Chris L. Bennett

Over the past couple of months, I've made several different packages with Open Web Studio. And after creating a simple content version package, article manager and a link exchange system I think it is a great tool for creating simple packages.

Unfortunately, the script language it inherited from listX is not exactly easy to master. Not only is the script language a bit difficult, but there is very little documentation available. I was able to learn it by starting with the listX documentation and downloading as many examples as I could find. Nonetheless this is time consuming.

I do believe they have started to address the short comings. Lately, the OpenWebStudio site has drastically improved the content available on the wiki. The lack of information is mainly why I published the packages I've created for this site. There is also a page on the wiki describing changes to the script language to make it more usable.

Even though OpenWebStudio is still fairly new, I would have to conclude that is a very good tool for creating fairly simple packages to augment DotNetNuke. However, if you intend to create complex modules, I would recommend creating native .Net code.

Extrenal drives are so cheap these days that anymore you can pick up a 500GB at Frys for <$70. Like many out there, I have several of them. Rather than have all of these drives connected to one PC. I would rather have a small home server. However, I'm not about to pay what HP wants for a Windows Home Server. Essentially the Home Server is a light weight version of Windows Server. So, I've looked for a long time for a small and inexpensive server that provides more functionality than just your basic NAS (Network Attached Storage).
Last week I came across an article on Slashdot regarding a new $100 embedded device designed to run Linux. The device is called a SheevaPlug. The specs for the device are quite impressive as listed below:

• 1.2Ghz ARM5 processor
• 512Mb Ram
• 512Mb Flash
• 10/100/1000 Ethernet
• USB 2.0
• SDIO Slot
• USB Development Port

Amazingly all this fits into the size of a "wall-wart" as shown in the images below:

 

Here is a block diagram of the internals

 

Out of the box you will need to go through a fairly technical process to install the boot loader and the file system image over the development board.  Once Linux is installed and running you can run the normal software you would expect in a small server Samba, Apache, PHP, MySQL, NFS, etc. Quite a few Linux distributions support ARM processors including Ubuntu and thus can be used on the device. The development kit is available in small quantities through the Marvell site.  I'm pretty sure I'm going to purchase one for myself to use has a light home server and as a remote backup for my virtual server.

Over the last couple of days I started using the iGoogle page. Well, it didn't take me too long to create my first Google Gadget. You can add my FightClub Quotes gadget to your iGoogle page below.

Creating the gadget was much easier than expected. Google has a very nice gadget editor that plugs into the iGoogle page. Basically a gadget is an XML doc with a configuration section and an HTML content section. The gadget also has a fairly simple set of javascript commands that provides access to remote content. One confusing thing I found when I started building the gadget is that there are 2 different APIs. The new and the legacy. My first reaction was to use the new format and go from there; however, currently the new format is not supported. For some reason this is not explained very well. I have a couple ideas for gadgets that I'm going to build. I'll publish them here as I complete them. To start creating your own gadgets click here.

Recently, my job has had me traveling quite a bit. So I haven't had much time to actually sit down and write down my thoughts in blog entries. Regardless, for some time I've been searching for good DotNetNuke data modules. Specifically I generally focus on modules which provide the ability to work with data. A couple of weeks ago I stumbled upon the open source module Open Web Studio. Open Web Studio (OWS) is an upgraded version of the ListX module from R2Integrated.

While OWS is open source they do have several different subscription levels. The subscriptions are a bit pricey from my persective; however, I'll let you choose. Mainly the purpose for the subscriptions levels is to provide commercial support. I've been using OWS without a subscription and so far, the user interface is quite good; however, it is quite a powerful and unfortunately has a steep learning curve. Recently a couple of basic videos have been posted to a Wiki on the OWS site. The Wiki so far is an ok reference for the package but not great. I also did some googling which gave me a couple of samples to start looking at. Another good source of information I found was to download a copy of the latest documentation for the ListX package off the dnn.bi4ce.com site.

Unlike other data module for DNN, OWS is extremely flexible in how the data is stored and how the user is able to interact with it. Essentially, OWS is a basic scripting engine. It works off of basic actions such as Query, Template, Email, If, Else, If-Else, Redirect, Variable and Assign, etc. By string together these actions is is possible to make choice based upon query parameters to allow adding/editing/deleting of records as well as data tables. OWS itself does not have a native format for storing data in the DB, instead it uses tables that you create. This provides for much better control and performance over tools such as XMOD.

In playing with OWS, I've create a couple different modules. One of these modules is currently active on this site on the About page labeled articles. While most of the articles module is completely hidden it has a complete admin interface which allows for adding/editing and deleting of articles. I have a couple of other modules that I've started, but are yet to be completed. I almost forgot ... in case you plan to distribute modules you create, there are instructions availabile on the OWS Wiki to create a Private Assembly (PA) for distribution.

In the near future I plan to post these modules to a software page and make them freely available. I have a couple of other things I'll release on the software page including a TCP client. 

