Sunday, March 11, 2007

The limits to DRM: my new car stereo

I'm not, by nature, an optimist. I tell my friends that not only do I consider the glass half-empty, I suspect the dregs are poisonous.

So it is not surprising that, for some time, I was pretty pessimistic about Digital Rights Management. The public had no reaction to the DMCA at all. Would the voting population figure out the real costs before Microsoft started selling DRMd retinal implants? I didn't think so.

I started to become a less pessimistic when I realized, a bit ahead of the curve, that it was very hard to manage any DRM solution in a world of disparate disconnected embedded computers [1]. I became slightly optimistic when I formally admitted that humanity's actions, on occasion, seem inexplicably less-than-dumb.

Recently, my new SONY car music player has tipped me into the moderately optimistic range. Ironically, and perhaps not coincidentally, this product comes from a company with a historic (truly) DRM fiasco. I wonder if Jobs was thinking of this sort of product when he wrote his "Nixon in China" essay.

So why does this car player possibly signal the doom of today's DRM? The key is that the player supports MP3/AAC [2] CD-Rs and USB mass storage devices in addition to the iPod. The iPod support is great, but the data CD-Rs are ultra-rugged, cheap, disposable, reliable and very simple to use for playlists. [3] They're also very well supported by iTunes -- as long as you don't have music from the iTunes store. The stereo doesn't support FairPlay, and it can't.

Until now I've allowed a few DRMd tunes to leak into my collection. Freebies mostly, one or two impulse buys, and some gift cert music for the kids. No more. Sure I can easily build iTunes smart lists to filter out protected music [5], but I'm a geek. Even so, it's a nuisance to have to burn my kids favorite tunes to a CD, then re-encode as non-DRMd AAC [4].

Americans live in cars these days. Large personal music collections are a natural fit and devices like this excellent Sony product will become ubiquitous. DRMd music doesn't work in this setting. We are going to hit the limits of this generation's stab at DRM for music very soon. I'll be a rampant optimist and predict complete collapse within 18 months.
[1] Our sump pump has an embedded computer system monitoring its health. That computer crashes every few weeks and has to be rebooted. My alarm clock crashes every few weeks. My car stereo has a reset button. The Daylight Savings Time transition is going to remind everyone how many embedded, disconnected, non-updateable, computers they own; we'll be resetting our camera and video clocks four times a year until the last one dies.
[2] And Sony's irrelevant ATRAC standard.
[3] See my review for more details. Using Apple iTunes it's trivially easy to burn a half-dozen data CDs made up of both AAC and MP3 tunes, and very convenient to pop them in a play them. I find it easiest to treat each CD as a unique playlist and not bother with folders, navigation hierarchies, etc. A 700MB playlist is not bad really.
[4] Sure there are applications to do this, but we have so few that matter the CD solution is easier.
[5] Omit KIND contains "protected".

Does Google understand hyperlinks?

Does Google understand how a hyperlink works? I'm beginning to wonder.

I rag on Microsoft Sharepoint 2007 big time for creating a supposed "web" product that uses the file system semantics to generate the URL, thereby ensuring that any file system manipulation will break any link, but really, Google is no better.

The original Blogger allowed users to change blog titles without breaking links, but after Google bought the first revision made the links reflect the title -- so editing a title broke links. The newest Blogger revision [1] changed that, the url is based on the original title but it doesn't change on editing the title -- however changing the date will break the link.

Then there's Google's Picasa Web Albums. Try this experiment:
1. Create an album.
2. Share a link to the album.
3. Edit the album properties and change the name. Note there's nothing to warn you of the consequences of this seemingly benign act.
4. Every link to the album returns a page-not-found. No redirects.
Google's supposed to be composed of very bright math majors, so I'm guessing this isn't mere stupidity. I can only speculate on what's at work, but it's more than annoying. I believe, but have not confirmed, that SmugMug, for all its recent buginess, does at least get this right.

[1] Google may have some warped beliefs about links and metadata, but I can now say, with great caution and trepidation and painful memories of a semi-botched data migration process, that the new Blogger is far better than the old one. Now that I've said that Google will certainly proceed to wipe out all my data ...

Saturday, March 10, 2007

Google and data lock -- not being evil

I was pleasantly surprised to learn that Google has taken a public stance against using data lock to retain customers ...
Matt Cutts: Gadgets, Google, and SEO � Not trapping users’ data = GOOD

... We build a very good targeting engine and a lot of business success has come from that. We run the company around the users–so as long as we are respecting the rights of end users and make sure we don’t do anything against their interest, we are fine,” Schmidt said. He noted that history has shown that the downfall of companies can be doing things for their own self interest. “We would never trap user data,” he said...
This is a radically different stance than almost any company I can think of in any industry. The use of data lock to retain customers is almost universal -- even when it isn't an overt tactic [1]. Google's stated opposition to this strategy is probably more a marker of exceptional confidence rather than true virtue, but I'll give them some "non-evil" points.

