Why NakdReality?

NakdReality is for entrepreneurs and developers who are frustrated with the time, cost, and high level of expertise required to develop their own augmented reality or location search.

NakdReality is an instant-on location search and augmented reality provider which allows anyone to quickly and easily create their own augmented reality or location-based search, without having to hire specialist developers.

What do you mean by "augmented reality"?

By "augmented reality" we mean overlaying location aware information onto a display of the real-world. Currently the most popular format for this is displaying hyperlocal data on the camera of a GPS enabled smartphone such as the iPhone 3GS or Android.

An example application running on Nakdreality is the THUNDRE product finder which finds products in nearby shops.

What is an augmented reality search service?

NakdReality's service is to configure, host, and maintain an augmented reality search-engine for you.

"Augmented reality search" is the selection or filtering of data to display, whether based on keywords or categories. Without a search engine, all your data will be displayed, which can result in users seeing irrelevant or extremely cluttered data.

Search for augmented reality consists of two components:

location search-engine

augmented reality interface

We currently use three different variations of location search-engine, but underlying each is a highly scalable spatial index that features two fractal space-filling curves (Also called "Peano" codes - named after the 19th centure Italian mathematician Giuseppe Peano who discovered them) that allow category or keyword search to provide constant lookup times irrespective of data quantity.

Our augmented reality interface of choice at this time is via the Layar Reality Browser. However we are ready to adapt to other augmented reality browsers or custom applications of your choice.

Will my data be locked-in to this service?

No. You are free to export your data. It's your data after all!

Where can I obtain your open source software?

Our Points of Interest Engine, which we also refer to as "The LightRod Engine", can be downloaded, with all the information you need to install it on your own server, from LightRod.org

Which search options are available?

Currently we provide two different search engines:

POI Engine (Point Of Interest Search)

This 'point of interest' engine does not filter results in any way, and simply displays all nearby results. It features a "spatial index" based on a pair of fractal, space-filling curves. To test-drive this engine, enter any data you want using our test-bed and see your data added in real time.

KISS Engine (Keyword Indexed Spatial Search)

Features the ability to search by keywords to find results based on their title or description. This sophisticated engine also offers the ability to create "venues" and enter different pieces of searchable information inside each venue. It additionally features category and person search using Twitter inspired # and @ symbols. The spatial index is based on a pair of fractal space-filling curves.

An example of a site using this engine is the local product finder THUNDRE which has over a billion spatially indexed keywords to keep track of. Notice how it has been set up with many spatially located products inside the same shops.

How do I upload my data?

We have an API for uploading data for real-time indexing. If your data is exceptionally large we can make other arrangements.

Your data must be 'indexed' to become visible to the search engine. There are two types of indexer programs available to operate on your data: a real-time indexer which is typically slowed to avoid any performance issues for users performing searches (but let us know if you need to remove this delay); and a non-real-time indexer which operates at full speed. Both indexers can be operating at the same time.

The non-real-time indexer operates separately to the search itself. Once the non-real-time indexer has finished indexing, its index will be switched on. On the other hand the real-time indexer applies changes directly to the live search index, so they are visible immediately.

If you have any sizeable quantity of data you will likely need to wait for the non-real-time indexer to finish indexing your data before launching.

What format should my data be in?

We recommend you supply data in XML or tab-delimited text format, but also handle CSV text format. If you would like us to handle some other import data format let us know, and we can make arrangements.

We will supply the specific fields for the appropriate engine after signup.

How do I launch my search?

Once your data has been uploaded and indexed, your search will be ready. At this point you can choose to use our pre-built templates for mobile and internet interfaces, as well as the Layar augmented reality interface; or build your own interface on the location search engine.

What formats do you output?

The export formats of our location search engines are XML and the JSON Layar format. However we also have templating systems designed to output HTML for internet and mobile web apps.

Do you offer real-time indexing?

Yes. You can update the search-engine while it is live.

How scalable is this service?

An example use of our keyword engine is the THUNDRE product finder which has more than one billion spatially indexed keywords. The spatial indexes we use offer a constant performance irrespective of data size or search radius. If your index is larger than this it will be split across multiple disks. Increased traffic will be handled by increasing the number of machines performing searches of your data.

How reliable is this service?

We will use Amazon's EC2 cloud computing service to host your search, providing a high level of reliability.

What is the process of submitting my augmented reality search to Layar?

We will provide you with step-by-step instructions. Layar vet each "layer" of data added to their service, so it can take several days to publish your layer depending on any issues they raise.

How international is your service?

Geographically speaking our engine is based on global latitude and longitude coordinates so isn't limited by geography. We don't directly supply geocoding services, with the exception of US zipcode or UK postcode lookup, but can interact with a third-party's service such as Google's.

Regarding internationalisation, our engines are based on a UTF-8 character set, and can both index and search for internationalised words.

If your front-end search interfaces require languages other than english we expect you to either handle the translations or pay us to do so.