Read Kevin Sites' open letter to the marine unit involved in the mosque shooting.  It's one of the best blog post I have ever read although I don't envy the responsibilities he was burdened with.  My best wishes to Kevin and the Devil Dogs.

As I pulled into the driveway after an emergency visit to the dentist this morning, I saw this large black squirrel running away with a persimmon.  If it was cute, I wouldn't mind so much.  This one looks like a very large rat.

The worst part is that this squirrel squanders food, eating only parts of fruits from our yard and then abandoning them.  Our neighborhood raccoons eat them too but they only eat what they need and eat one completely before touching another.

Update:

While raccoons may not waste food, they sure make up for it in volume.  My wife told me today that many of the branches were broken and she found 65 stalk-ends at the bottom of the tree.  It must have been a family of hungry raccoons that did that.  Well, they are now a severely constipated family of raccoons.  Sweet.

Dave has an interesting idea this morning.  A reader suggested in my Right Roles for Standards Body post that Tim and I have a public debate on Atom vs. RSS.  Using the suggestion, Dave formulated a new type of debate that takes place on the blogosphere over a much longer period than traditional face to face debates.

Very cool.  Dave's blog-based debates can be implemented using a new type of blog category: Conversation Category.  The idea is for a small number of people to share a single blog category and converse over a long period of time through their blogs.

A conversation aggregator subscribes to the category feed of all the participants and merge them into a single feed and publishes a mini-website dedicated to the conversation.  The 'referree' of the debate or the conversation moderator gets editorial rights over the merged feed and the mini-website.  Hmm.  This stuff is very close to what I am currently working on so I think I'll slip this feature in while I am at it.

This idea reflects what is going on in each of our blogs.  Note that I often mention Dave Winer and Tim Bray by their first names.  That's because there is a persistent context that reduces ambiguity and an ongoing history to refer back to.  This context is not just limited to my blog but includes blogs of everyone I frequently mention in my blog (i.e. everyone on my blogroll).  A conversation category formalizes that context and gives it a more refined shape.

Frankly, this is the kind of innovation we need to be thinking about instead of reinventing what is already in wide use.

Looks like ActiveGrid finally unveiled itself.  One sentence description of ActiveGrid is: open source development tools and application server for grid + LAMP.

There is an arguably patentable technology underneath it all though.  If you were at NetDynamics while I was there, you might remember the cascading network of holes and plugs idea I mentioned in a few occasions.  This is exactly that.

I often feel as if I am living in a river of time.  When I was young, I didn't really care what might be downstream.  As I got older and experienced many harrowing turns of the river, I found myself looking farther and farther ahead.

What I just realized was that my sense of now changed over the years to include the future, near and far.  An event that will happen feels almost as real to me as an event happening now, just as the shape of the river downstream affects the flow of the river upstream.

FYI, I can now receive Skype calls.  If you are a skyper as well, you should update Skype.  Old versions of Skype had an buffer overflow vulnerability in the 'callto' URI handler.  The trouble starts as easy as clicking on a 'callto' URL longer than 4K.  Ouch.

Some software recommendations while I am here:

• Sam Spade is great Windows app for digging information on IP, DNS, etc.
• Fujaba creates UML diagrams from Java source.  It has an awkward UI and not much layout smart, but it's useful for figuring out what is going on in typically class-happy Java code.  ArgoUML is a better UML tool overall, but Fujaba is handy for this task.

NASA's unmanned jet X43A, powered by an experimental scramjet engine, flew under its own power for 10 seconds off California coast and reached close to mach 10.  Awesome!

X43A is the small (12 feet long) black thingy
at the tip of the white booster rocket
hanging under the B-52.

X43A accelerating to scramjet speed

I couldn't find any image or video of X43A flying under scramjet power.  Maybe they are classified.

BTW, NASA's scramjet project is being phased out, thanks to Bush's Vision for Space Exploration initiative which aims to send humans to Mars.  *eyeroll*  Damn it.  I wish we had more say in where my tax money goes.

Responding to my post Is Atom Ready for Prime Time?, Tim Bray wrote:

The right role for a standards body is to wait till the implementors have deployed things and worked out the hard bits, then write down the consensus on what works and what doesn’t.

I'll agree to the following version of the statement:

The right role for a standards body is to wait till the implementors have deployed things and worked out the hard bits, then write down the consensus on what works and what doesn’t without unnecessarily breaking or competing with what already works.

Atom feed format as it stands now creates too much distruption for too little gain.

Tim, do consider what works and what doesn't work in RSS 2.0 and fix it by adding to it or tightening loose parts of it instead of attempting to replace it entirely.  Then you can declare victory and everyone, even this grouchy donkey, will wholeheartedly join the victory parade.

As competition in the search space continue to heat up, I think the space is about to enter a critical phase in its evolution in which walls and roads will be built around and into popular or valuable areas of the search space.

The idea is not to restrict access but to restrict discovery or limit quality of search results, meaning that, while anyone can read those pages, they will have to use a specific search engine to find those pages or to get quality search results.

Walls will be erected around popular areas of the Web so that those areas can be searched only through a particular search service.  Major areas will be bought, shared, or locked in via some affiliate restrictions.

MSDN could be searchable only using MSN Search.  Likewise, Blogger.com blogs could be searched only through Google.  Amazon website could be made searchable only through A9.  Throw in alliances, affiliates (Amazon), and service dependency (AdSense) and time into the picture and the result is a the segregated Web.

How this phase unfolds doesn't have to be as ugly as I described above.  More palatable scenario is to open up previously unsearchable areas of the Web and make it available only through a particular search engine.  Another approach is to enrich search result using metadata not available to crawlers.

Update:

I am neither worried nor find the notion of walls in search space chilling as John Battelle wrote in reference to this post.  All businesses are, in essence, selling access to goods and services.  For goods that cost little or nothing to reproduce, selling access or quality of access is a legitimite business model.

Today's search engine results only show what they can scrape from the surface of the Web.  In order for them to provide higher quality search results, they have to dig deeper, into areas not directly accessible from the Web.  Whether or not to sell/leverage access to those areas is a business decision, not a moral one, unless there are privacy or ownership issues involved.