Ranking factors for Your Google Earth Video and Other Objects

Earlier this month, the ability to watch YouTube Videos In Google Earth was reported. The Google Maps User Guide also describes how videos can be added to Google Maps, and how those maps can be viewed in Google Earth.

A couple of new patent applications from Google look at how objects can be shared in Google Earth, how they might be ranked against each other, and how the very large size of the data base, holding different file types, might handle dynamic uploads of new information.

Google Earth indexing Challenges

The challenges behind Google Earth in indexing, updating, and searching the system exist because of:

Different file types — images, maps, and tables can all be involved in such a system.

Large size of system — When you cover a very large area, like the entire world, and include expansive amounts of information about different points in that area, there’s potentially a lot of topographic, demographic, and historical information that may need to be stored.

Graphical data — including 3D structures, can also be large and unwieldy.

The Google Earth Index

Content in Google Earth is stored in a multi-level hierarchical index having information about descendants at each level of the index. Arranging data like that allows items to be retrieved quickly, even when the number of stored items increases to a large number.

We’re also told that this layered structure can generally be updated dynamically as objects are added without a need for to timely rebalance the index tree.

This index setup may provide for one or more of the following features and/or advantages:

Objects within the system might be quickly located, even though associated with a very large geography such as the earth, and within a very large database of objects.
Bounding box queries can be made, where a smaller area can be searched based upon boundaries that a searcher defines in latitude and longitude space (generated by a graphical interface such as Google Earth or Google Maps).
The amount of items show can be puposefully limited, so that too many items aren’t shown at the same time.
Items can be ranked to show them in a preferred order.
Third party (non-google) geographic information providers could benefit by being able to more easily store information about geo-located objects, and may be able to readily search for and recall such objects for display with a related geographic area.
Dynamic updates to the index can happen without requiring a rebuild of the index, if set up right.

Choosing and Ranking Objects

The system uses layers can contain objects associated with the geography, such as 3D models of building, bridges, or other structures.

They can also include geographic data such as earthquake data, weather data, or the like. Because multiple servers are used, objects can be layered over the geography, so that viewers can look at different layers, and can interact with them easier.

Requests from users of the system can determine what information in geography database is needed by the user. That kind of information may include latitude and longitude and a bounding box that the user is interested in.

There’s an object server that contains an object database, which stores information about objects that could be associated with certain areas.

Information about the objects can be displayed in a geographic representation, along with hyperlinks to additional information.

A label, such as “Eiffel Tower models” could be displayed over a map of Paris. When someone selects the label, which could look like a hyperlink, they may be presented with one or more images showing various 3D models of the Eiffel Tower, such as in a pop-up window.

Those models could be ordered in various ways, such as by placing most popular models near the top or placing models from certain “trusted” object builders at the top. The number of user downloads or clickthroughs may be used to measure popularity.

Selecting one of the images could cause it to be placed on the geography shown.

Creating and Submitting Objects

People can submit objects to system, that they have created through programs such as SketchUp, AutoCAD, or 3DStudio, so that they can be displayed as part of the geographic information system.

They can provide information about the object at a web site associated with system, such as a URL where the object can be accessed so that others can choose to incorporate the object into geographies they are viewing.

Those creators can provide additional information, such as an example image of the object and labels for hyperlinks associated with the object. This might be done through an interactive web form or some other interface.

A user navigates to a geographic location, such as by entering a lat/long combination or an address, or by moving a graphical representation of the geographic area into view on a display. He or she can see some other objects or textual information for that location, and could possible put a placeholder pin in the location where the object should be stored.

Content uploaded to Google Earth may be subject to being viewed, used, and modified by other users without any intellectual property violations by the system or by others. Those users might be provided with a notice saying thta the uploaded content is subject to a Creative Commons license, so that others can use it.

When uploading objects, the user may be asked to add meta data related to their object, and an image of the object as well. They may also identify other users or classes of users who may have access to the model, and define levels of permissible access. While that might mean making an object “read only,” or “view only,” the system might restrict the ability of people to do that because it would not encourage later user interaction. does not encourage later user interaction.

Password Protected Objects

Creating a password and private identifier for an object would enable the user to make it available only to a select group of other users. For example, an architect could create a model for review only by a client at the location where it would actually appear. Emails could be sent out to the intended viewers, with a separate email sent out for a password.

Shape Indexing

Users of the system may also be able to search for particular shapes, or by other physical features of a model. For example, a model may be viewed as consisting of a number of “primitive” shapes such as spheres, 3D rectangles, toroids, and so on, and models may be composed of Boolean combinations of those.

That model might be associated with one or more of those shapes. A score might be assigned to the shapes used in the model, by looking at how many times a particular shape in the model, and how prominently, in terms of size, the shapes are when they appear.

Conclusion

The patent applications don’t specifically mention videos embedded in Google Earth, but they also don’t exclude them. They use 3D models as an example of how objects can be placed within this Geographic Information System (I like how they refer to Google Earth in that manner).

One of the underpinnings of the patent filings seem to be the desire that people interact with this system as much as possible, and create and add objects that could be modified by others. World building may work best if there are a lot of hands working together. While such modification is possible, they also provide the opportunity for people to look at, and work upon earlier versions of objects.

Some private use of models is anticipated, with the sharing of URLs and passwords in separate emails, though it does seem that the focus of this system is more upon a public usage, and a collaborative effort.

I’m encouraged everytime I visit the Google Earth Community by the community effort in creating a rich wealth of different maps and objects. Some people are working upon some pretty interesting maps, based upon different concepts and locations. If you haven’t spent any time there, and are interested in learning what’s possible, it’s worth spending time in that group.