At MEDC I had a conversation with Seth Demsey about making a guitar tuner with a Pocket PC.  The challenge is getting analog data into the device and I proposed using an external ADC with an I2C interface using the serial port control lines for the I2C bus.  I'd spent a considerable amount of time over the last couple months working with the SPOT processor  and writing synchronous serial code and our conversation got me to thinking that the work was actually salvageable. 

I refactored a lot of the code so that now it can be used for the Compact Framework or even the Full Framework.  Couple this with the SDF's serial port classes and you could have an I2C bus running in under 5 minutes on a device or the desktop.  What fun!

Before you say “well the docs are nice, but where do I get the binaries?” let me preemptively answer.  I'll post binary downloads in the near future as soon as we get the licensing model ironed out.  These classes are not going to use the OpenNETCF Shared Source License.  Instead they will be commercial, for pay items.

Because I still think community access and learning is important for the community, I can tell you that the current plan is to offer the binaries as free for personal, non-commercial use.  Sorry, but only paying customers will get the source.

Well I just read a simple question in the newsgroups that had to do with an InitializeComponents method exceeding the 64k JITted code maximum so I figured I'd post a link to the FAQ on the OpenNETCF Wiki.  I went to get the link and lo and behold once again the wiki was just full of spam.  Rather than go through it page by page an find the changes I pulled down the DB locally and eradicated the spam.  So there's 45 minutes that I intended to spend working on a white paper that instead went down the toilet. 

I'm at the point that I've got no idea how to prevent the spam, and I just don't have the time to routinely go in and clean the damned thing.  Don't be surprised if in the near future the Wiki becomes unwikified and we just disable all posting or redirect it back to the forums. It seems that what had the potential for being a useful tool for agregating and disseminating information has been ruined by a few complete jackass spammers.

I was just saying that the SPOT stamp price would have to come down to be competitive.  As if to underscore that idea, look at what you can get for $100.  And that's a PXA255 with flash!  Heck for $120 you get bluetooth.  WEDIG is starting a project called GumSTIX to run CE on it.  This is way cool.

I wrote this a couple months ago, but never got around to publishing.  Here it is though:

Profiling the National 2F43ES (SPOT) Microprocessor

Are you working on a Compact Framework application and need some assistance?  Looking to start a new project and need an experienced resource?  Do you want the bragging rights of having industry leading experts to work on your application?  Well you've come to the right place!  The one and only Neil Cowburn is looking for contract work.  If you're in the UK great, so is he.  If not you're still in luck, we've got experience in remote development as well.

Click here for his resume and contact info.

So now that I've gotten my article on the TinyCLR published in .NET Developer Magazine, I'm starting to get email with comments and questions.  I'll provide what little info I know about here to prevent me from having to repeat it.

First, I want to thank everyone that emailed me for the positive feedback on the article.  It's good to know that people actually do read what I write and it's encouraging to see public interest in the technology.  I think it's that interest that's going to be critical to the success of the platform.

When I wrote the article I was using a development system that I had purchased from dotnetcpu for $500.  Of course I figured that the price was a bit outlandish considering that I think it's designed for the microcontroller space and you can get PIC, Zilog, Rabbit, Atmel and a host of other micro dev kits for much less.  The stamp price of $200 was also a bit steep.  You can get a cadillac PIC for under $12 - that addition of flash and some other stuff does make $190 to me, especially for a simple ARM7 processor.

At any rate, I realize that early adopters pay the cost of startup and that the price would likely come down over time.  We (my partners and I in Onward Technology) were hoping to see a sub-$40 price in a year or so.

Well after the article, I spent time writing another more detailed white paper on the characteristics of the processor (interrupt latencies, jitter, etc) as well as writing a substantial set of classes for several peripherals. 

I also noticed activity on the dotnetcpu site dried up rather quickly, and about a week before I went to MEDC I noticed the site just gave an error (which it still does today, hence I'm not giving any link to it).  Their apparent parent company, A DOT Corporation, is still there and lists the product, but doesn't give much info.

At any rate, I attended MEDC and tried to use my networking and beer-drinking skills to get any useful bits of info from Microsoft insiders on where the SPOT/Ollie (codename for the CPU itself) project was.  Unfortunately I didn't come away with much.  Those that I talked with couldn't give me much because of NDAs, though I was promised that I'd be informed of any status changes.

My takeaway for looking at developing with the processor was that the future of the processor as the stamp project is right now in some sort of limbo status.  There are advocates inside Microsoft pushing to get it to commercial viability, and currently it looks like things will move forward in a time frame of months.  So if you're working on SPOT today, you may want to hold off, but don't throw anything away.  Of course as with anything of this nature things inside Microsoft may change and it may go into the trash bin, or it may turn and come out publicly in a month.

I am still trying to make contacts inside Microsoft to try to get any scrap of info I can, and I'll try to keep the community informed of what I find.  If you're interested in the technology please post a comment here.  Maybe a little external support and interest can help drive this technology out of the log jam it appears to be in.

If you have any other questions, feel free to ask as well.  I hope to publish my detailed white paper in the near future (this week), even if the processor remains unavailable for some time.

Sorry, couldn't resist the pun.  Here's an article I did for .NET Developer's Journal this month.  It's a peek at what I've been working on the last couple months.