Have you ever created a TFS 2010/2012 Build definition, and tried to make sure that unit tests were executing correctly, only to check it and notice that the Unit Tests were not quite executing, yet TFS Build server displayed it was. To make matters worse, the unit tests were never ending, and the TFS Build was sitting there waiting for the test to end. To top it off, these builds were “Gated” check in builds, which basically meant no one could check in their code successfully until the build completed successfully, which in my case wasn’t happening because the unit test never completed, heck they never truly started.

You tried everything and not once did it occur to you that you were running Visual Studio .Net 2012 unit tests with TFS Build Server 2010. TFS Build Server 2010 is designed to work with Visual Studio .Net 2010. Even though it will work woth VS.Net 2012, TFS Build Server 2010 knows nothing about VS.Net 2012, including it’s unit tests.

So… I searched… http://social.msdn.microsoft.com/Forums/en-US/winappswithcsharp/thread/ab53380d-36fd-40e5-8494-3cb9560578b5 and I asked http://social.msdn.microsoft.com/Forums/en-US/tfsbuild/thread/252f96cd-4b67-4075-ba82-57aa322f69ec

Luckily, one of the MS TFS Experts: John Qiao, pointed me in the right direction and gave me the answer…

First thing I had to do was to create a new Build Process template based on the default template, so that I would mess up any default settings and build definitions.

Next thing I had to do was modify the section which invokes the Unit Testing engine: MSTest.

Which was easy enough, just open up the Custom template I created, which just happens to be a WF 4.0. Navigate to the Run MSTest for Test xxx, activity, and modify it’s ToolPath property settings:

That was it… My Visual Studio 2012 Unit Tests were executing in TFS Build Server 2010 successfully… (BTW, I know my path is mis-spelled, when installing VS.Net it was manually configured and typed in wrong… I didn’t do it- promise J)

Today I take a break from the more technical underlining of development tasks to peek into the Infrastructure and Operations management side of things. This brief talk will explain how easy it is to take Microsoft System Center 2012 Service Manager, Orchestrator Runbooks, and Microsoft Azure Virtual Machines to integrate and automate provisioning of new VM instances.

Before I get into specifics, let’s discuss these topics from a high overview to make sure everyone understands the topic at hand.

Overview of System Center 2012 SP1

System Center 2012 SP1 is a suite of applications produced by Microsoft for managing enterprise servers, client machines, inventory and asset tracking, enterprise applications, request offerings, problem management and incident reports and many more features. System Center 2012 is comprised of Operations Manager, Virtual Machine Manager, AppController, Data Protection Manager, Configuration Manager, Service Manager, and Orchestrator.

Operations Manager is used to monitor servers, such as SQL database, BizTalk Server and IIS Web Servers. This application can monitor when these servers go offline and when errors occur on these servers. This application can also notify groups of individuals responsible for the health and maintenance of these servers.

Virtual Machine Manager is an application used to manage, create and monitor virtual machines created with Hyper-V. A virtual machine is a virtualized computer that is running in the memory process of a Host Hyper Visor application (Hyper-V). The Hyper Visor application divides up it’s hardware, memory and other assets to provide resources for the virtual machine image.

AppController is an application that allows companies to manage their virtual machines created in Virtual Machine Manager from s self provisioning perspective. A Web site can be configured to allow internal employees request and self provision virtual machines across a private internal network, or even the public network known now adays as the public cloud.

Data Protection Manager is an application that allows you to backup and restore virtual machine hard drives, databases, and applications stored on network drives and network shares. It helps to automate the backup and restore process of large files and systems.

Configuration Manager is an application that allows companies to keep track of all hardware, and versions of software installed in a network. It can also be used as an inventory and asset tracking repository.

