I was just reading "8 Tips and Tricks for Better BizTalk Programming" in the May 2007 MSDN when the word "Frisbee" in the author's bio caught my eye.

I was pretty good with a disc (3 world records, all of which have been long eclipsed), but Scott Zimmerman was the Michael Jordan of disc. I think he held every throwing record, probably multiple times.

Reading that just sent me down nostalgia lane: Skippy Jammer, Crazy John Brooks, the Coloradicals ... Southampton, La Jolla Cove, Mission Beach Rollercoaster ...

Geez, now I'm going to have to learn BizTalk just for the offchance of jamming with Scott at some development conference...

After Pele rolled over, she's gone into a deeper slumber and the Big Island is experiencing the lowest level of volcanic activity since 1983. Such pauses have happened twice before during the current 24-year-long eruption and previously lasted just a few weeks.

The effect on the Kona (leeward) side of the island is dramatic. For the whole time I've known this island, the 14,000' Mauna Loa creates a huge atmospheric eddy in which the volcanic aerosols are transformed into "vog" (volcanic smog). This makes Kona "normally" hazy, with an indistinct horizon and, for some, noticeable effects when exercising. A few times per year, when the winds shift, and the sky becomes blue, it's literally like a scrim being lifted. It's been like that every day for the past couple weeks.

No one expects this pause to be long-lasting, but for the moment, it's marvelous.

I am setting the publish date of this post to tomorrow (6/30) at 6:00:00 AM, although I am publishing it on the 29th at 7PM.

Wii has announced that they will be opening the door to small developers on the Wii. The first take on this is that "WiiWare" is primarily about downloadable games a la XBox Arcade, but we all knew that was coming. More surprising is that Nintendo says that they will not apply their notoriously thorough and expensive vetting process to such games. No details on the SDK: whether it will be C-based or some form of managed language and no word on whether it will be free.

Apparently, that my XBox 360 lasted from the day it went on sale until today is unusual. I was negotiating a turn in Forza 2 and the system froze. Rebooted, it froze during race startup. Rebooted, and now it's "the Red Ring of Death."

According to this site, I can expect to be told that it will cost $140 to get a refurbished replacement, but if I bitch I'll get 25% off. Oh well, maybe the replacement will be quieter -- there's no way the fans I had on this thing were up to spec.

We have a database (SQL Server) that has a field in which we store fairly-hefty (1-4MB) XML documents. We need to create reports. Lots of reports.

The schemas upon which the documents are based are quite complicated and we have thousands of records.The reports need to be modifiable by inexperienced programmers, so a visual report designer is an absolutely necessity. Any opinions on good tools? (Big Faceless has high Google rank. Anyone use it?)

I d/l'ed a demo of Crystal Reports, but:

• Found it hard to configure my data source and schema (and even though I thought I finally got it right, although I can now see the "tables" and "fields" from my schema, it still says Records: 0)
• Am not sure I can configure it to use an XML database field as a datasource

Am I a fool to use Crystal Reports or just a fool to not have figured it out?

Anyone looking for some freelance report-generating work?

Now that IBM has a petaflop supercomputer, the question becomes what to do with it:

10. Develop a coherent exit strategy from Iraq

9. Crank call Gary Kasparov at 3 in the morning and ask in that creepy computer-voice: "Would you like to play a game ... bitch?"

8. Write a script that saves the Geico Cavemen sitcom from being canceled after 3 shows

7. Port Rails to VBScript

6. Explain the plot of Pirates of the Caribbean: Dead Man's Chest

5. Open the pod bay doors (Or not -- let the computer figure it out.)

4. Hook up with slutty, esteem-challenged teraflop computers by telling them you think they're still hot

3. Explain Paris Hilton's popularity

2. Calculate shorelines post-Greenland icecap meltdown. Give seminars on how to buy to-be-oceanfront real-estate, particularly distressed and defaulted properties, for no money down!

And the #1 use for a petaflop-capable supercomputer...

1. Barter it for an iPhone

Rationale is an interesting program that allows you to visually represent arguments:

It is essentially a domain-specific Visio or MindManager. Since it is domain-specific, it is faster to construct a complex tree and additionally export the structure as a text outline. This also limits it a little, in that I couldn't find a way to associate a single point with multiple higher-level positions ("Ruby is easy to learn" relates to both training developers and maintenance). While it could be argued that one should continue to decompose one's arguments until they fit into a neat hierarchy, that seems a little over-zealous for the general case. (Incidentally, I didn't use the product to structure this paragraph, but it turned out quite close to the sort of text generated by Rationale.)

It's fun to use and you can rapidly construct arguments either top down ("We should use Rails for our next application" because ... ) or bottom-up ("Ruby's interactive console allows active exploration" supports ... ). It does not attempt to parse your logic (it won't balk at "because I say so") but it can help structure an argument for conversation and review.

