According to Tim O'Reilly's always interesting quarterly analysis of the book industry, Ruby is doing extraordinarily well, with a 689% quarterly increase in sales and is now approaching Perl in terms of book sales.

Caveats include the (some would say profound) difference between book sales and use. Most Perl programmers already have accumulated books on the topic. Similarly, C# showed a spike in book sales, but that is almost certainly largely a reflection on the release of VS2005 and not an indicator of a sudden shift towards that language.

Having said that, an important component of programming language popularity is "buzz" and the perception created by a sudden increase in books/seminars/articles. With sites like Digg creating "flash-interest," and increasing the volatility of the marketplace for attention, "buzz" may play an even bigger role in whatever The Next Big Language is than it did in the success of C++ and Java (two languages whose success was undoubtedly boosted by the amount of associated teaching / discussion).

On June 19, I posted 3 articles, "15 Exercises to Known a Programming Language," which came to the attention of Digg and was, for a few days, on the front page (and even the top item) in the Programming theme.

I've finally gotten around to reviewing my logs. While the article got about 8 times as many hits as my next-most-popular article ever (about ~40,000 hits), the click-through to the second and third articles dropped dramatically (5297 and 3919). I'm not disappointed by these click-through rates: each article was several hundred words long.

It's difficult to determine how many people subscribe to your RSS feed, since there is not a 1:1 relationship between hits to your XML file and "eyeballs." My blogging software (dasBlog), like many, puts a Web bug in each post, though, so I use the average number of aggregator-based reads as an indicator of whether my blog is gaining or losing ground. By that standard, it doesn't appear as if being Dugg made a long-term difference. From my baseline rate, I saw a 17% spike in June (the month the article appeared) and a return to the baseline (actually 94%) in July. (Hmm… I should switch that to median rather than average…)

In terms of immediate economic boon, I have a minimal AdSense presence on my Website. AdSense TOS forbids discussion of your actual income, but in the spectrum of latte-book-graphics card-rent, 40K hits from Dugg was in the high-latte / low-book range.

Thoughts:

The article was pretty on-topic for this blog, so I was hoping to pick up some long-term readership. While I may have done that, it's not apparent in the data. I have been having a long-term growth in my aggregator reads, and while there was a spike, July actually fell back a little (on the other hand, my posts in July have been pretty lame, because of my hardware problems).

The 40K direct hits are nothing to sneeze at. I imagine that if I had been pitching something (a book, a tutorial, etc.) and had embedded an ad in the article, that might have led to some income. However, I think in-post contextual ads are beyond the pale, so for me it highlights the fact that ad revenues for a programming related blog are trivial. (Having said that, I am convinced that blogging is easily the most cost-effective form of marketing there is for a consultant. Absolutely worthwhile.)

I finally got a dual-processor motherboard that boots (a Tyan S2885) only to discover that I only have 2GB of "registered" DDRAM, making my other 4GB of RAM worthless on this computer. Well, easy enough to deal with later, I think, happily screwing the motherboard down into my case. Now, slide the drive cages into... Now, slide the drive cages into... Wait a second...

While my Antec case technically "fits" an EATX motherboard, the drive cages are perfectly aligned in the airspace above the primary CPU slot. I tried all 3 cooler formfactors I've acquired in the past 3 months: none fit. So now I need to buy another case ($170 + $50 shipping).

Oh, and I forgot that while the motherboard booted, it didn't with 2 CPUs, so I spent an hour debugging RAM configurations before I noticed that the 8-pin CPU power plug only had wires going to 4-pins. So that was another order ($5 for the cable + $20 shipping).

I love Hawaii, but boy do I miss Fry's.

AMD unveiled new chip pricing today, trying to steal some thunder from the outstanding reviews garnered by Intel's Core2Duo chips. Personally, I've got a socket 940 thing going on, so I'm sticking with AMD. I'll probably build an Intel-based desktop when they ship quad-cores, which they claim they'll do by the end of the year.

This site claims there is a car with powered wheels that can do 0-60 in 1.67 seconds.

You'd have to have wheels made out of glue to do that, but assuming that's possible, I make that out as 1.6 gees as the _average_! Must be pretty hard to keep straight...

