The [GIS]: Desktop Concept is one of making a transparent easy way to get geographical data as well as uploading data that is meaningful to the user. GIS data is currently data which is normally easy to get in small amounts but the companies who collect it make it very expensive to make use of the data. Desktop users would be able to make a lot of good use of local data generated by like minded people.


First of all this would be a great selling point for OpenSource and OpenSource Desktops in particular. Secondly it would be a good basis for expansion for GIS data gathering. If an application/framework can be simple enough for most average people to make use of it then it's likely that other parties would be very interested in using it or leveraging the datasets created.

Use cases

The simplest example of how this can be used is the obvious resturant scenario. Someone needs to find a resturant and can't be bothered to look it up. They type in the name and since local searches are default it pulls up an overlay with the information on the resturant and where it is. The person clicks to relate it to their current location (stored locally) and gets a little readout. Other more complex searches might involve cell phone coverage or market appreciation.

* Jimmy wants to keep a track of Real Estate in his area so he makes notes of places and keeps a track of them on the GIS Desktop


Really it could be any data that people could be interested in as long as it protects the privacy of people. There should be some review process to allow spot specific data so that vandals or trouble makers do not post information about individuals that could be used to harass them.


There are at least three sections

1. The user interface of course.

2. The backend store and filtering system

3. The Data that is being transfered.


Still pending. The working out of the data structure may trump any other design considerations. As near as I have come is that there must be

* a standard data format that can easily package the information and be stitched together to have a queryable(sp?) set of related layers. As the system grows more complex relations can tested and exposed.

* An agreed framework that if possible has distributed (centralized) storage and distributed (shared) bandwidth would make the system scaleable. Of course the management server should be done so that anyone can create their own system with restrictions on what type of data can be flagged.

* A front end that can optionally come with basic local maps (as high as resolution as possible) or world maps. The user should be able to mark areas both as a spot location or an area with optional notes. These areas should be categorized with one or more flags. The categories can be hierarchal with the system understanding that some low level flag implies the ones further up the tree. It would also allow allow queries based upon categories and user input.


Data preservation and migration

Outstanding issues

The most Outstanding Issue is creating a Framework to get data from ordinary people in such a manner that it is useful when compiled with the data of other people who are not in collaboration, allowing flexibility for people to put in any information that they find interesting and still keeping private data out of the system.

BoF agenda and discussion

==Notes to Note==


GISDesktop (last edited 2008-08-06 16:23:26 by localhost)