As long as everyone agrees that decomposition is a productive route and as long as people aren't so cagey that they recognize that structuring a debate is 3/4 of the way to winning it. That's my biggest problem with the product; at $199 for a perpetual license or $100 per year, it's a little pricey for something I don't think you'll be using in many meetings. However, as an educational tool, I think it's a good bargain. The educational rate is $49 for a perpetual license or $24.50 annually, and I think this would be a very good tool for students learning rhetoric, debating, logic, or composition.

You can download a 30-day trial version from Austhink.

 Charles Petzold has a post in which he talks about the unique experience of reading a book . In the post, Petzold mentions a device that caused a buzz at a recent O'Reilly conference; the device was a book that somehow embedded a screen for displaying hyperlinked content (from the picture, it looks like it used some kind of flexible screen). Petzold says that while this sounds good initially, it will inevitably lead to more and more ancillary "fat."

I think he's right. I remember that the first few Shakespeare plays I read were from school-provided texts. They had footnotes for virtually every sentence and they were incredibly distracting ("Fardel: A burden.") It was only when I learned to willfully ignore the footnotes that I began to understand why people love Shakespeare. (Exception: ignoring footnotes is not recommended when reading Nabokov.)

eWeek is quoting an Evans Data survey in which Borland's StarTeam was rated the best source control management (SCM) software, beating out CVS, IBM ClearCase/ClearQuest, Microsoft Visual SourceSafe, Microsoft Visual Studio Team System, Perforce, Serena/PVCS, and Subversion.

It's an interesting result, because surveys have a strong tendency to correlate with marketshare, but surely StarTeam does not have dominant marketshare. Meanwhile, SCM tends to be a "good enough" solution, where people generally stick with what they already have. Is StarTeam so exceptionally good that it can overcome those tendencies?

Update: Evans' methodology is to survey users of the tool in question, which ought to overcome the "marketshare == survey results" problem that one gets in most "reader's choice" surveys. On the other hand, it may inflate the influence of cognitive dissonance; people invested in niche products (StarTeam has about 9% marketshare, according to BZ Research) have a psychological pressure to praise them more than do people invested in market leaders (who, after all, know the market has endorsed their decision, making criticism come a little easier to the tongue).

I don't have deep opinions about SCM tools (except about Visual SourceSafe, which I despise), but I asked one of the judges in that category for the Jolt Awards his opinion. He recalls Evans "finding that among Java IDE users, the preferred IDE belonged to Rational. This struck me (and others) as being implausible, but not impossible. This survey result, however, does strike me as quite impossible."

Having said that, I solicited Eric Sink (of SourceGear, a competitor of Borland's) for his thoughts and he responded in the comments section. Sink characterized StarTeam as a product that is in the category of SCM tools that are "mostly or somewhat liked" by their users.

I've only used StarTeam momentarily, quite awhile ago, so my skepticism about the results may be sheer ignorance of a great product.

Words cannot express my admiration for this.

A few weeks ago, Scottevest Cargo Shorts were on sale. I've never owned a Scottevest product before, but they're well-reviewed, and if there's one piece of clothing a Hawaiian geek requires, it's capacious cargo shorts.

They're quite good looking and can handle a full load of iPod, wallet, digital camera, and phone. However, they have a critical flaw, of which I'm surprised given the company's clear understanding of their audience: there is no pocket that accommodates a Moleskine Reporter Notebook (or the slightly smaller and more casual Sherbert Notes 7"x5"). The "big" pockets on the Cargo Shorts are cut with an angled entry that writing-sized notebooks can't negotiate (see photo).

Of course the shorts can handle notecards or memo pads, which are sufficient for to-do lists and Hipster PDAs, but have you ever tried to record a non-trivial thought on a memo pad? Doesn't work.

Perhaps the next release will solve this critical bug.

Alan Turing was born 95 years ago today. Less than 100 years ago. I know that at the physical level, information processing is nowhere near as dramatic as flight or the rise of the car, but it's still astonishing to reflect upon the advances. I've been drafting an article about the connections being discovered between computation and physics (both thermodynamics and Riemannian geometry), fields where there is a palpable sense of impending breakthroughs. I've never understood why there's so little discussion of the science of computation and information, which is still a field that, like biology in the 18th and 19th centuries, is broadly accessible and one where I am convinced amateurs and dilettantes can make major contributions.

 NVidia's Tesla C870 Graphics Processing Unit (GPU) will be the basis for a "deskside supercomputer" add-on that will provide highly-parallel high performance computing (HPC) capabilities, presumably programmed with NVidia's CUDA toolkit.

Dedicated hardware for HPC has always been a treacherous market -- one year's darling is next year's has-been (people used to buy Cray Supercomputers at auction and resell them for the gold in the connectors. True story.). Dedicated processing boards for desktop computers have always been especially troubled, as the system bus is such a bottleneck and Moore's Law used to provide such wonderful free lunches. (No longer true, although the bus issue is potentially more dramatic than ever.)

