Posted by jive
on July 2, 2005 at 6:21 PM PDT
I was in SanFran and did not go to Java One - what could possibly be so important?
Where 2.0 was really about one thing and one thing only - making lots
of money from maps.
There are three significant things going on:
#1 Google released an AJAX API
#2 Where 2.0 is the Question? Address is the answer
#3 The preception of how much data should cost is going down
How does this relate to Java? Well I am involved in the top three open source Java GIS Projects.
is the second conference I have been to since udig is has been released. And I actually had a udig user recognize the shirt and come say thanks (she also happened to be one of the only women at the conference but that is that). Number one request in this conference is Mac support.
I went low profile the first day by wearing a suite (so I could blend in). The second day was a tshirt so I could meet the "hackers".
Google released an AJAX API
This is the next step of the web, first thing I can honestly say is Web 2.0. What this gives us is a decent web component that is hackable. You can see a few others (like Flickr - which was used in every 3rd demo by hackers).
The commitment of a major web player to an API at this level is actually a big thing. It will be remembered long after maps are so absorbed by the internet that they lack interest.
Ideas: Make a Kamap Flicker demo
Ideas: If this had been done as a calendar app it would still be signigicant - and the conference would be called When 2.0.
Concern: It is not at all clear that anthing that does not go through AJAX is welcome on web 2.0.
Where 2.0 is the Question? Address is the answer
If the answer was place, or route (polygon or linestring) this train would of ground to a halt. This is not a GIS community it is a two other things - both of which are interesting.
Where 2.0 is hackers: one of the few people to get applause when they steped up to the podium was the hacker that combined craiglists with google maps. This is *why* everyone showed up at a conference about location technologies.
In short the fact that hackers "had an itch to scratch" and did something about it is exactly thing that everyone was excited about. The hackers were excited about it because it was possible. The spawning industry is excited because it means people will make up their own super specific services, long-tail was mentioned so many times it turned into a drinking game.
Where 2.0 is data miners: one of the only things that you "discover" when you do data mining is location. Seriously.
Location is often the strongest corralation of any data you can find. The companies "giving away data" claim they don't know how they can recover costs. It is very obvious that they do infact know, and the answer is golden.
Location represents the keys to the data mining kingdom. In the large, if location is known, you can adjust your sales figures for population and spots trends. GNat (the conference organizer) almost wept with joy when he finally relized that he could make sense of book sales. It took James McGill to finally explained this to him.
In the small location represents they key to know darn well most things about you (there is a reason it is on all the register your product cards). For your buying patterns they found that people were willing to trade airmiles. For location they found people are willing to trade driving directions.
And they all hope the hackers find the next thing people are willing to trade. Or if they don't they will create a swarm of services leading people towards the same AJAX interface.
Ideas: Why wasn't OGC here? If there was every a place they need to be it was "where". Several hold overs from the OSG2005 conference put in a good showing.
Ideas: I explained what a Capabilities file was to each of the big three, the deal is the first engine to crawl this stuff get a community of GIS types behind them.
Ideas: We had to call off our Java BOF (chances were that Java developers were elsewhere as Java One going on - to the point that billboards were up advertising to Java developers *everywhere*).
What was needed was an open standards talk. We did fairly well, but what the location community could not understand was that there was no standard for *address*. They got a bit demoralized when they relized what a hard problem it actually is, however VCard does have structured location information in it.
Ideas: Make a WFS that serves up VCard as a format
Ideas: Hook up uDig to an address locator (have done - the code is in my laptop).
The preception of how much data should cost is going down
GIS customers for data are not willing to pay as much, this has the people drive roads scared. We all know that that the value is in the attribtues (apparently 50 groups are driving the US roads collecting different bodies of attribution). This is what turns something from a pretty picture into something we can reason with.
But google being able to "give" away the information (or at least the appearance of the information) is causing a reset that we should even feel in the GIS community.