Some of the earliest posts on this blog were about how powerful location aware services could be.  I spoke about Mobile Mashups and SOA, as well as Location Based Services (LBS) and AGPS and finally the value of Location (and how I thought it was relevant, but not quite sure what to do with it).  You’ll notice this is “so long ago” in Internet History (last year) that I refer to Touchstone, now called Particls.

So it’s with great interest that I can see some of the things I was blogging and thinking about beginning to become a reality.  I was a user of Plazes for a while, but ultimately gave it up because to some extent I just didn’t get it and I think having seen BrightKite now, it was missing the fine-grained access I needed to feel comfortable at that time (it may have it now I haven’t re-checked).

BrightKite is a very compelling and interesting service, provides reasonable privacy settings, but also links in to Twitter by default to set either your location or tweet it.  I’m going to see how it evolves, but I can see BrightKite becoming part of my tweeting in the future (especially when I get a new phone in the next couple of months).  It will potentially be my photo-log of where and what I’m doing, auto-feeding this in to Twitter as appropriate, while I’ll use Twitter directly through it’s various other clients for my 140 character tweets on the world and the community.

Another interesting feature of BrightKite is FireEagle. FireEagle sounds like it’s the DataPortability answer to Location Based Services, an open exchange for location information and the fact that a player like Yahoo is making a move in this space suggests we are at the cusp of big things to come.

Dopplr also adds to the picture of location based services, although in this case Dopplr fulfills a “Where am I going” rather than a “Where am I” feature.

I speculated early on, that LBS would potentially be of use for the Enterprise, but I couldn’t quite touch on the use.  The big area I could see is staying in touch with the people who are part of your Enterprise as they move around, with BrightKite and it’s GEORSS feeds, Dopplr and it’s “where am I going to be” and services like FireEagle, I can see that in the next 6 - 12 months, this will start to become a reality.

To paraphrase myself from over twelve months ago:

As Architects, understanding how location works and what location information is needed to tie the system together will be critical steps that can be taken now to ensure early mover advantage when it finally becomes a reality.

Farewell Plazes.  Regular visitors will know that I’ve had a Plazes locator on the top right of my sidebar for several months now.  After trialling it for a while, it hasn’t made the cut and I’ve finally decided to remove it.

Now lets start by saying there really isn’t anything “wrong” with Plazes, it’s a well presented tool (little bit buggy in some early releases but nothing terrible and the team fixed this), but ultimately it just wasn’t… well, useful!

 I feel that location based services really are just an iteration away from being “the next big thing“, but right now, while we can get excited about the fun things that they do (I love geo locating photos in Picassa), for business there really isn’t a requirement.   I’ve firmed up this after using Plazes from a position of thinking there probably wasn’t a requirement for corporate, to now being quite certain about it.

I’ll be keeping an eye on it, but at the moment it’s a bit like being the only person around with a Fax machine.  Cool technology, but not a lot of point!

Oh and to the Plazes team if they read this — keep it up guys, it’s a great idea and is touching on what I think will be a successful area.  I just don’t get it yet and no doubt am in the wrong demographic.

I came across two interesting sites by David Troy, http://flickrvision.com/ and http://twittervision.com.

 They are both elegantly constructed mashups that automatically match location from Flickr and Twitter to maps.  Apart from the pure simplicity, what struck me yet again was how this kind of location based information adds real depth to what is otherwise a photo board or shared SMS service.

Twitter takes on a whole new dimension when “Making my coffee” pings up on the map as day breaks in the US, and watching photo’s appear on the map from Flickr was surprisingly enjoyable.

Of course, the cynic in me is still not sure what practical use they serve, but they are both excellent sites and a worth spending 10 minutes to experience.

I’ve touched on location based services in a few articles on AGPS and Plazes.  All which has me thinking more and more about logical extensions for this — how do we get location “off the map” and into a real application?  I was pointed to GeoRSS which adds geospatial support to RSS posts which allows you to know WHERE a post is relevant to, amongst other things.