There is infinite demand for HPC from 3 well-funded sectors: economics (trading), bioinformatics, and chemistry (bio- and otherwise). These sectors will absorb any amount of information processing capacity available. Whether that can be translated into commercial success for NVidia, or whether they unlock additional markets, is far less certain.

I wonder if Google will buy a couple boards.

Takeaway for programmers: Feverish hardware activity relating to concurrency continues. Software lags, with only relatively low-level toolkits available for exploiting the system. Keep your C skills sharp.

Ziff Davis is selling for $150M its Enterprise Group, whose assets include Baseline, CIO Insight, eWeek, and microsoft-watch.com. Unfortunately, that leaves ZD with still around ~$240M in debt, which they must pay off using their Consumer/Small Business Group (which published PC Mag) and their Game Group.

The purchaser was Insight Venture Partners. I can't imagine that they've got a plan to flip it -- I don't think any publishing company is hankering to make such an investment. So that leaves slicing-and-dicing.

Miguel de Icaza has unveiled "Moonlight," an implementation of Silverlight on Linux by way of Mono. The project was done as a 21-day sprint and while just a prototype, it makes Microsoft's new in-browser managed platform available on the 3 major desktop contenders.

I remain of the opinion that Silverlight is going to be a major platform for Microsoft, siphoning off a lot of developers who otherwise would be looking at .NET / desktop CLR. And while Mono has not seen the uptake that I think it deserves, the availability of Silverlight on Linux is important for Silverlight's acceptance.

I was just scanning my latest copy of one of the very last independent software development magazines (independent as in "copy not subject to approval by vendors") and saw an article on REST. It seems intuitive to me that if you're a programming magazine today, you compete on clarity and authority. The article, in fact, was written by one of the magazine's contributing editors and I thought "Ah! A 1500 word overview of REST -- how valuable!"

Take a look at the core code of Listing 1:

protected void doPost(HttpServletRequest req, HttpServletResponse resp){
...
Employee emp = lookupUser(userSSID);
String medPlan = emp.getMedicalPlan();
String dntPlan = emp.getDentalPlan();
String retPlan = emp.getRetirementPlan();
Response = "User " + emp.getFullName()
   + " has medical plan: " + medPlan
   + ", and dental plan: " + dntPlan
   + ", and retirement plan: " + retPlan;
out.println(response)

Believe me, I understand that I have a lot of glass in the walls of my house, but it's a really big mistake to have a tutorial on REST that uses the HTTP POST verb to retrieve existing data. The use of HTTP verbs for distinct purposes is central to REST principles. The author seems to be unaware of this.

One reason I've not posted the name of the magazine is that, although this magazine is likely to have been in the hands of its many tens of thousands of subscribers for weeks, I've not heard the slightest ripple of outrage in the blogosphere and I'd like to see if that continues.

Obviously, this prominent magazine's feature article on an architectural topic of great interest and passion has gone unremarked. What does that mean? There's been a great deal of buzz lately about "whether alpha geeks have given up on Windows" but I find it even more disturbing to think that the alpha geeks have given up on ink on dead trees. Surely one of the roles of experts is to police the "mainstream media" and not simply to piss on each other about esoteric corner cases.

Or is it the case that this particular magazine has lost its credibility and that such a mistake is considered no more worth pointing out than the White House making an overly-optimistic prediction about Iraq?

A big earthquake swarm on the SE side of the island is "consistent with a shallow intrusion of magma" at Kilauea / Pu'u O'o. They don't predict eruptions, but I have a feeling that Pele might be restless. Luckily, that's 60 miles away and on the other side of a 13,000 foot mountain.

This isn't "fool me twice, shame on you," it's like "fool me every freakin' year for the past decade." At least Norton 360 doesn't seem to consume huge portions of my CPU constantly.

I hesitate to call Mark McKeown's Brief History of Consensus, 2PC, and Transaction Commit (via just about everyone, but let's say Bill de hÓra) a "blog post." It reads much more like a darn good professional article.

If you're interested in having an informed opinion about concurrency (as opposed to waiting half-a-decade and accepting what the market has decided is "good enough"), the article is a must-read. As we get to the manycore era, the amount of asynchronicity within a single machine will be significant. We're already seeing hints of this with Non-Uniform Memory Access (NUMA), in which different cores can access main memory asynchronously.

So in order to think knowledgeably about what concurrent programming ought to be like, the best crude model is distributed programming. With a huge caveat, which is that you can't think just about the network messages as "the system," you have to think about keeping the local processor busy. So don't think about Google Maps, think about Forza 2 on XBox Live. (Mmmm... Forza 2 on XBox Live ... )

Can't talk about it, but I demo'ed something today. I love it when the chewing gum and baling wire doesn't show.

That was a lot of fun. It's been a long time since I've done a big stakes demo.

Ambiguous? Of course. But c'mon, there's clearly one reading that gives David Chase credit for being brilliant.

Having said that -- Could anyone stand watching more than the first 15 minutes of John from Cincinnati? I only made it that far because it was David Milch.