We’ve implemented more ArcIMS sites than I can recall over the past few years. The chief developer on my team has probably as much AXL experience as anyone I know. But times change and clients have different expectations than they did in 1999 or even 2006.
“Because you didn’t come here to make the choice, you’ve already made it. You’re here to try to understand why you made it.”
On one our latest projects, we are developing a site using SharePoint and what would have been ArcIMS. The difficulty we’ve had getting the WebADF to work the way we want has put a damper on our excitement that we had when it first showed up. It is just way too difficult for the “simple” and “quick” development that we have to do. Plus, as nice as the WebADF is, clients still are expecting a Google Maps type interface and the WebADF is not that (nor does it try to be). With the new licensing of Virtual Earth, we’ve decided that MapDotNet Server 2007 connecting to ArcSDE is the way to go. The front end will be based on Virtual Earth so folks will feel comfortable with the interface (it seems every ArcIMS web mapping site is different and that causes usability issues) and they’ll be able to work with the data rather than fighting the interface.
Now this isn’t to say that we are abandoning ESRI server products because that isn’t the case. We will continue to develop ArcIMS sites for clients who want them (I still say it makes sense to leverage existing licenses of ArcIMS or move over to ArcGIS Server Standard) than dump all that work and start new and ArcGIS Server applications for clients whose requirements need Geoprocessing. But for quick and simple web mapping I think MapDotNet and Virtual Earth will be the killer combination for ESRI .NET developers who are already familiar with the tools. The simple fact that folks won’t have to abandon any of their existing ESRI servers (ArcSDE is still there) and desktop clients, the ease of which we can develop applications and the speed of MapDotNet will give our clients that quick, easy to use, great looking web mapping tool that they have been clamoring for.
Maybe the ESRI REST API will change things for us (I wasn’t at the 2007 UC so I have no idea how it looks or works), but for now we are going in this direction. We’ll see what the 2008 Developer Summit brings for the REST API and rapid development of ESRI web mapping applications.
I find it interesting to see another ESRI developer look outside the ESRI stable for a replacement to MapObjects. There was some concern among many developers at the 2007 Dev Summit that ESRI was abandoning the smaller developers and focusing on enterprise level GIS tools. Steve, who posted on his blog about .NET SIG at the Developer Summit wrote:
Damian [Spangrud] talking ‘ discussion about pricing ‘ tension between large enterprise customers who expect it to cost more and smaller shops that think it is too much (like me).
That just scares me working with ESRI server software. I feel like I’m being priced out of the marketplace with their new tools. The days of writing simple and cheap Avenue or MapObjects application are over. Now you need superstars who know ArcObjects in and out and clients where price is no option. Maybe the RESI API will change this (or maybe not), but if you look around there are tons of tools available for you to use that won’t mess with your workflows and might just allow you to improve you output without spending tens of thousands of dollars.
So we’ll see where this all leads. We still may decide that MapDotNet isn’t for us and go back to trying to figure out the WebADF and its task framework.