This is leading to some interesting applications, if you haven’t seen it, check out World Explorer from Yahoo Research Berkely which analyses Flikr tags to produce a geotagged map of the world, weighted by the number of tags for a location – explore the world through the eyes of others in a way that makes sense.

So what should Enterprise Architects be thinking about?  In my overly lengthy post on AGPS (I’m going to try get better at more succinct ideas I promise!) I suggested you should be thinking about location type services, but I’ll go one further — you need a consistent global taxonomy for location in your organisation as it will become crucial.

Some 16 years ago, I worked for a fertilizer company, and we experimented with MapInfo to map sales data.  What we found was that based on the invoice address, most of our product was shipped into the centre of the city.  Now of course, this was explained as farmers either being owned by corporates, or having accountants who are located in the CBD.  Still, it wasn’t what we expected.  Even moving to delivery address didn’t help, as product was delivered to one of 6 major national distribution centres where farmers sent a truck to pick up product.

The point is simply this, location offers unique opportunities, many which I propose we don’t fully understand at this moment in time and others that will arrive.  To some extent we’ve been doing this in business for a while (hands up who has systems that know what office a staff member is located in), but we probably don’t understand this to the same degree of relevance about customers and in particular our own information.

I’m still remain sceptical that LBS will prove to be a ”killer app” for office based corporates, but I’ve no doubt that it will provide some incremental level of benefit.

I saw an article in the local free-paper on the train (MX), which spoke briefly about AGPS being launched by carriers in Australia in 2007.  There is some support for this in The Age which also suggests in an article there are plans for AGPS in  2007.  

If you are wondering about what AGPS, or Assisted GPS is, it’s essentially a better GPS service for mobile phones which takes advantage of the base station network to remove some processing power from the phone and provide GPS service even where you cannot get line of sight to the GPS sattelites.  iMobile has a great summary article here as well as coverage of some phones coming out with AGPS capability and more support that AGPS is on its way.

Stepping away from innovation and invention for a moment and back to the “true-purpose” of this blog (if it even has such a thing), I wanted to consider what some of the implications of AGPS might be in an office based business, if any.

It’s obvious that companies with traditional GPS based applications, like transport or logistics companies could benefit from AGPS, and from a consumer perspective, I think it may well be the next “killer app” on the mobile for any number of reasons, but what about corporate office users?

 No doubt, AGPS phones will eventually become the corporate norm, in the same way that despite limited business use and initial opposition, camera phones are more or less the standard corporate hand piece these days even when there are limited business applications for your average office worker.  This is a very legitimate strategy, even if only as a value add to your staff who will find “normal” uses for these things as the phone crosses equally between the boundary of personal and business use.

So what are some of the uses of AGPS technologies for different business sectors?  I’m going to ignore privacy concerns at the moment as undoubtably these would need to be resolved, or may in fact be prohibited.  Additionally I’m looking for applications where a “mobile” location is key — static information like “where is my store / customer” on a map can already be achieved.

1. Manufacturing: Potentially a use in occupational health and safety situations.  Personal GPS devices could identify where staff are located and act as an additional safety check to traditional tagging mechanisms.  While interesting, I doubt there is much in here to drive this use and phone have other distractions and use issues which should see them kept off the shop floor.
2. Retail: For retailers, there may be some usage in tracking store locations, shipments and other logistics type operations.  The biggest growth area here will be in sophisticated customer loyalty programs tied to stock control systems that can work magic like “a known customer is within 200 metres of my store, they have purchased x in the past, this top will match well and I’m over-stocked here — send them an SMS voucher to be redeemed in next 60 minutes”.  This is an area that I believe savvy retailers will look at very seriously.
3. Sales / Relationship businesses: I could envision a version of LinkedIn on steroids but I’m not sure how practical this would be — I don’t particular want my sales rep to know I’m walking down the street!  However, from a sales persons perspective, something linked to the CRM system that can alert them when they are within a set distance of customers, according to parameters, may be of use.
4. General Office use:  No killer apps that I can think off, but some location services may be of use.

