I can say that when I decided to go to Mix this year, my main goal was to meet some of the folks involved with WPF and to learn more about what was called WPF/E. I suspected something was up however when Microsoft announced Silverlight a few weeks ago, knowing they were planning something much larger. So in the keynote they dropped the bomb that Silverlight will include the entire .NET runtime. It was actually ironic because I was talking to someone from Adobe (formally with Macromedia) before the keynote and I asked him what he thought about Silverlight as it related to Flash and he just commented that it didn't include any type of engine. I bet he as well as the rest of Adobe dropped their own bomb when they heard that news.

So what are the advantages of .NET in the browser? Listening to the keynote, the big things I can see are faster execution, better controls, better DOM and the fact that they are including LINQ! Yes, LINQ in the browser. Scott Guthrie even did a quick demo on a chess game he wrote that pits the .NET engine against the Javascript engine in Silverlight. It showed how many nodes per second were processed by each and it became obvious how much faster .NET will be that JS. Hey Brett, just think, drawing the freehand paths may actually be acceptable now!!!

Another cool announcement was the Silverlight Streaming Alpha, which is a web site you can upload videos to (up to 4GB total) and get a direct URL back for viewing on any web site, which can be easily used in Silverlight applications. Sounds like Microsoft's YouTube to me, but either way combining this with the great media abilities of Silverlight, web sites should be better than ever. Check out http://silverlight.live.com.

With keynotes taking up most of the morning, the afternoon was left to breakout sessions. The first one I attended was Silverlight and Javascript, but it was way too crowded and I got to sit between two guys bigger than myself. The three of us started producing way too much heat so I decided to call it early. I did get to see the presenter use the Javascript object model to access to access WPF entities built with Blend.

The next session I attended was something called Accessing Data Services in the Cloud. I didn't really understand what I was walking into but it was recommended to me. Ended up being a really cool breakout, as Microsoft announced something called "Astoria". I will not do any justice to this technology, so I'll just keep it brief Astoria exposes data as a service, allowing access via HTTP as the exchange. The coolest thing I saw it used for was with JSON, which allows the data to simply be exposed as objects inside of Javascript. There's also an online service available, but keep in mind "Astoria" is a new technology in just preview form. I have no idea of what type of product it will end up as. You can check out more information at http://astoria.mslivelabs.com.

I tried to go to more sessions, but both of the ones I went to were cramped and hot, and I just didn't feel like sitting through another 1 ½ of it. I'm heading to the Mashup tonight though, so the evening is still to come.

Chris

In mid May my Imaging Server team will release an update to the KnowledgeLake Imaging Server 2007. This release will include something not in the original version, which is the ability to scan directly from and into a SharePoint Document Library. For current users of KnowledgeLake Imaging, those with maintenance contracts actually, will have access to this new version that includes the scanning functionality. KnowledgeLake Scan will also be available as a standalone product for SharePoint servers sometime this summer.

Those of you familiar with KnowledgeLake's capture products know that we have always focused more on the production side of document capture, rather than the desktop capture user. With this in mind, it might a good time to flashback to where KnowledgeLake Capture evolved from. Unlike almost all of our other products, capture was not originally built for SharePoint. Back in 1998, I was working on a FileNet implementation at a large brokerage firm and FileNet was just releasing their new 32-bit (yep 32 whole bits) scanning application called Panagon Capture. To keep this short, let's just say that the release was an absolute disaster and the customer was not happy. In my complete arrogance, Ron Cameron suggested I use the Kofax Toolkit to write an application to scan and save to FileNet. I'm still not sure if he was serious or not, but I thought for sure that if I had a decent toolkit this couldn't possibly be that hard. After hiding in locked room for 30 days I emerged with an application that was simple, but did the job. The customer accepted it and we slid my application in instead of Panagon Capture.

Three years later, KnowledgeLake was born and we decided to start things off by building and selling something called "Tablerock", named after a lake here in Missouri. Taking what I learned from my first dive into the Kofax Toolkit, I improved on some things and we launched our first product back in 2001. We sold the application as a framework rather than as a package so our services organization could modify it accordingly for each implementation. This was especially important for indexing as it seemed every customer needed its own modifications. The product really came to life in 2004 when we starting building software for SharePoint. Changing the saving mechanism to allow for release to SharePoint, we launched the first production capture software for SharePoint. We also built some tweaks into SharePoint to allow document libraries to scale. The product was renamed KnowledgeLake Capture and we dropped the lake names from all of our products at that time.

Right before the SharePoint products were born we also built KnowledgeLake Desktop Capture (actually "Clearwater" at the time), as our sales team was finding more and more needs for desktop scanning software. I didn't get it at the time, mainly because of the FileNet focus, but this started to make a lot more sense as we moved into the world of SharePoint. For 2007 we actually merged the Capture and Desktop Capture products together, which was made possible by rewriting the application from the ground up and no longer around the Kofax Toolkit. Although we do offer compatibility to Kofax drivers via a specific adapter the application was no longer dependant on it. This allowed to also build Twain and WIA scanning adapters to meet the needs of both Desktop and Production scanning in one platform.

Seeing other companies take the approach of scanning directly from a web application, and determined to stay in the lead on ramping content to SharePoint, we knew we had to at minimum add the same abilities. A prototype was built months ago, but we just didn't have the time and resources to get it incorporated into a real product as our teams were very focused on releasing our products for the new SharePoint versions. This changed after February however and we used the knowledge from our 2007 capture product and took the time to build Scan directly into SharePoint. KnowledgeLake Scan (as it's known right now) is our first web based scanning application and I can say without doubt I think we provide a better interface than any other web based capture application. We minimized the button clicks it takes to scan a document and make it easy to index and save the images as PDF, XPS of TIFF files. We plan on taking this interface a lot further in the future adding web based scanning into our KnowledgeLake Capture Server product, and with that we will introduce some abilities that I believe will separate us from any competition.

I do have to give some major kudos to my teams on this as the Imaging and Capture teams combined efforts on this to build it together, making the core components usable to each team in the future. So guys if you are reading this, let me say thank you!

Mixing it up,

Chris