So now NASA is considering using robots to repair Hubble instead of a manned mission.  The primary reasoning being that it is safer and more cost effective.  While laudible reasons, I think that using robots should be the preferred method for a completely different reason.

Let's face it, putting a man up there and doing repairs, maintenance and upgrades isn't technically challenging (though it is rocket science).  We've done it all before, and doing it again really doesn't teach us anything or push the boundaries of space exploration for us.  Using a robot, on the other hand, is a new advancement.  We've not done it before and by pursuing it we'll be improving our ability to understand the complexities of the job.  THis will further our ability to make robots that repair other craft and machines even further from Earth.

So by all means, scrap an astronaut visit. It's costly, risky, and just plain boring.  If we want to push further in our abilities to explore space, we need to push our abilities to do work locally, and in this seemingly rare case what is good for science may also be good politically.

This article predates my days as a developer (I was a seismic geologist in those days) but it's unbelievably valuable.  Sure, it talks about Windows 3.1 and DOS and 95 was the latest, but if you develop on any windows platform you should read an understand it.

Well I think I may actually join the digital age and get a digital music player - I'd love to get a PMC when they come out, probably the one from Creative (the demo I saw in Redmond was phenomenal).  In preparation I've started ripping my large CD collection to MP3 format using CDex.  A nice app, but just becasue of the number of CDs I have to rip, this could take months.

First up:

Mad Season - Above

Coming soon....

OpenNETCF Development and Consulting Services.

Need help architecting or implementing a .NET-enabled Mobile solution? Need advice on design or just help writing code?  We are looking for possible contract work for our upcoming DCS branch.  If you would like more information or you have a specific project in mind, feel free to contact us.

How's that for a fatalistic title?

I've just read Randy's Blog by way of Mark's Blog and it got me thinking about the work I've done in the past and work I'm in progress on.

The gist of it (as I see it anyway) is this:  with Microsoft's internal members becoming more publicly available, and the push for Microsoft to let the public in on products and technologies ever and ever earlier, what's the point of printed material and training courses, especially when delivered by people with a name that isn't instantly recognized as a technology leader?  Let's face it, the development cycles are getting ever tighter, but capabilities more robust.  It's taking longer to explain complex technologies, but they come out ever faster.

Paul Yao, a guy I respect as a leader in CE programming and training just released a fantastically informative book on the .NET Compact Framework, yet I can't help but think that it's nearly obsolete already.  In the meantime, anyone with a question about the CF has been able to go right to Microsoft's newsgroups and ask questions and get answers right from either MVP or people on the CF team itself.  And the public already has had a couple drops of the next version of the CF.

So are tech books, articles and training becoming obsolete?  For that matter, how about web sites?  Let's face it, many sites hold static inforrmation about a system that is becoming more and more fluid.  I find that information even in the popular Forum paradigm, which is far mor dynamic than a book, is beginning to become stale - old information clutters up the knowledgebase and makes searching more difficult.

I'm running into this problem with the knowledgebase at Applied Data and at OpenNETCF and I can't even remeber the last time that I updated the stuff on the IDSS site.  It would take a small army of full time experts to just constantly sift through all the data, throwing out the chaff and recognizing the wheat, plus keeping everything new coming in.

So what are we to do?  One format that I'm toying with is the Wiki.  I like it because it's a very fluid medium.  It can be updated instantly and by anyone who's reading it.  Sure, oon the surface that's pretty scary, and I've hit some resistance and skepticism about it, but I think that's simply fear of the unknown.  With a Wiki you essentially have the potential that every reader is a web master.  If a reader find stale or bad content, they're free to change it.  If something's missing, they can add it.  No notifying the web authority, no submitting change requests, just change it and be done.

Sure the potential for misconduct is there.  Someone could easily come in and deliberately delete information, or worse add incorrect content.  The point is that you expect 99% of people to behave, and those 99% will continually fix what the 1% fuck up.  For the compulsive admin, there are changelogs and IP logging so filtering is quite possible.

So let's look at an example.  The "official" CF FAQ is maintained on MSDN and it started out as a great idea.  It provided quick answers to common questions.  Then Microsoft decide they would update it frequently.  Sounds like a positive thing, right?  Well it didn't take long for it to become the monolithic beast you see now.  It's become too damned hard to find things in it, and I've stopped using it or referring people to it altogether becasue it's simpler to actually answer a question than try to find the answer in the FAQ to point to.  Add to that the fact that the FAQ has some errors in it (mostly formatting), some info is stale, and some of the questions could be answered with more information since better online stuff has appeared since the item was posted in the FAQ.

Since I'm a fan of openness, I've decided to try to both "fix" the FAQ and start a fluid running knowledgebase about the Compact Framework in Wiki form.  The FAQ isn't even fully populated, and what is there is simply a copy and paste from the MSDN stuff - not updated as I'd like, but I've got a start on it and put it at wiki.opennetcf.org.  Once I have the FAQ punched out, then I figure we'll make a "public announcement" or some such.

The hope is that the CF community will maintain it and it will organically keep current.  It's a topic-specific encyclopedia that's never finished, yet hopefully always current.  It's not intended as a replac ement for a newsgroup or a Forum.  It's not a place to ask questions.  It's simply a place that hopefully can contain more answers and knowledge than any single person or group could easily provide.  The hope is that the whole (the wiki) will be greater than the sum of the parts (each individual's knowledge) and with it we can all have an easier time doing what we like to do.

One of my co-workers has a set of "rules" posted on his wall.  Basically quips that he's made up, and he uses them in conversation regularly.  Things like "Don't confuse a problem with a desirable situation."

i hadn't thought much about it, but I have my own list, it's just not on paper.  It came up when I used one of my "rules" and he asked if he could use it.  Not like I'm trying to patent them or something like "You're fired" (what a jackass).  At any rate, here are two of them:

It's better to have something and not need it than to need it and not have it.

This one started when someone asked me ages ago why I was buying my Glock, but I now like to use it for power tools, hardware accessories, car parts, and just about anything I want but don't need.  Of course my wife still doesn't buy it.

"Idiot proof" is unattainable.  They'll always invent a better idiot.

This comes up a lot when trying to decide just how far to take a sample or how much to document something.  Trying to satisfy the dumbest possible user just can't be done, so now I simply try to cover the upper 80-90% (depending on my mood).