System Center Service Manager 2012

Service Manager is an application that allows enterprises to manage activities, change, incidents, problems, release and service request items. It also allows you to create your own form of item management. These items, are usually referred to as work items and can consist of something called a request offering, which is simply where an IT department can offer various capabilities to other internal departments. These capabilities, referred to in System Center as “Services” can refer to anything from Purchasing new laptops, to fixing the office telephone. Typically, when these services are offered, there must be a cost associated with them and managed so that the services can be billed to the appropriate department requesting the service. This is where Service Manager comes in. This application allows enterprises to create one or more of these services, as it applies to your business, and create a Request form to expose to internal departments to request this capability. There is also a self service aspect to this application as well, similar to the AppController for Virtual Machines. The self service aspect allows employees to request a service from a Web site configured to host these self service forms.

Which leads us up to the purpose of this posting. Service Manager easily allows you to create forms to expose “Services” or capabilities which an IT Department offer to its co-workers and departments it must service. AppController and VMM, allows employees to create Virtual Machines, and provision them within their private network (Private Cloud) or external network (Public Cloud). With these three applications, there is no bridge which allows you provision Windows Azure Public resources, such as Mobile Services, Azure Virtual Machines, Azure Storage accounts, and Azure Storage containers, Azure Networks, and etc.

With the implementation of Public Cloud offerings such as Windows Azure, it only makes sense to support this with a powerful suite of applications within System Center 2012 sp1. In order to do this, we must configure and develop business logic that can automate workflow and call out to the Public Cloud api’s.

Overview System Center Orchestrator 2012

Enter the world of Orchestrator. System Center Orchestrator contains something called “Runbooks”. This is an application that allows developers to further extend System Center suite to call out into the external world of Public Cloud offerings such as Windows Azure. A runbook is a graphical designer that allows a developer to configure and create some very fancy and complex workflows. It can do things like send an email when a request for a new telephone line is generated, or even automatically provision a new Windows Azure Virtual Machine instance when a request is generated. For those BizTalk Dev’s out there, think of a Runbook as similar to a WF Workflow, or a BizTalk Orchestration, except that it has it’s own editor as opposed to being hooked inside of Visual Studio .Net.

Overview of Windows Azure

Windows Azure is Microsoft’s public cloud service offerings to everyone. It offers a multitude of capabilities. It supports Web Sites, Virtual Machines, Mobile Services, Cloud Services, SQL Databases, Storage Containers, BigData processing, Virtual Networks, SQL Reporting, Media Services, Active Directory and more.

When working with Windows Azure Virtual Machines, this is a service that Microsoft allows you to create your own virtualized computer and have it hosted and maintained in Windows Azure. This means if you have company, and you need to hire a contractor or employee to do some work, you can create a virtualized computer configured in your internal network to access your sensitive information. The Data never has to leave your controlled and monitored environment. It’s very easy to setup a server, or a workstation for virtualization. Below is a screenshot of a new VM wizard showing you just how easy it is.

How it all fits together to Automate Windows Azure Virtual Machines with System Center 2012 Service Manager

So now that you overstand what System Center 2012 SP1 is, and you overstand Windows Azure Virtual Machines, let’s discuss how we can integrate the 2 together to automate the Windows Azure wizard. First thing we need is a way that Service Manager to maintain and monitor our requests for provisioning a Windows Azure virtual machine.

Service Manager gives you the ability to create a Service Request item. The only catch is that the out of the box Service Request items does not contain any field entries for DNS Name, Image Size, Image Name, Password, Affinity Group or Locations as the Windows Azure Wizard supports. This means we will need to extend the Service Request item in Service Manager. Extending the Service Manger Service Request item requires that you create your own class and derive from the base class to extend it. This is a topic for another post, however I will way it very easy to do, and once it’s done you can create a custom form to display inside Service Manager to keep track of these special fields and more.

After you have created a custom form, you can use System Center Orchestrator to create a custom Runbook to retrieve the filled out entries in the Request form and automate the process for provisioning Windows Azure Virtual Machines. Below is an example of a runbook created to monitor the the Request Form being filled out and crated, and then invoking the Windows Azure Runbooks activities to provision the Windows Azure Virtual Machines.