Some of  the general location services that I think all types of users could use might include things like — who from my company is in this location?  Now this may seem a strange idea, but as someone who travels a fair bit back and forward to Sydney, it’s a reasonable bet that there is at least one other (and possibly many more) from my work in the queue for a cab at the airport.  If I could easily locate them, we could share a cab, reduce some green-house gas, save some money and more importantly network.

Similarly, I’m often in a hotel where I know there are other people staying — it might be nice to share dinner or something similar.

Touchstone has got me thinking a lot about attention based services as well, and I think that there is some merit in considering location as part of an attention profile.  This might mean that when I’m travelling for example, I get some local news.  If I could locate you to a floor in a building then I could SMS you as you enter the floor about things that need your attention now.  In effect an event trigger system, better known as Location Based Services (LBS).

LBS uses could be very wide ranging, from letting people know I’m at my PC (because I’m in my office), on my mobile, or just triggering a notification at a time I can do something about it, LBS offers opportunities for managing attention that broadcasting information may not.

From an architectural point of view, the best place to start would be with JSR179 which is implemented by both Nokia and Blackberry as well as a host of others.  This lengthy standard gives a great deal of information about what a LBS is capable of providing to a phone based application.

In researching this, I found an excellent summary here which talks about several different types of location services:

1) Location blind services – users manually enter their location.
2) Location aware services – location determined automatically.
3) Location precise services – accurate location, and a proactive system.
Geoff Hendrey. Managing the wireless internet. RF Design, March 2001.

We are now entering the first generation of level 3, with AGPS enabling accurate location.  I’m still searching for information on how the pro-active nature will work — while a system based of a JSR179  application on your phone might work, it’s not pro-active enough for real event based LBS which need to work without an application required to be useful.  Once I find some more I’ll post it here.

In summary, AGPS is coming, it’s not here yet, but there for some businesses it will bring significant advantage if implemented well.  For others, namely office based businesses, it may provide some interesting niche applications, but no real business drivers yet.  This isn’t to say something won’t happen, I believe in the war for talent that business is increasingly driven by the expectations of it’s staff, not just ROI in the traditional sense.

As Architects, particularly in retail businesses understanding how AGPS works, what location information is needed to tie the system together (is it just a mobile number) will be critical steps that can be taken now to ensure early mover advantage when it is launched for consumers later this year.

I found this Wall Street Journal video here at The Lost Remote.  One of the things I’ve been working on at the moment is a retail sector publication on trends and issues for 2007.  This video sums up succinctly the message I’ve been trying (like many others) to get across - the world of communication (as in Internet and Mobile Phones) is fundamentally changing.

Apart from being an entertaining little video, it issues some real challenges which is to get our heads around some of these types of features.  Particularly for retail (ie consumer focussed sectors) understanding where the internet and mobile technology is going will be key to early mover advantage and future success.  Taking advantage of the application driven nature of the Web 2.0 and plugging services together with proprietary systems is the compelling challenge for companies today — even if they don’t realise it!

From an Enterprise Architecture point of view, an SOA view of the world is becoming even more critical.  It’s all a service, wether it’s an internal or external application, the boundaries are changing.  With a strong services focus internally, then mashups between internal and external apps become compelling value adds for consumers and the companies that serve them.

Here’s a simple example, using Plazes, another interesting startup I’ve come across recently.    There’s a similar service shown in the video.

Plazes is at first glance locating you on a map. Kind of neat, but when you explore it under the hood a bit you realise it’s also a locater service for friends and others in your area.  For a retailer, the scenario could play out like this.  Set up Plazes, place your retail outlets, let your customers hook up as friends then use the Plazers service to let you know when your friends (ie. customers) are near your location.  This kind of location mashup could be put together quickly with a prior investment in SOA internally, enabling you to then link the friends to your backend systems, work out what they’ve purchased recently then contact them to tell them that there is a special offer available now if they come into the store.

What the Web 2.0 evolution is doing is removing the requirement for us to wait for big vendors or vertical applications.  Enterprises should be acting now to ensure that they are ready to grab the best of the web and put it together in innovative ways for their customers. 

That means sound SOA strategy so that agile developers can pull together the systems they need to create compelling applications.