Thursday, March 31, 2011

New Blogger Looks

Google announces “five new dynamic templates in Blogger that you’ll soon be able to customize and use for your blog. These new views use the latest in web technology, including AJAX, HTML5 and CSS3….”

Here's what they look like on this blog.

Interesting, you can apply them to any blogspot blog by appending /view to the URL. That's an interesting approach to customizing.

Tuesday, March 29, 2011

YADC: The Deathstar

A [VIDEO] tour of Lucasfilm's/ILM's data center by Greg Grusby of ILM, blogged by Arik Hesseldahl and via Data Center Knowledge.   (YADC:  yet another data center).  It is interesting to see him point out the old machines they used for Star Wars Episodes 1–3.

Saturday, March 26, 2011

MIT NSE Nuclear Information Hub

The MIT Nuclear Science and Engineering blog is an excellent source of information and explanation on the nuclear situation in Japan.  The short answer:  There's a lot of unnecessary fear, uncertainty and doubt, some of it malicious.  Though the situation is serious, there's just not a lot to worry about.

What's This About IPv4?

We've run out of addresses so now what?

I decided a few weeks ago it was time to think about this a bit and try to come to some reasonable conclusions. I wrote a really long and rambling, i.e., useless blog post which is still sitting in the draft bucket. Now I'm going to take a stab at a shorter summary.

So, how will this problem affect you and me?

It may cost money

There's now a shortage of IPv4 addresses meaning they are scarce resources. That's the definition of an economic problem and the response will be rising prices. I think that the more you care about what your IP address is, the more you'll have to pay. If you don't care, maybe you don't even know what an IP address is, then the impact may not even be noticeable.

For the most part, the ones who care about IPv4 addresses are those providing services. They'll have to pay more and, in some cases, charge their customers more.

There will be down time

Still, a lot of changes will need to be made in the networking infrastructure that underlies the world most of us live in. That means large projects will be undertaken, some in a panicked rush, and mistakes will be made, even in the best of cases. For you and me, that means web sites will go down, we'll lose access from time to time and other inconveniences will pop up without warning. Think of big highway construction projects.

And finally,

IPv6 is not a solution

IPv6 is not a solution to the IPv4 problem. It's an entirely new undertaking that's so huge, I claim it's an entirely new problem, even bigger than the IPv4 challenge.  Claiming IPv6 solves the IPv4 problem is like saying we can fix some local problem on Earth by terraforming Mars and moving there. Well, in an indirect way that would render the original problem irrelevant, but it's not an easy or affordable solution and it has a whole lot of details that aren't worked out yet.

I continue to think that the current IPv4 problem can be architected away using network address translation (NAT). We've pushed off the scarcity of numbers this way over the years. Now it just needs to be done on a bigger scale, and it will be accompanied by the problems I've predicted above.

Warning! Take Heed!

Fake USPS Emails in Circulation posted on Softpedia by Lucian Constantin. “A wave of fake United States Postal Service (USPS) emails currently making the rounds are trying to pass a trojan downloader for a shipping label.”

The spam emails pose as failed delivery notifications and bear a subject of  “Post Express Information. Your package is available for pick up.”

The contained message claimed that an error in the shipping address caused the package to be returned to the post office, from where it can be retrieved.

“Your package has been returned to the Post Express office. The reason of the return is ‘Error in the delivery address’ Important message!

“Attached to the letter mailing label contains the details of the package delivery. You have to print mailing label, and come in the Post Express office in order to receive the packages” the emails read.

The attachment is called Post_Express_Label_ID_[number].zip and contains a malicious executable of the same name.

My personal advice continues to be the same: Avoid clicking on links in the email message. Instead, type in the address by hand in your browser to go to the page and see what's up.

OO vs. FP

Finally, a concise and perfect resolution to the question of object oriented programming vs. functional programming, assuming the debate is even an appropriate one in the first place.  I've stuggled along with this issue myself without finding a satisfactory resolution.

Today a blog posting announced, and was echoed on Slashdot, CMU Eliminates OO Programming for Freshman [Sic.  I presume it's for more than one freshman].  In the ensuing discussion, bradley13 posts a reply that captures and resolves this issue in the most concise and accurate way I've seen.

OO is practical for lots of problems, because it makes modelling real-world data easy. However, it is not useful if you want to give students a solid understanding of the theoretical computer science. OO is fundamentally data-centric, which gets in the way of algorithmic analysis.

To give a pure view of programming, it would make sense to teach pure functional and pure logic programming.

That's it!  FP is useful for expressing and teaching algorithms!  Now, having a programming language that doesn't even include I/O makes sense.  Now I understand why colleges wanted to use FP (Scheme and such) for their introductory classes.   You can just settle into looking at the pure algorithms with the same mathematical purity as you study  2 + 3 = 5 or a proof in geometry.

That's also why FP seems so difficult to use in the real world, at least to me, in spite of well-written and -spoken arguments, and perhaps even well-documented examples, to the contrary.

On the other hand, OO is indeed well-suited for writing real programs and solving real problems.  It is imminently practical.

Now it makes sense.  The theoreticians scoff at OO while practical coders are sometimes bewildered by FP.  (Okay, it could be that only I am befuddled, and the FP proponents do keep saying that you just have to see the light).

There are some underlying themes in the discussion on Slashdot I agree with.  One is that I don't think any education program should start with OO.  I agree that a program should begin with the simplest programming environment possible   Most folks around my age, and some younger, learned FORTRAN or BASIC first.  That seemed to be a perfect introduction.   And yes, I'm well aware of the FP crowd that is shouting at this point:  FP is the epitome of purity and simplicity!