In a future post I drill more deeper into how I created the Runbook, Custom Request Form, and extended the ServiceRequest Item in Service manager. Until then, happy reading…!

Just a quick note…

The Kinect for Windows SDK v1.7 is now available for download as of March 18^th, 2013…

You can get the download here. You can also read from a fellow MVP and colleague Tim Huckaby about it here

I’ll post more details about what you can do with it in the following days…

Let’s start ‘Kinecting’ Things Together Shall We?

As I start down the path of building my Kinect for Windows (K4W) Windows 8 Application, confusion starts to set in about why I can’t just create a .Net Framework based application in the programming languages of C# or Visual Basic.Net. After all, the last 11 years of Microsoft based computer programming has been rooted in the .Net Framework. We started with the .Net Framework 1.0, and today in the year 2013 we are now in the .Net Framework 4.5. The K4W SDK, and toolkit provide a very simple .Net Framework assembly: Microsoft.Kinect.dll. This assembly is a .Net Framework based assembly, which means as a developer you can quickly and easily build .Net Framework based applications to use the device’s capabilities. The interesting part about this is that this is the only assembly and dll Microsoft provides for interacting with the K4W device.

Question: You mean you can’t write a Kinect 4 Windows 8 application with the .Net Framework?

Well, yes and no. Windows 8 comes in multiple flavors. We have Windows 8 Pro, Windows 8 RT, and Windows 8 Embedded. Windows 8 Pro is the full edition that allows developers to create 2 different type of applications “Desktop” and “Windows Store”. As of right now, these application types are not interchangeable, meaning you can’t develop a “Desktop” which switches to a “Windows Store” type application. You could develop two applications that function alike, thereby giving the impression that the “Windows Store” app is essentially the same as the “Desktop” version, but this is easier said than done. “Windows Store” applications are a new paradigm in programming. The Design model is completely different, as well as the application programming interfaces (Api’s) and libraries used to implement it.

The “Desktop” type of application allows for backwards compatibility to older versions of Windows. This was your typical application that can run in Windows 7, Vista, and before. The “Desktop” type of application supports the full .Net Framework, version 2.0 all the way up to version 4.5. The K4W assembly was written with the .Net Framework 2.0 base, and refreshed to work with the .Net Framework 4.5. As a developer you can take full advantage of this assembly inside your applications.

On the other hand, “Windows Store” applications are based on the new execution engine called the “Windows Runtime” or WinRT for short. The WinRT engine uses a new programming library and set of API’s which include a subset of the .Net Framework. Basically it is the .Net Framework for WinRT. This is not the full fledge .Net Framework 4.5 we developers have become accustom to. The K4W .Net Assembly has not been tested, nor has it been given the official “stamp” of “support” for this framework. There is no guarantee that this assembly will work with this new .Net Framework for WinRT. So, as a developer, if you want a K4W device to work correctly in the official supported way with a “Windows Store” application you have 2 choices.

1. Wait for Microsoft to release a new Microsoft.Kinect.Dll for WinRT…
2. Or, create a wrapper library yourself, following the new design and guidelines for the WinRT engine.

One would figure that Microsoft will eventually do this, however Windows 8 is out now, and it could be another 6 months to a year before we see or hear anything about choice #1. Thus, the answer to the above question is yes and no, it depends on the OS you are targeting as a developer.

Continuing on with the other flavors of Windows 8, we have Windows 8 RT. Windows 8 RT only supports “Windows Store” applications, and it must be compiled for the ARM processor chip. This effectively means that the only set of developer API’s available to you are the WinRT api’s and the .Net Framework for WinRT.

Windows 8 Embedded supports both Desktop and Windows Store applications, however it depends upon the configuration of which core Windows OS components are enabled. Windows 8 Embedded allows you to remove or disable core components, such as the Desktop application support, and Windows Store application support and a wide host of other things, like keyboard support, network, printing and etc. Windows 8 Embedded also allows you to control which processor chip you want to support: ARM, Intel, x86, x64, and others. Windows 8 Embedded is the only Windows 8 based OS flavor that allows you to create a “Kiosk” on a touch screen device and completely “harden” it or lock it down.

