The new memory arrived in the mail right on time, and I could hardly wait to get one of the two new memory chips out and into the ELF2K to see if it would work with a new chip.

Of course, I had to wait, there are always things around the house that I need to do. But eventually I got a chance to head into the den and put the new RAM chip into the socket. Put it in, powered the ELF up, flipped the RUN switch up, and son of a gun the little computer worked on the first try.

Only I didn’t have a terminal hooked up yet. So I needed to download a terminal program (TerraTerm) and set it up (baud rate, port), hook up the serial cable (take the shield off the DB9 socket on the ELF, my serial-to-USB converter sucks), power everything up again – and it still worked.

I then spent a happy half an hour toggling programs into the ELF – starting with the one from page 66 of the March 1977 issue of Popular Electronics. That’s the one where when you flip the input switch once it turns the Q LED on, and when you flip the input switch again it turns the Q LED off. That’s the first program that I ran on my original ELF, and the one that I showed my Dad. I showed it to Katie – the nostalgia was thick in the air.

I’ll be playing with the ELF now from time to time, it’s a bit of nostalgia after all. I’m probably going to have to get the I/O expansion board so that I can hook it up to my Picaxe network, but that’s a project for next winter I think.

For my birthday this year my wife got me a Spare Time Gizmos ELF 2000 kit, a re-interpretation of the original 1802-based computer that I (and many others) built from plans in Popular Electronics.

Just like the first one that I built, this one had a few problems when I first turned it on. Finding and fixing a bad connection in an IDC socket was fairly easy, and I found the cold solder joint on the switch panel after just a little more work. What was harder was the fact that some memory locations just didn't seem to be changing, and I couldn't figure out why.

So I built a test rig for the 32K static RAM on the ELF2K. I used a Picaxe 40X1 as the brains, two 74HC595 chips to latch the 16-bit address required, and a handful of LEDs to show me what's going on.

I used PortC to read and write data to the RAM chip. I had some difficulty getting the bi-directional data bus to work until I put dirs= statements in the code to explicitly change from output to input when writing then reading the test data.

Since PortC was in use, I couldn't use the hardware SPI port, so I used the simple bit-banged serial protocol from the manual.

Once I had the test rig up and running it started indicating memory errors throughout the RAM chip, but especially on the last page of memory, the page that the ELF2K uses for it's system data page, and where the OS on the ELF2K was indicating there was a problem.

I've ordered a couple more RAM chips that should be here tomorrow. Once I've got a known good RAM chip to install in the ELF2K I'll be one step closer to getting it to work.

If you visit our site often (and how many of you do, really?) you may notice it's got a new look. I've switched to using BlogEngine.NET to manage the content. Hopefully using BlogEngine will make it easier for me to keep the site up-to-date since I'll be able to edit from any Internet connected PC. For example, I've already fixed a couple of typos on the front page that were just too much trouble to clean up with the old way of working with the site.

Over the next few days (weeks) the site will keep changing as I work on getting BlogEngine to display it the way I want it to look.

Here are the things I'm thinking about tackling next:

• Adding an editing form for custom attributes. I have this done in my local copy of the application, I'm just waiting for Peter to release the 0.3 version of the app before I check it in.
• Adding a context menu to the tree view so that it's easier to add, move and edit nodes.
• Updating the editing form so that it follows the UI guidelines.
• Adding pages that enable color changes to item backgrounds and text.

That will keep me busy for a while, I think.

The first set of changes that I submitted to the Pocket Freemind source were some clean up work and adding an editing form for notes and HTML content for the node.

I re-factored the source code for the base class (MindMapItemBase) and moved some redundant code from the other node types into the base class. Now nodes that only have attributes, like >link< and >attribute< don't have to have any changes made to the ToXml and FromXml methods, the base class versions do all the work.

My other big change was to add an editing form so that you can edit >richcontent< nodes. These nodes contain free-from text or HTML that makes up the nodes content. I've got some ideas in my head about how to create Web pages from the contents of a mind map's content nodes, and I want to be able to edit the mind maps on my PDA while I'm riding back and forth to work.

One of the things that I do often as a writer is brainstorm a set of content, whether it's a documentation set at work or a set of pages that I'd like to upload to this Web site. One of the tools that I use when I'm brainstorming is a program called FreeMind, an open source mind mapping tool that you can find on SourceForge.

One of the things that I'm trying to do with FreeMind is use the XML that I create on the mind map to drive other projects. For example, at work I use a FreeMind mind map to structure the table of contents of my documentation set. I have a XSL transform that I run against the mind map XML to create a sign-off worksheet for each release of the docs. I want to try to use the mind map to actually construct the TOC and run a tool against the XML to create a set of DocBook files that 'XInclude' the appropriate documentation files.

Anyway, I use FreeMind on my desktop PCs, but I wanted a tool that I could use on my Pocket PC as well. There are a couple of tools out there that people have created, but the one that uses the .NET Framework is called "Pocket FreeMind." It's a project started by Peter Carroll, with some help from a few other developers.

And now I'm one too.

Peter accepted some patches that I created to edit notes on the mind map nodes and added them to the project, and asked me if I wanted to be a part of the development team.