The C99 language standard added support for variable length arrays (VLAs):

void foo(int len){
  float myArray[len];
  ...
}

In porting some C# code to C++, I had a method signature like that, so I tried to use a VLA. Sadly, VC++ 2005 produces "error C2057: expected constant expression," "error C2466: cannot allocate an array of constant size 0," and "error C2133: 'myArray' : unknown size". I've confirmed that this is a compliance thing: the code works with

gcc -std=iso9899:1999

Word is that Borland has come to an agreement with a still-secret savior to purchase and support "DevCo": the former languages and tools division.

Details as they emerge, but that won't be for a few weeks

NaturallySpeaking 9, coming out in August, claims to dramatically reduce the time it takes to model your voice, achieving the best-possible recognition soon after opening the box.

For some people, that best-possible recognition is said to be 99%. Maybe. I've probably gone throught the "voice training" process a dozen or more times over the years. Not only have I never achieved 99%, I've never achieved anything usable.

There are several factors: one is that "tethered to your computer, wearing a noise-cancelling headset, and watching the dictation in realtime," is not appealing to me. The second is that when you make a typo you are off by a coupe letters and then you get back on track. When a voice-reco system fails, the error mode is a parlor-game chain of semi-homonyms "wrecks a beach" == "recognize speech".

I'm ever optimistic, though. As a writer, I'd love to be able to do significant amounts of work using a digital recorder (PDA, smartphone, what-have-you) on the beach. I've even thought of trying out those lost-cost (human) transcription services. Maybe I'll give that a shot this National Novel Writing Month.

The last thing I saw last night was something about "Do not turn off or remove the power cord from your computer" for some no-doubt-critical update. This morning, my desktop doesn't POST. It hangs right at the point the RAM count happens.

Now, I don't think an OS upgrade can effect POST, so perhaps it's a bad RAM chip. But I pulled all but one chip this morning and it still didn't boot. Then I replaced that chip and it still failed. So the odds against it being the RAM seem high.

Meanwhile, on my new system that I'm building to replace the motherboard whose hard-drive controller blew 6 weeks ago, the replacement motherboard doesn't POST at all: no BIOS announcement, no nothing.

Power's coming through a UPS, which should condition it adequately.

At the moment, I have no desktop. I _so_ don't need this.

Oh, and you have to be nuts to buy new chips and motherboards before next Thursday, when Intel launches their new chips and AMD will undoubtedly respond with huge price cuts.

When Software Development was killed, I predicted that Dr. Dobb's wouldn't change markedly. Boy, was I wrong. Editor-in-Chief Jonathan Erickson and Publisher Stan Barnes clearly decided that the time had come to create what is essentially a new magazine: I don't think Dobb's has changed this much since at least the late 80s.

The "new" Dobb's really has taken a page from SD's playbook and dramatically increased the amount of technical-management-focused editorial. There's a fraction of the source code that used to be Dobb's signature.

They've dramatically changed the column lineups, which is a real surprise given Erickson's loyalty to his long-term writers. Mike Swaine is still there, Scott Ambler and Rick Wayne came over from SD, and Pete Becker is writing the C++ column. This was a bold move must have been a hard one, both for the columnists and for Erickson. But  think it was a good choice.

They also redesigned the pages. Note to publishers: Do we really need to go through the whole "the new font is too small," "you're right: we've changed it back!" charade every time? It looks like the new page layouts are more flexible, although at least initially, I think the readability has gone down.

As an old-time competitor to the "old" Dr. Dobb's, they've walked away from some of the things that made them hard to compete against: the "signifiers" of technical depth that came from their source-code and low-level articles. Pages of source code cue programmers "there is immediate value here." When flipping through a magazine, an article on, say, computer security, will be much more eye-catching if there is accompanying source code: the programming reader stops and "checks out" the source code to see what's going on. "Soft" articles, on the other hand, have a harder time catching the eye and coming to mind when the renew / resubscribe decision comes about.

It's gotta' be tough managing the editorial of a programmer's magazine nowadays.

One of the cable channels (FX, I think) has been playing "The Core" in medium rotation lately. I've been trying to expose myself to it in small amounts, to inoculate myself and accept it as cheesy "so bad it's good" fun(ref. "The Fast and the Furious": One of my favorite movies of the past few years.)