Question: So you mean to tell me you’re going to write a wrapper around the Microsoft.Kinect.Dll Assembly?

It appears so… I’ve already started down this path. I have the Kinect Sensor Count, and Kinect Audio capabilities working, the rest is to follow. In part 3 I will discuss what it takes to build a wrapper for the WinRT engine. Maybe I will get lucky and the library will compile for ARM as well, this way the wrapper can support all the flavors of Windows 8.

It has been a while since I wrote about the Kinect for Windows (K4W). The reason is because I’ve been busy learning all about the Kinect, and thanks to the Kinect for windows team at Microsoft, I’ve also been learning about the upcoming changes and additions to the K4W toolkit, SDK, and device.

Let me say this without saying much… It’s an exciting time to be a device developer, especially a Kinect Device developer.

With the recent rumor about the Kinect 2.0 device, and soon to be released Kinect Fusion
(1), developing applications for the Kinect Device and K4W couldn’t be more fun and entertaining!

While I can’t specifically speak about any other upcoming changes and additions let me just say again… It’s an exciting time to be a device developer, especially a Kinect for Windows device developer.

As some of you already know, Microsoft recently release it’s flagship OS and products line: Windows 8, Windows 8 Surface RT and Windows 8 Surface Pro devices. This OS and these devices once again are making a splash into IT industry. I won’t argue the goods and bads, being you all know I’m a Linux Gentoo Geek since 2009. I will say that both the OS and line of devices have changed the way developers develop and design applications. Simply put, it is another game changer, much like the Apple IPods, IPhones, and IPads, not as major, but still a changer none the less. It is not as much a game changer like the Xbox Kinect device but again a changer none the less.

Understanding the development changes, and design changes of Windows 8 and its devices will require a learning curve which I too am not excused of its effects. Thus the purpose of this posting is to talk about my experiences when developing an application for the Kinect 4 Windows device on a Windows 8 OS and its many devices.

First off right out of the box, there are many things to think about when designing and developing a K4W application for Windows 8. There are too many things to discuss in this one post alone, however I will briefly bring them up in a bullet points. As I develop applications, I will go back through these bullets and most likely talk about my experiences with them in many posts during the year.

Design Challenges:

• Windows 8 devices are Touch Devices
• K4W is a Touchless Device
• Windows 8 has areas outside of the pixel for user interactions
  • Such as Swiping from Top to bottom
  • Swiping from Left to right
  • Swiping from Right to left
• K4W has no pixel representations of “Area”, it’s area is the whole space surrounding the user (within the device limits of course)

Development Challenges:

• Windows 8 RT devices can only install applications from a Windows Store
  • Currently no driver exists for K4W device and its many sensors
  • There are currently no Windows Store applications which contain the K4W device and drivers
  • There are currently no Windows Store applications that are compatible with K4W
• Windows 8 Pro devices do support the K4W device and drivers however depending on the GPU of the Pro device, it may not be powerful enough to support Kinect Fusion
• Windows 8 requires the WinRT libraries for development
  • VS.Net 2012 is the only development tool for WinRT which by default doesn’t support K4W
• K4W can be developed with C#, VB.Net, C++, DirectX, and XNA
  • No library exists right now to work with WinRT

These challenges are just a few as more will probably be added to this post as I develop my Kinect for Windows 8 application.

In part 2 of this post, I will talk about the steps I have used thus far to get the bare minimal framework working for K4W based applications.

Introduction

TFS Server 2012, and Team Foundation Service (TFS in the Cloud) allow you to take advantage of the power to provide source control, collaboration, agile planning, test execution and continuous builds and integration.

Currently, I am working with Nimbo Technologies, and we recently had a project where we needed to implement continuous builds, and create a automatic packaging and deployment mechanism. We utilized Team Foundation Service (TFS in the cloud) as our source control, however the default feature set of TF Service did not allow us to customize our packaging and deployment process according to our requirements.

