exchange but its way more then that – document manager/archive/backup/search, wiki, integration point for all things MS Office not to mention Groove for connected/disconnect applications. Even a few CMS qualities as well. That, and its ability to easily consume and utilize GIS make it a great intranet/portal technology. Way better then most out there and I have seem my share of intranets that give the Shivers now a days (*Vingette anyone* YIKES)
Now with MS offer it as a SAAS solution its even easier with a “pay as you go or your need increases” it saves my company (web dev shop >20 employees) from the hassle of rolling our own freeware/opensource solution that its ***FREE*** yet costs a ton of man ours supporting/triaging/etc… when this is done as a service the cost is negligible…
Also – your point of “we have gone google for office apps” is like saying moving away from ArcGIS Server because Google Maps API gives you “everything you need”… umm… sure.
regards,
Kevin
Kevin, what useful feature can exchange do that Google Apps can’t?
Ok I’ll roll the dice… You get keep physical ownership of your own data?? Just a thought…
Dude, that’s like sooo five years ago, man.
[Of course that's exactly the prevailing attitude in just about any organization larger than fifty people...and don't EVEN get me started about the locked-down fiefdoms inside our stuck-in-the-mud Federal bureaucracy.]
At a recent cloudcamp, I listened to a lot of people talking about how government shoudl be solely in the cloud, and that cloud computing, SaaS, and open source should be the standards in all parts of government.
That’s when I came up with an interesting criteria for them (which relates to Kevin Dunlop’s points below)…
Would you run your county’s 911 and dispatch center on the cloud?
I think the same question is appropriate here, since a lot of 911 operations are tied into active directory, sharepoint, and office (scary, isn’t it?).
Would you run your city’s or county’s 911 and dispatch center on Google Apps?
I’ve done E-911 work and the answer is not only no but hell no! Ideally I’d like my apps and data in the computer next to me protected by me and my .40!!
3 different orgizations I deal with (2 private, 1 gov’t) are moving or already have moved to SharePoint. SharePoint is a powerful tool if set up correctly and actually used. When my office moved from a file server to a sharepoint environment, the hardest part was to make people use it for everything instead of sneaking back to the file server. But now everyone uses and it is a major part of our workflow.
One advantage of SharePoint over G Apps is that it can sit inside the firewall. This decreases your internet traffic which is often slow to start with for some ISPs.
As predicted, GIS has hit the iPhone with the iTunes App store release of iGIS v1.0
http://www.geometryit.com/index.php/igis
one more 2010 prediction: Many users will continue to drop into ArcView 3.3 once in a while to “just get it done”.
or ArcInfo Workstation — the COVERAGE model is still alive and well with some of us Luddites. Believe it or not.
ESRI Tech support folks were trying to tell a friend of mine at ESRI that no one is using workstation anymore. The truth is a) it is very very solid and b) what’s the point of reporting a problem with workstation — users just fine a workaround!
Totally agree with the above statement. Still use coverages and workstation. Love my workstation, if it ever went away, it would be a sad, sad day. IMHO, it’s the best product ESRI has made so far.
PIOS was pretty cool too.
I vote for ESRI’s AUTOMAP II with the SYMAP ruler!
Pingback: Emergency Management Software - Topic Research, Trends and Surveys