ASP.net is not ajax exclusive btw, it’s just a platform for serving http content. The MVC site the SVD is based upon makes good use of Ajax (via jQuery).
SVD is generic, it’s intended to run under it’s own domain which we’ve already bought (stolenvehicledatabase.com and .org). It’ll list all vehicles types to attract as much use as possible and be internationally focused.
The application and database is already done and the website framework, what’s left is the UI itself. The Google Maps stuff is critical and what I’d love to collaborate with you on. That and the data-warehousing side as well!
We’ll be hosting it on our uber-servers as well, so performance and capacity will never be a problem.
This is going to be a popular system, and we need good data warehousing capabilities, which is why I’ve chosen SQL Server 2008 R2, which is also what we use as our primary db platform here at LB.
I know. What I meant is that my stuff isn’t using ASP.NET (as in pages, code behind etc), as everything is done through AJAX (google libraries suplemented with jQuery), with simple handlers serving JSON (and other export feeds like RSS), so no MVC needed or anything. Much cleaner architecture for something simple like this.
Practically, I’m deploying .html pages and dlls, no asp.net pages.
Funny, I was looking at those the other week, but decided on stolenDb.org (not bought yet) as I didn’t really want to limit it to bikes or even vehicles. To start with, it would just be bikes.stolendb.org …
Happy to collaborate on the maps stuff, but what I’m missing is a designers input on the UI.
Honestly, have a look at Google’s Fusiontables for all the public data, it’s a must of large maps integration and really really good.
As for data storage, I was going to go for MongoDb as a document db (no-sql) is great for this stuff. It will be much faster and easier to support changing model (no schema). For example, I’m to add loads of data capturing to help the inteligence, like security etc. For the UI, this doesn’t need to be structured as such, so a document db is ideal, then a seperate process will take it into the warehouse for reporting.
Happy to go through this and chat, I’ve done a lot of thinking about this.
Wow jay/driesie your words have lost me but sounds cooool:D
failing that the low tech mad-dog method. Bait bike, van, rope & lube:w00t: that way they off the street and my sexy life doubles:cool:
Drisie, Fusion Tables do indeed look very promising, though they’re not usable for large projects right now I fear (100mb limit per table), and then there’s issues over data ownership which causes problems with the UK DPA.
The visualisation aspect is very interesting, as well as the collaboration factor. Perhaps it would be possible to link up to it later-on via exports.
Where do you get the 100Mb limit from? I don’t think that’s true.
I’m not worried about the data ownership issues, we’re only talking about storing the data that is visible on the website, the rest get stored in an internal DB. So the user will need to give permission to publish the data it’s giving.
I really believe this should be a community project, everything I write will be open source, and it will be possible to take a feed of all the data.
My plan was to get a basic version to work (nearly there), then release all the code (probably google code or codeplex), so everybody who wants can improve it. Each stable version will be installed on StolenDb, so people who want to contribute improvements will just be able to.
Drisie, what you’re doing is great, the architecture is sound. I’m intrigued to see where you take it, but am dissappointed there’s a duplication of effort here. Perhaps there will be room for a mash-up later.
Jay, let’s talk.
As I said before, I don’t want to “compete”. I’m happy to join up efforts, but I have a couple of principles I’m not going to be stepping away from:
All code in the front-end and the back-end integration must be open source
It’s a true community project and anybody can take a feed of the data and re-publish it (I have no problem with forcing consumers to credit the source though)
The reason for this is that I think the project will only work if it’s “open” and transparent. The reason why I think it should be open source is that people would be able to write add-ons, improvements or variations and contribute this to make it a better project.
If you can agree with that, I’m happy to integrate all the google maps/ui stuff with what you have, and I’m happy to build a warehouse/intelligence solution from your database, as long as I can have some input in what sort of data to collect (that’s what is going to make the trend analysis etc work or not).
If you like to take this “off-line”, send me a PM with your details, or we could meet up some time to discuss over coffee or beer and I can show you what I have so far.
I’ve got the code ready for a first release. The last thing to do before putting it live is to give it a quick once over design-wise. I’m quite happy with where it’s at, but it’ll only have the basic functionality, the more advanced features will come after.