Enter the world of the Hybrid

We wanted to take advantage of Team Foundation’s Source Control in the cloud (TF S ervice), and still have the ability to customize packaging and deployment processes. At first we thought this was going to be difficult, however to our surprise, TFS makes hybrid designs a first class citizen with it’s core new functionality.

TF Service does provide a default build server. It’s limitations are that you can only build pure Core .Net Framework applications. This includes: ASP.Net, WF, WCF, WPF, and Windows Forms. Whenever you want to build Server specific product applications such as SharePoint Server, BizTalk Server, or anythng that requires server components or third party components to be installed and registered, you’ll have to ask Microsoft for the permission to install it upon the default TF Service Build server instance. This means it’s left up to Microsoft to allow you to do it or not. At the time of our project, SharePoint nor BizTalk was supported, and thus we needed to look elsewhere for a solution.

The Setup

We basically created two on premise TFS Build servers. Each server was configured to retrieve it’s source code from TF Service Source Control.

The following diagram represents its design:

The Steps to Build it Yourself

1. Install TFS 2012 Server
2. After installation, the wizard starts, select TFS Build Service
3. 
4. 
5. Click Next.
6. 
7. Click the Browse button:
9. 
10. Click Servers -> Add.
11. Type in the Name of the TFS Server:
12. 
13. You will be presented with a TFS Cloud service Login screen.
14. The Login you use here will be used by the TFS On Premise Build agent/service to access TFS Cloud.
15. 
16. Click Sign In to connect to TFS Cloud Service. Once logged in, you should have access to the TFS Cloud Service:
17. 
18. Select the Default Collection and click on Connect:
19. 
20. Next Click Next:
21. 
22. For Build Services use the default settings:
23. 
24. Click Next.
25. Use the default build service account settings:
26. 
27. Confirm the chosen options:
28. 
29. Click Next.
30. If you run into any issues, check that your Windows Live ID account used to sign into TFS Cloud is apart of the Project Administrators group.
31. You should see success screens such as this:
32. 
33. Click Close twice, open up or navigate to TFS Administration Console. You should see the Build Controller and Build Agents successfully configured and running within 5 minutes:
34. 

In Summary

TFS Server 2012, and Team Foundation Service, allow you to take advantage of having a major Cloud player like Microsoft, host your source code, and at the same time customize build processes, packaging and deployment.

The Kinect device has remarkable physical capabilities. It can scan the room using its infra-red sensors, listen for sounds with its array of multiple microphones, record video from its *web cam* like camera, and move/tilt its head up and down from -27 degrees to + 27 degrees.

However when ever I tell people that I am developing applications for the Kinect, there is a pause, blank stare and then a question that is usually uttered… “Oh so you’re developing games for the Xbox now…?”

The quick and short answer is this: “No, (however if anyone is interested, that is a possibility…) I am creating designs and enterprise/business applications that can take advantage of these physical capabilities.”

InfraRed Capabilities

Just stop and think for the moment. If you have an infra-red camera, and projector,which the Kinect has, you have the basis for a “See in the Dark” functionality. Which means you have a potential for guiding persons with an aid to see obstacles in the dark. From a security standpoint, you can be notified when an intruder has invaded your space in complete darkness similar to passive infrared motion detector (PIR). Now on top of this, envision the software that can be developed and run on the multitude of devices. Visualize a cloud based mobile device such as a cellphone, or tablet that can easily notify you of any impeding danger, where the Kinect is mounted. Danger such as that annoying toy car with its sticky points innocently resting on the floor which you’re about to step on in the middle of the night as you make your midnight refrigerator raid. Imagine steps before you reach this painful moment in time, a cellphone rings, or even better, a speaker embedded in the house walls yields a quick chime or sound to warn you to look down.