Last weekend at the Indy Tech Fest I was talking with Bill Steele regarding the Microsoft Robotics Studio. We started talking a product he came up with called a VirtualHUD he came up with. Basically the VirtualHUD is a handheld heads up display unit for single engine aircraft. Why single engine? This is the part that I didn't understand at first. The reason for the single engine is that the VirtualHUD projects a heads up display on the back of the propeller. Without a doubt this has to be one of the more interesting ideas I've seen in a while. Bill had a video handy from test run on his plane at night. Quite Impressive, it very clearly projected a horizon line and the other vitals. The item in the video I did not notice at first was the video projected behind the horizon. Yes behind. He has an infrared camera on the plane and it was projecting the live video with the horizon and other information overlay. I'm sure once he is in full production this one will be a big hit in the general aviation industry. Be sure to check out Bill's website to find out more details. To be able to perform all of the processing necessary Bill's product certainly needs a hefty processor or quite a few small ones. The solution he employed was to use a new processor ironically called The Propeller. The Propeller processor is made by Parallax. Parallax for years has been known for their role in education through their Basic Stamp processor and kits. The Basic Stamp is really an off the shelf processor from another manufacturer that is loaded with a Basic language interpreter. This processor is mounted on a carrier board with the support components (crystal, memory, voltage regulator, etc). The entire package is sold as the Basic Stamp.



The Basic Stamp is a bit expensive for the processing power it offers and its lack of C compiler does not make it an attractive for use in commercial applications. The Propeller processor is a complete new design as it represents Parallax's first custom silicon solution. The Propeller still has some of the hobbyist qualities in that it does have a built in interpreter. This time for a different language Spin. But it also has the ability to execute assembler code something the Basic Stamp lacks. This then opens the door for a 3rd party to create a C compiler. ImageCraft has created a C compiler and it is currently available for roughly $250. The Propeller's architecture is quite unique. Number one it is fast at 80Mhz (20Mips). It is a full 32-bit processor. Oh yeah, and I almost forgot it has 8 cores. Yes, the ability to simultaneously execute 8 threads in a part that is in the $11 range. Below is an image of the architecture of the part.

 

Each of the cores in the processor act independently and each contains 2k of RAM, timers and a Video Generator. Yet another unique feature a Video Generator. Many of the demo boards include connectors for VGA and for an NTSC signal. The cores also share resources including 32k of RAM and IO. The 32k of RAM is used for both data storage and for code space. A central component called the Hub takes care of sharing the resources. Since the RAM is shared by both data and code it must first be loaded with the application. This is done by pulling the code from an external 32k SPI EEPROM. This opens the door to pull code in from other sources. And I've been able to locate several drivers for accessing SD cards formatted in FAT 16. On that not, there is a significant number of objects available for use in SPIN to communicate with all kinds of components. Just a highlight of a few I've found are: TCP stack with SPI to Ethernet Driver, 4 Serial Ports with hardware...

I never really thought about what was the fastest ride on earth until I saw a recent Modern Marvels episode on the History Channel. I knew the Mclaren F1 would do 240mph and vague knew of another that could come come close to 250mph. Then the land speed record which is currently held by the Thrust SSC Team who broke the sound barrier in 1997.

I never really gave much thought to the military .... But as I was watching the show I expected to hear a speed of 3, 4 or maybe Mach 5. In fact they have a sled at the Holloman Air Force base capable of Mach 8.5 or just over 6400mph. It was an unmanned vehicle and it ran on rails. To achieve this speed the team used 4 stage rocket and carried a payload of 25-pound. Not only did it achieve Mach 8.5 it only had 3 miles of track to do it in. Your average car cannot achieve 60mph in 6 seconds. This thing went from 0 to 6400mph in 6.04 seconds using up 3 miles of track. When I saw the brief video of the test I instantly had several questions like how did they keep it on the track. Unlike other railed vehicles the rocked sled does not use wheels it uses a shoe type device that wraps around the rail to hold the sled down. And how do they keep those from melting at 6400mph? Unfortunately, they did not talk about stopping the sled and how could you. If it took 3 miles to reach Mach 8.5 how much room would you need. At the end of the the track there must have been a huge creator.

Given the way our military works, the Mach 8.5 record has probably been superseded by another speed record. Have they reached Mach 10 yet? I guess we will find out in a few years after it becomes declassified.

Although, I'm not an expert at all in particle physics, its good to see that after years of work, the Large Hadron Collider (LHC) is now online. Well, not exactly ... The LHC is functional from the stance that it is able to send particles in both directions around the ring; however, no collisions have taken place. It is amazing to think that the 17 mile ring will collide particles at 99.999999% percent the speed of light. Yep, you read that right. Granted we are talking about micrograms of matter traveling at that speed. From what's been in the press the first collisions are to start taking place sometime in late October. The high energy collisions are scheduled to take place in about a year.

So what are we actually going to learn from this? I think there is a lot of speculation around what we will learn, but no one actually knows. Many think they will locate the Higgs boson particle which is believed to give matter its weight. I'm intrigued by the fact that Steven Hawking has bet $100 that we will not find the particle. His view is one where we will probaly find something even more interesting.

No matter what we find it should be interesting. Something I didn't know until all of the press regarding the LHC came about is that the USA actually start building a similar test facility in the early 90's called the Superconducting Super Collider (SSC). Unfortunately the project was cut in '93. The US version of the LHC was actually planned to be much larger and far more powerful (54 miles instead of 17 and nearly three times as powerful).

Last weekend I was listening to a live recording of a podcast called Leo Laporte "The Tech Guy". During the show he did an interview with a physicist friend involved with the show. Some of the information she (I'm sorry I do not know her name) discussed related to the amount of data that will from the detectors when the LHC is running. I was completely shocked to find out that the LHC detectors produce nearly a petabyte of data per second. Today it is not common place yet to refer to data in petabytes as is gigabytes or terabytes. To deal with such a volume of data they LHC employs several hardware solutions to perform a great deal of filtering to skip any "uninteresting" data. Still the amount of data that must be stored must be huge. I would guess they are planning to use a grid of some time maybe event use the SETI@home concept.

In the end, I'm sure we will learn quite a bit from this big science project.