Jalaloedien has committed the RSS stuff he has been working on to trunk in SVN. It’s pretty cool stuff! Currently, as a test for the system, the last transactions of documents you subscribe to are available via an RSS feed and there is an RSS client dashlet for KnowledgeTree that you can use to view it - or any external RSS feed for that matter . Soon it will be extended so all documents and folders and even saved searches can be “watched” via your favorite RSS client or the new RSS client dashlet in KnowledgeTree. I am trying to get him to blog some of the details for us.

In the meantime here are the obligatory screenshots…

Remember you can always checkout the latest unstable development code from our SVN repository on SourceForge…

We try really hard to keep the trunk and all branches in SVN working all the time and all code is reviewed before committing it to SVN. So if you want access to the “bleeding edge” of KnowledgeTree, our SVN repository is the place to get it!

For those that can’t use SVN for whatever reason we will be creating a snapshot source file for download from SourceForge every Friday - given sufficient time and code changes to warrant one . I will post a message on the forums when I have more details.

Our new Continuous Integration (CI) environment, that I am currently busy with, will soon (yes soon) provide nightly builds for download and go a long way to helping us integrate all the code contributions and patches we get from you in a timely manner.

So jump in and help make KnowledgeTree all it can be!

Thanks
Kevin

PLEASE NOTE: This is prototype work and not yet on the release roadmap. Don’t take anything said here as hard and fast or even remotely true. I am only speaking for myself and the whole team must be involved in any decisions. Some features mentioned here may never make it into production. Some may not even be possible. These could very well be the ravings of a lunatic. With all that said, on to the ravings…

I have spend quite a bit of time getting back to my C++ roots and wrapping unmanaged Interfaces and COM interop objects in a fairly generic managed code Shell Extension library. Then I re-wrote it several times just for good measure and probably will do so again. I do not intend going into any detail on that process. Suffice it to say that that territory is dark and mysterious, sparsely documented and there should be a dedicated Saint to watch over the travelers in that region as Saint Christopher watches over the travels of others. Okay, perhaps a slight exaggeration - but only slight.

The first topic I will look at is Namespaces. You can click on the thumbnail below to see a larger image of where Namespaces fit in.

• Namespaces

What is a Namespace? From MSDN: “Microsoft Windows Explorer provides a graphical representation of The Shell Namespace combined with tools that allow users to interact with Shell objects. With a namespace extension, you can take any body of data and have Windows Explorer present it to the user as a virtual folder. When a user browses into this folder, your data is presented as a tree-structured hierarchy of folders and files, much like the rest of the Shell namespace. Users and applications are able to interact with the contents of this virtual folder in much the same way as with any other namespace object.”

“Wow, that sounds great! The body of data thingy as a virtual thingy, that’s just great! I have no idea what that means, but it sounds great!” Okay, let me show you a few “thousand word” pictures then. Click on the thumbnails below for examples of a KnowledgeTree Namespace in action…

As you can see from the above images KnowledgeTree is “integrated” into the Windows Explorer as a virtual folder similar to a network drive, but oh so much more powerful (as we will see in our future Shell Integration adventures). All the Windows Explorer functionality you are used to like drag ‘n drop, right-click context menus, views, etc. become available to KnowledgeTree resources. The KnowledgeTree virtual folder even shows up in a Word Save/Open dialog. This allows you to choose a location on a KnowledgeTree server to store or open a document just as easily as a location on a local hard drive. In fact, any application that has the ability to open or save documents is provided automatically with the ability to do so using a KnowledgeTree resource. Even other shell extensions can work from the existing context menu. Select a KnowledgeTree folder, right-click and “7-Zip -> Add to Archive…”, it’s all possible.

“Fantastic! I can easily open and save *new* KnowledgeTree resources from pretty much any application, but what about the documents I retrieve from the KnowledgeTree resource to my local file system? They are marked as checked out in that virtual folder thingy.” Ah ha! Someone’s been paying attention. Although we are jumping the gun slightly, now is probably a good time to introduce the “DotKT” or “.kt” concept. DotKT does it’s level best to look after your KnowledgeTree documents while they are away from their KnowledgeTree server home, so to speak. DotKT is actually a pretty broad topic but I will only touch on some of the Windows aspects for now. While the Namespace concept relates well to the relative safety of the “server side” of things, on the “client side”, in the wild, things are, well, largely wild. There is currently very little “supervision” of a document once it leaves a KnowledgeTree server (i.e. it’s checked out) and finds itself on a users local file system. Using more shell magic, DotKT under Windows endeavors to keep track of checked out documents on the client side. The Namespace and DotKT interact on the KnowledgeTree CheckIn/CheckOut boundry where they leverage what the other does best. Namespace does server best and DotKT does client best. They both know enough about each other to successfully “handoff” a document to the other at the CheckIn/CheckOut boundry.

Enough! The idea was to introduce Namespaces and I hope that I have at least whet your appetite for more information about the power they can bring to KnowledgeTree Tools.

Until next time then…

I have always had a disturbing (to others) obsession with technology and as a youngster dreamed of being chosen for a voyage of discovery to the stars and beyond - quite happy that it would probably be a one way trip. By 12 I had written quite a few programs and had built up a large library of code. Yes, even then I was into code reuse. I did not know it was good design, just that I was too lazy to keep rewriting it. My mentor at that time was a UCT CompSci student friend of my brother. After she told me they where too difficult for me, C and assembler quickly become my favorite languages. I would beg, borrow and steal her lecture notes and textbooks to feed my ever increasing appetite for coding knowledge. Looking back now, I see I was a total control freak! I loved being the “creator” that made “stuff do stuff”! This is probably what led to my first big dev gig. “Develop a system that controls the fire systems, environmental systems, lighting systems and access controls for the new Head Office of one of the largest insurance companies in South Africa” they said. “I’m in heaven!” I said. Since then I have been involved in several large development projects ranging from a couple of months to several years. The majority of them have even been successful!

Today I find myself in a very enviable position. I work for a great company; with great people; on a great product; in the most beautiful city in the world! I truly believe that with this much going for it KnowledgeTree is only going to go from strength to strength. I hope that during this journey I can share some useful information and give back to the community that has helped me so much over the years.

Thanks
Kevin

If it moves, recompile it!