Microphone Capabilities

The Kinect also has the ability to be trained through software to listen for sounds, speech, and other audible things. Imagine the ability to listen for when your new born starts crying in the middle of the night, where you can immediately monitor the crib, for an empty bottle, and possibly have a lullaby play through speakers in your house. What about voice recognition? This ability is also possible with a Kinect. Envision a security device which needs three factory authentication, such as a password, a security key card, and some form of biometrics: (voice, fingerprint, retina or facial scanning) to prove identity. The Kinect device can can provide the voice, facial, and password requirements by listening through its microphone, recognizing a person’s face, and using speech recognition through software to decipher words.

Camera Capabilities

Lastly, the Kinect device can record video and take pictures with its VGA camera. Imagine software which can identify clothes, purses, shoes along with their styles, fabric and makers from a video or picture. With a Kinect device inside a clothing or department store, discounts, offers, advertisements can be given to a customer when the walk past the device or even queue up to purchase an item. Let’s say because I’m wearing the latest Nike Air Jordans, the Nike jumpsuit I’m about to purchase will automatically be discounted 10% due to my patronage; along with an offer to purchase a Nike Spandex body fit exercise shirt… nice!!!

So do you get it now???

In summary, Kinect development means working with the Kinect Device for its capabilities, and developing software which can propel humanity into the future, or as I like to say, bring us up to speed with the ancient cultures that existed before us

If you want to see some more ideas on the Kinect device visit Kinect Hacks.

If you’re curious about how to use Node.js with Windows Azure on Linux, there’s some good documentation on the WindowsAzure.com web site here:
http://www.windowsazure.com/en-us/develop/nodejs/

Clicking on the Linux Download, gives you with Azure Source code, where you can compile, and install the NPM module for Azure for Node.js

The steps for the most part are simple:
1. Install Node.js from http://www.nodejs.org
You can emerge it if you’re running Gentoo
or use any package manager for you flavor of Linux

2. Install Azure for Node.js
Compile, and then install from the source code downloaded earlier.

As I made my way through these steps, I was presented with a small issue. Azure for Node.js requires Node.js version greater than 0.6.15.

The error I kept getting was:
‘Node version needs to be higher than 0.6.15′

Originally I had an older version of Node.js installed, and uninstalling it, and re-installing the new version didn’t seem to fix the issue. I also refreshed my environment vars, with env-update command, and created a symbolic link to node version 0.8.3 from the downloaded source tarball. All to no avail. It still didn’t work for me.

So I put my developer cap on and decided to look at the “configure” script that you run when getting prepare to run “make” to compile the Azure source code tarball. Low and behold it’s a python script that’s checking for the version of Node.js, and it appears to be incorrect, or at least the values are not correct. Thus I just commented the check out with a ‘#’ mark, and re-ran the command:
sudo ./configure
Success!!!

Next I ran the command:
sudo make

Then:
sudo make install

And now I can run Azure commands to monitor, and provision all things with Azure from my Gentoo Linux Laptop…

I have embarked on a new opportunity dealing with Kinect, and RaspberryPi. I have also been doing some Secure Coding/Certified Ethical hacker/Web Pen testing type stuff as well recently. So stay tuned for some posts on all these things mentioned above.

Now to the announcements. Microsoft has made some splash in the news over the past few weeks dealing with Windows Azure, and Office 365. In my oppinion Microsoft is really taking the competition to the likes of Amazon Cloud offerings, and Google Docs cloud offerings. In the Office 365 space, Microsoft is making it extremely easy connect work on Office products in the cloud, build custom applications and reports, and deploy them with a click of a button. You can read more about the fancy tips and tricks here:

Office Next Blog
http://blogs.office.com/b/office-next/

TechNet Office Tech Center
http://technet.microsoft.com/en-US/office

Office 365 Preview Blog
http://community.office365.com/en-us/preview/blogs/office365preview/default.aspx

As far as Windows Azure, Microsoft is making it easy build, develop, architect, implement, and deploy all other types of Cloud solutions. So on one hand, you have Office Product Solutions with Office 365, then everything that is not covered under Office can be done through Windows Azure. Windows Azure offers the ability to create Web Sites with the click of a button, or type in a Node.js command from your Gentoo Laptop (As I’ve done), to monitor, stop, provision and manage web sites. Azure also supports BLOBS, Virtual Machines, Identity Management, Service Bus, and more. Just navigate to
http://www.windowsazure.com/en-us/

Or check out Scott Guthrie’s video showing a good 1+ hour demo of some if Azure’s capabilities:
http://www.meetwindowsazure.com/conversations#ScottGuthrieMeet

Stay tuned for more updates.

