We have been having “slow” SharePoint issues at work (hosted on the over side of NZ-US WAN), and the IS guy gave me a trail URL to test with. Much better performance, but how do I show this. Camtasia to the rescue. Currently rendering the 1280x1024 video to flash, to send to the IS so he (and the Management) can see the difference.

It’s just finished, and it’s built all the .html .js and .swf files needed to be Internet-Ready™

In following my referrer log for my Curse of the Azure Bond project, I’ve found some quite interesting discussions about games re-makes. The scummVM forum had a link to the REWiki. This is a very cool idea to document file formats and the likes for old games.

This would seem like a good place to put my CotAB file format information.

Oh my god, Steve is back, and on fire. That Old Marshmallow Maze Spell is fantastic, I think I found two or three sub stories that fit me. Not sure what the whole story was now. Just that it flowed, changed and fit the whole time.

I’m still feeling shaky, story in a story, with some meta story in the story story. Not sure how, but I just feel profoundly changed.

The US military has an project called JRAC (Joint Rapid Airfield Construction) that has some of our gear as part of the solution. Recently a US colleague emailed a URL to read the blog posts of the current trial in Australia.

But the page will not load from New Zealand, so off to Google to view them in the cache, just make sure you click the cached links,

then click the cached text link, otherwise you have to wait for the picture requests to timeout..

So the bigger question is, why can’t I view the blog, and if it’s because the military want to “keep it secret” isn’t Google side stepping that via the cache? Or have the military not heard of robots.txt?

For my Queens Birthday weekend, I spent the public holiday building a wooden bookshelf for Jacob. Software developers like to compare building software to different things, like building bridges or houses, but building a bookshelf really felt like a small software project to me.

I had a customer (wife) who had a vision, “build a bookshelf” and I had a time frame “now”. I spent a few days getting the needs the bookshelf was solving, the maximum space constraints and the general aesthetics of the desired result.

Michaela went past a hardware shop to spec wood, and I designed the result to get the lengths of wood required. Which were then purchased. I then wrote a wood cutting plan, so I knew how to cut the 2.4 meter lengths to get the correct parts.

So then came build day Monday, 3 hours of mitre box cutting and jig sawing, the build could begin.

All along the way small refinements to the design were made. With the customer representative present quick decision were made.

Just in time for the children’s bedtime, the project was complete.

Today was the third CHCH DNUG Architecture Chat’s. Once again at the Bohemian Cafe.

We had seven attendees today, with a couple new faces.

Topics discussed:

• Google Analytics, and general site performance feedback systems. Bryn discussed the power of monetizing the cost of down time on a production server, and using this to improve development processes.
• Windows Workflow Foundation, I inquired weather anybody had any experience with it yet, Gary said he was starting to look at it, but had some useful articles from the web about others findings.
• Office Ribbon and task oriented customization, we talked about the office ribbon, and about other methods of solving the task oriented activities. I was firmly in the ribbon makes sense, “let-go of your fear luke” camp.
• SOA, followed on from the ribbon.
• A little gadget wars.
• And generally Bryn telling good tales from the dark places.

There was a standing in the door discussions also, where I partook in about TDD, and more WWF/WCF/DCOM architecture stuff.

Woot! I just did my personal IR3 tax return online, all very painless. Quite satisfying really.
Until I remembered I’d not mention donations.. luckily donations rebates are submitted separately on a IR526

Thumbs up for the IRD and making it painless.

Eric Sink has just posted that SourceGear’s DiffMerge is free. Woot a new tool!

As a subversion user, I thought I’d check it out, compared to the default TortoiseMerge. Well it beats the pant off that, but still left a few things lacking..

The first part getting TortoiseSVN to use the external tool.

Using the command line parameters %mine %merge %theirs

Hay presto it works. Well you get the Subversion failed merged result in the middle. As DiffMerge only supports three files so you can’t see the base and the result as the same time. But as the Subversion merge failures are messy, DiffMerge finds the merge failure points really easily with the Next Merge button (need to learn the keyboard short-cut).

A subversion merge failure look like this

The nice thing is you get highlighting between the mine/theirs so here you can see I inserted a HWND cast, and the replace with this block works real nice.

So first before I note my wish list, I will be using DiffMerge over TortoiseMerge. This will help as a stop measure until I get to trial the new Beyond Compare Merge tool (hope it lives up to my expectations)

Simeon’s Free DiffMerge Tool improvement list:

• Do four way merge: Mine, Base, Theirs, Result. This way I can see what each person changed and have a more informed edit.
• Allow me to manually match the 2/3 documents. Like I can do in BeyondCompare 2
• Improve the auto matching code, if in the above code picture, the left (and thus top of the merged “working” code was indented differently then the for loops would not align. This could be solved by fixing the indentation in the result file, or by deleting the working line.
• Give me a bigger screen at home, wow, you really can’t see much code on a 17” monitor, running 1152x864. Well this is my own problem to solve.
• Support the subversion method of marking the result set…

That’s it for now. Thanks Eric and SourceGear, this really is better than what I had before.

And all I have done so far is spill blood, trying to add Windsor to a working DCOM service.

So I’m going to try go back to square one, get the basic examples working, and progress from there…

Update: So I was trying

Container = new WindsorContainer(new XmlInterpreter());

or

Container = new WindsorContainer(new XmlInterpreter(new ConfigResource("castle")));

but always got:

System.Configuration.ConfigurationException was unhandled
Message=”Could not find section ‘castle’ in the configuration file associated with this domain.”
Source=”Castle.Core”
BareMessage=”Could not find section ‘castle’ in the configuration file associated with this domain.”
Line=0
StackTrace:
at Castle.Core.Resource.ConfigResource..ctor(String sectionName)
at METScomms.MainProgram.Main() in C:\dev\METSsharpe\METScomms\MainProgram.cs:line 23
at System.AppDomain.nExecuteAssembly(Assembly assembly, String[] args)
at System.AppDomain.ExecuteAssembly(String assemblyFile, Evidence assemblySecurity, String[] args)
at Microsoft.VisualStudio.HostingProcess.HostProc.RunUsersAssembly()
at System.Threading.ThreadHelper.ThreadStart_Context(Object state)
at System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state)
at System.Threading.ThreadHelper.ThreadStart()

So after discovering that there is a third page to the basic intro (after page two) I now have the programmatically setup container working… At least it’s a start…

Following the FeedDemon instructions I now have a public link blog feed new google shared feeds, so I’ll be posting link’s to other people’s blogs that I found funny (Scott Adams) or Tech Geek or just plan interesting.

Now I wont have to do link posts, so it’s will be just easier to share. Enjoy!