WebCom OSGeo Site Focus
This page is intended as a collaborative brainstorming effort in defining how the web site should be designed, and is intended to be based on the WHO developed in coordination with VisCom VisComTargets.
OSGeo Web Site Focus (WebCom)
== Target
Who *does* want to talk to us?
=== Users (Got Feature?)
"George"
ESRI Proffessional getting sick of status quo...
Senario: Told by a manager to look into what is going on?
Recomendations:
- Need to speak in terminology accessable to the target audience. Diagram of ESRI Archtiecture --> FOSS Architecture.
- Page 1: Starts on the OSGEO home page (that his manager sent him from our latest press release)
- Page 2: Will click on the AUTODESK link right away as that is a named brand he remembers from the 80s
- Page 3: Sigh I don't know how to get this guy to the information he needs :-(
Here is the problem "George" just wants to know *what* to use, selection of components is a real problem here and we need to have some "certified" software stacks that he can try out, or write a memo on (ie we should evaulate this stuff that I read works together).
Would be better politically if we can ask some of the members or commercial providers to cough up a CD that can run on windows
Constraints:
- We loose points for linux distributions (sorry but true)
- Not interested in open standards (yet), that is part of a long play against vendor lock-in that George's manager is trying to step out of.
- George wants to see a feature list (aka answer the question what can I not do if I choose FOSS)
- Needs to see product set he knows, compared to what we got
- it ** will do us no good to talk about the benifits yet (cause George does not know what an SLD document is).
George wants a definitive "yes" or "no". Yes = we are ready for biz and we just stole ESRI's lunch money.
"Lui" GIS Professional looking for cheap way to try out open standards
- Once again start with the big picture, this time focus on interactions between components
- We are allowed to list the specificaitons (they probably did a web search to find us)
- Expected workflow:
- Page 1 - Google hit centered around specificaiton (ie SLD), need a pretty picture or diagram so he knows that a) we got our act together about the SLD spec b) we can make it sing
- Page 2 - Clicks on a "product" page of one of the programs that does SLD (ie GeoServer SLD page)
- Page 3 - If Lui lasts this long it better be a demo that shows an SLD file hitting a live geoserver
- Page 4 - Off to the GeoServer site to sign up for mailing lists, and get to know the gang.
User Expectations: - Video flash via wink (ie Show don't tell) - Expect each and every page to be "home", people arrive here via google and will not navigate the site, each page thus has to make obvious the story - we have at most two pages (clicks) to show them a pretty picture and give them what they came for - it is very hard not to talk down here, we have a different set of technocati terf staked out - each with our own terms - we are both right but we need to speak the correct langauge.
=== Developers (Got Code?)
"Sarah" Undergrad looking to have her homework done:
- she can check out our toolkits (and even read the source code looking for the answer)
- her project is in a specific lanaguage, or against a specific file
- We can be blunt about asking her to consider our projects for graduate student work
- many projects (like uDig) have additional facilities that are open to academics
And what to they want :-)
- Simple functional site, see jakarta
- No flash, it wastes time
- Want to find the right *project*
- These people *will* arrive via web search, there is no way they would start from our home page...
Structure Ideas
- Need to be user-focussed.
- New users? (What can I do with OSGeo projects, What is OSGeo, etc?)
Feature Ideas
- Segmentation:
- Education
- Government (small, large)
- Service Provider
- Solution Developer
- Convincers:
- Community profiles (profiles of individuals activel in open source geo)
- Project profiles (with lots of pretty screenshots)
- Sucess stories (these are big; need to show sucessful implementations)