In fact, I've heard that some teaching programs are switching to Python and I think that's an excellent solution.   Python can serve in all of these roles, from a simple, BASIC-like beginner language, to a functional language for studying theory, to an immensely practical language for solving some of the most difficult  computing challenges.

To bradley13 I say, Thank you for clarifying this discussion for me in such economic terms.  Well done!

Your Kidding Me! Thunder?

Who ordered that?

Friday, March 25, 2011

What's Wrong with the World

Color syntax highlighting.  I could cry.  I used to write beautiful code that was completely readable on a properly printed black on white laser-printed page in Courier font.  The programs were somewhat literate and completely manageable.

Now I seem to be completely dependent on colored syntax highlighting because my code is brief and basically illiterate.  Thank goodness I found go-mode.el for emacs for my Go language programs.

Sunday, March 20, 2011

The Super Moon

I saw the so-called super moon last night after it had well-risen high in the sky.  It was a beautiful and bright full moon.  However, it didn't look that different to me from any other full moon.   Phil Plait always does a good job of explaining these kinds of phenomena.

Amazing Earthquake Visualization

From the University of Alaska Fairbanks Geophysical Institute, here's an amazing report on the Sendai 2011-Mar-11, 05:46:23 UTC Earthquake featuring GPS Kinematic Solutions and accompanied by fantastic videos.

Happy Vernal Equinox!

Happy vernal equinox today at 19:21 EDT  (23:21 UTC).

Sun 2011-03-20 19:21:00 -0400
Sun 2011-03-20 23:21:00 -0000

Tuesday, March 15, 2011

Google's Quest to Build a Better Boss

From the NYT by Adam Bryant.

Fukushima Nuclear Accident

A simple and accurate explanation by Barry Brook.  This seems to be a well written and correct explanation of what has happened in Japan, sent to me by Phillip.  The article is long, but not as long as you might think from the web page size—most of the page consists of the follow up comments.

Note that I don't (necessarily) endorse the web site where the article is found.

Here's the most disheartening moment in the middle of the article—the point where the actual failure occured.

When the diesel generators were gone, the reactor operators switched to emergency battery power. The batteries were designed as one of the backups to the backups, to provide power for cooling the core for 8 hours. And they did.

Within the 8 hours, another power source had to be found and connected to the power plant. The power grid was down due to the earthquake. The diesel generators were destroyed by the tsunami. So mobile diesel generators were trucked in.

This is where things started to go seriously wrong. The external power generators could not be connected to the power plant (the plugs did not fit). So after the batteries ran out, the residual heat could not be carried away any more.

Here are the first three of the author's final conclusions.

  • The plant is safe now and will stay safe.
  • Japan is looking at an INES Level 4 Accident: Nuclear accident with local consequences. That is bad for the company that owns the plant, but not for anyone else. 
  • Some radiation was released when the pressure vessel was vented. All radioactive isotopes from the activated steam have gone (decayed). A very small amount of Cesium was released, as well as Iodine. If you were sitting on top of the plants’ chimney when they were venting, you should probably give up smoking to return to your former life expectancy. The Cesium and Iodine isotopes were carried out to the sea and will never be seen again.

The Power of a Flood

It wasn't that long ago that I came to realize the real power and  peril of a flood isn't just the rush of a wall of water, but it's the fact that the raging water is full of rocks, trees, cars, houses, etc.  This video illustrates that with disturbing clarity.

Saturday, March 12, 2011

Caught Up

Well, that catches me up for the week, I think.  I've decided to favor blogging over Twitter for posts that I think are somehow significant since Twitter has become so ephemeral.  Right now that means I email to them to myself instead of retweeting them but then there is a delay before I get around to posting them.

There may be some tool out there that allows re-blogging a tweet, but that capability isn't in my favorite Twitter client, Seesmic on Android.  Such tools I have tried don't construct the blog post in they way I want it and so just aren't useful.

Tsunami in Northeast Japan


AMD Building a Large Data Center

Near Suwanee.  From the Gwinnett Business Journal.

Happy Birthday ZX-81

The little computer was released on 1981-03-05 as the successor to the ZX-80 (1980).

I fondly remember a kit version of the ZX-80 that a group of us bought and built at the planetarium.  I wrote software for it to reduce variable star observations we made with a photometer we'd bought for the observatory.  The work of building it was really done by one of the planetarium associates, Ron, and he took the interesting approach of building it in something like an old portable case for some type of tape recorder I think.  He also built a little keyboard with actual push buttons as an alternative to the little membrane keyboard the kit (and the ZX-80 computers) came with.

Secret X-37B Space Plane

A launch was planned for 2011-03-05.   Apparently it succeeded!

The Artificially High Price Of Academic Journals

And How It Impacts Everyone by Mike Masnick at

Go Computer Language

Go is a new programming language designed and written from scratch by Robert Griesemer, Rob Pike, and Ken Thompson (Wikipedia).  The main web page is at  There's a wonderful little playground tool there which let's you try out Go right in your browser.

A little informal study group at work has formed to look at Go so I've finally started playing with it a bit myself.  Colleague Jeff McNeill started this off with a series of blog posts on Go, which are quite well done.

Here are a couple of outstanding videos which tell you a lot about the language.

Talk by Rob Pike

Andrew builds a URL shortener

Interesting Ngrams

I finally got around to playing with the Google Ngram Viewer.  Here's an interesting example comparing IBM, Microsoft, Google, Linux and UNIX.

What they say is true, you can spend a lot of time playing with this!

Thursday, March 03, 2011

Hooray! Seesmic Gets

My favorite Twitter app on Android now supports the URL shortener:  Seesmic on Android.

In the meantime, I've really been enjoying Tweetdeck as a Chrome App on the desktop.