A few years ago, during a bout of business travel, I saw "The Day After Tomorrow," like, 7 times on various airplanes and I really thought that was the worst that could be done, but I just saw the last 10 minutes or so of "The Core" yesterday. OMG. So bad. So very, very bad.

InfoWorld's Tom Yager wrote a column on the benefits of native code, but then went off the deep end with:

Here’s a native code prediction that’s way under your radar: We’ll see more use of assembly language. ...Developers coding for new, controlled deployments can afford to set high requirements that include a 64-bit CPU, OS, and drivers. And if you know you’re coding for Opteron and you’re ready to write to that architecture, baby, life is a highway.

[via James Robertson]

I worked with assembly language. I knew assembly language. Assembly language was a friend of mine. And I have this to say: Assembly language isn't coming back to the mainstream.

The point that native-OS code is worthwhile is dead on. The point that native-hardware code is worthwhile is, for numerics and media programming specialists, true: if you're working with huge blocks of 8- and 16-bit integer data, packing the registers and using the wide-data ops is going to be worthwhile (assuming we're talking about code that will be run hundreds or thousands of times). But the only reason to drop that far low-level is parallelism and assembly has very little (if any) advantage for expressing that. Even in the graphics world, where concurrent ops are already the norm, the trend has been towards higher-level languages (relatively speaking: shader languages look like C).

The idea that the concurrent revolution is going to be solved by old tools is dead wrong. Low-level C-derived tools? Possibly. (Or possibly not: a higher-level language that did for concurrency what Java did for memory management [solve 90+% of the problem in a relatively performant manner] could very well sweep the industry.)

When I feel listless, I sometimes try to whet my brain by rubbing it on quantum mechanics, which requires math that's absurdly difficult for a dilettante to understand. For years I've tinkered at implementing a simulator for programming quantum computers and really haven't gotten anywhere. Well, now I can use André van Tonder's Scheme-based DSL for quantum computing. I really doubt that I'll ever understand it well enough to create a unique algorithm, but the point is really just to have something "whose reach exceeds my grasp."

"Java in a Nutshell" weighs in at 1264 pages. Matt Croyden, sez:

[Y]our programming language just might be complicated when you have trouble telling the difference between its Nutshell book and a telephone book.

[via James Robertson]

This is somewhat unfair, as the bulk of "JiaN" is a library reference, but it's certainly true that Java and C# have grown more complex as they've evolved, while certain other languages (Lisp, Smalltalk) have seen continuing simplicity as a feature of the language.

I think that one force in play in the market for programming language popularity is pressure towards a "collapse toward simplicity." It's not the only force, in my opinion it's not likely to be the major force, but it certainly played a part in the rise of Java. Of course, Java was equally an example of the force towards familiarity: it seemed quite like C++. Similarly, I think one reason why Ruby is currently the belle of the ball is its similarity to Perl.

According to "How UML Is Used," an article in the May 2006 issue of CACM, the UML diagrams that most commonly "provide new info" above-and-beyond use-case narratives are:

1. Class diagrams
2. Statechart diagrams
3. Sequence diagrams

Interestingly, "usage rates are not well explained by how much new information is provided." Statecharts, the 2nd most useful diagram, are used in most projects by only perhaps 1/4 of practitioners. Use-case diagrams, in comparison, are the 2nd most commonly used type of chart, but are one of the least effective in terms of adding value to use-case narratives (well, yeah...). Class diagrams are both the most useful and most used, while sequence diagrams are commonly used by about half the practitioners.

Rounding out the studied diagrams (they skipped Object, Component, and Deployment diagrams), Collaboration and Activity diagrams are, when used, considered useful by more than 60% of practitioners.

With all my AI posts lately, I'm sorry I hadn't realized that the May 2006 issue of the CACM had a theme on the language-action perspective, a critique by Terry Winograd and Fernando Flores that dates from 1986 whose essential point the CACM summarizes neatly:

[S]killful action always occurs in a context set by conversations, and in the conversations people perform speech acts by they commit to and generate the action. Expert behavior requires an extensive sensitivity to context and an ability to know what to commit to. Computing machines, which are purposely designed to process symbols independent of their context, have no hopes of becoming experts.

It's a cutting insight and goes, I think, to why expert systems, for instance, initially seem very exciting but, in the real world, generally fail to provide a lot of value. (They're great for training operators, though!)