How to get LibFreeNect and OpenKinect working on Windows 8 RC

- first install Visual Studio .Net 2011 – Express C++ should be fine.
- Next install .Net framework 2.0 – 3.5 from here https://www.microsoft.com/en-us/download/details.aspx?displaylang=en&id=21

- Next install Git Extensions from here: http://gitextensions.googlecode.com/files/GitExtensions232SetupComplete.msi

- Once git extensions are installed, run the following commands inside a Git Bash Shell prompt:

$ git clone git://github.com/OpenKinect/libfreenect.git //downloads the project, creates local master
$ cd libfreenect //cd to source directory
$ git branch –track unstable origin/unstable //creates a local unstable branch for remote unstable
$ git checkout master //check out the master or unstable branch
$ or
$ git checkout unstable
$ git pull

- Next download libusb-win32 from here: http://sourceforge.net/projects/libusb-win32/files/libusb-win32-releases/1.2.6.0/
- Extract this file into a folder of your choice

- Next download pthreads from here: ftp://sourceware.org/pub/pthreads-win32/pthreads-w32-2-9-1-release.zip
- Extract this file into a folder of your choice, make sure its different from the libusb files previously

- Next download OpenGL Utility Toolkit (GLUT) from here: http://user.xmission.com/~nate/glut/glut-3.7.6-bin.zip
- Extract the file into another folder of your choice

- Follow the Driver Installation instructions from the OpenKinect.org web site on Windows 7: step-by-step walkthrough found here: http://openkinect.org/wiki/Getting_Started#Windows

- Download CMake for windows from here: http://www.cmake.org/files/v2.8/cmake-2.8.8-win32-x86.zip
- Extract and files and then run: cmake-gui.exe

- Set the Source Code path to the location of where you git cloned libfreenect
- Create a directory for building the VS.Net solution project for libfreenect. The CMake tool creates a VS.NET solution with all the projects for libfreenect, glut, libusb and pthreads.
- Select the Grouped and Advanced checkmarks and then click on the configure button.
- Select the Visual Studio 11 generator
- Expand the BUILD Entry and only check mark: BUILD_C_SYNC and BUILD_EXAMPLES

- Add one new path entry: GLUT_INCLUDE_DIR set the include directory to the Visual Studio VC include directory, by default it’s c:\program files (x86)/Microsoft Visual Studio 11.0/VC/include
- Add one new filepath entry: GLUT_glut_LIBRARY set its value to the glut32.lib from the GLUT Extracted files

- Set the LIBUSB values: LIBUSB_1_INCLUDE_DIR to the include directory of the libusb extracted files
- Set the LIBUSB_1_LIBRARY to the libusb.lib path from the libusb extracted files

- Set the threads values: THREADS_PTHREADS_INCLUDE_DIR to the include directory of the extracted pthread include files
- set the THREADS_PTHREADS_WIN32_LIBRARY to the pthreadVC2.lib file from the extracted pthread files

- Click the Generate button- VS.Net solution files will be created.
- Open the VS Solution from the “Where to build Binaries” folder you selected earlier.

- Inside the GLUT extracted files, create a directory named GL
- Copy the glut.h header file into the GL directory

- Inside VS.NET, for the projects: Regview, glpclview glview, hiview, inside the project properties, add an include directory path to the GLUT header file.

- Rebuild the solution.

- Copy the glut32.dll and pthreadVC2.dll into the Build output folder.

- Run glview.exe and if everything works… SUCCESS!!!

… NEXT UP, how to get the Kinect 4 Windows working… now this is a challenge!!!