Not a lot non-evil points, however. As far as I can tell Google's photo service (Picasa web albums) still locks in data. I don't believe there's an API for transferring images and metadata, or even an XMP-sidecar export option. If Picasas changes course, then I'll give Google full credit. At the moment: B-. Of course everyone else has a big fat F.

[1] The de facto covert approach is to fund development with feature-based ROI. There's no ROI for building migration capabilities, so it never gets funded. Migration capabilities can be expensive. Of course customer demand might change this, but consumers haven't figured that out. Or rather, they hadn't a year ago. I think that may be changing.

How to fight the Cheney/Bush torture machine -- assault the weak points

I'm surprised this hasn't come up sooner. It handn't occurred to me. The American Psychological Association is a legitimate target for non-violent pressure -- starting with shame....
Hamilton Spectator - News

... next step -- we need to metaphorically kneecap the doctors, nurses and psychologists working on the literal kneecappings within Guantanamo Bay. Every one of these interrogations is supervised by people who have sworn a Hippocratic oath to do no harm. They should be immediately stripped of their medical licences. The American Psychological Association (APA) has refused to do this. In an attempt to bargain for more cash and influence from the Pentagon, the APA have even placed a loophole in their ethics code that allows any supposedly legal military order to supercede all the other rules: effectively, the Nuremberg Defence. An organization representing liberal professionals should be easier to shame than the Bush White House; if we can, we can severely hinder the smooth operation of these torture chambers.
I'd have preferred a word other than "kneecap", even when preceded by "metaphorically". APA members should be threatening to resign en masse, if they aren't doing that then they are playing abetting torture.

I'm going to write the American Academy of Family Practice and ask what the AAFP's policy is in participation in government sanctioned torture. If I get a response I'll post it here. If I don't, I'll post that instead.

One caveat. Licensure is typically set by the state, not by a professional organization. The board organizations, however, can take a stance.

Friday, March 09, 2007

Beyond the Gonzales Eight - the prosecutors who did the GOPs bidding

Paul Krugman continues to make me feel good about paying the NYT $50 a year just to read his columns. His is the first article I've read that talks about the darker side of the latest GOP scandal -- the prosecutors who played along ... (emphases mine)
Department of Injustice - New York Times

...The bigger scandal, however, almost surely involves prosecutors still in office. The Gonzales Eight were fired because they wouldn’t go along with the Bush administration’s politicization of justice. But statistical evidence suggests that many other prosecutors decided to protect their jobs or further their careers by doing what the administration wanted them to do: harass Democrats while turning a blind eye to Republican malfeasance.

Donald Shields and John Cragan, two professors of communication, have compiled a database of investigations and/or indictments of candidates and elected officials by U.S. attorneys since the Bush administration came to power. Of the 375 cases they identified, 10 involved independents, 67 involved Republicans, and 298 involved Democrats. The main source of this partisan tilt was a huge disparity in investigations of local politicians, in which Democrats were seven times as likely as Republicans to face Justice Department scrutiny.

How can this have been happening without a national uproar? The authors explain: “We believe that this tremendous disparity is politically motivated and it occurs because the local (non-statewide and non-Congressional) investigations occur under the radar of a diligent national press. Each instance is treated by a local beat reporter as an isolated case that is only of local interest.”

And let’s not forget that Karl Rove’s candidates have a history of benefiting from conveniently timed federal investigations. Last year Molly Ivins reminded her readers of a curious pattern during Mr. Rove’s time in Texas: “In election years, there always seemed to be an F.B.I. investigation of some sitting Democrat either announced or leaked to the press. After the election was over, the allegations often vanished.”

Fortunately, Mr. Rove’s smear-and-fear tactics fell short last November. I say fortunately, because without Democrats in control of Congress, able to hold hearings and issue subpoenas, the prosecutor purge would probably have become yet another suppressed Bush-era scandal — a huge abuse of power that somehow never became front-page news.

Before the midterm election, I wrote that what the election was really about could be summed up in two words: subpoena power. Well, the Democrats now have that power, and the hearings on the prosecutor purge look like the shape of things to come.

In the months ahead, we’ll hear a lot about what’s really been going on these past six years. And I predict that we’ll learn about abuses of power that would have made Richard Nixon green with envy.

We have balanced on the knife edge of history ever since the Dems barely took control of the senate. I think Krugman is correct that Cheney/Bush will eventually make Nixon look squeamish. It will take ten years to reform the GOP.

Software evolution and the DST mess

[Update: I'm wrong about the Java update. Sun went with a utility that updates the JREs, rather than a JRE replacement. In retrospect that makes far more sense, they had to make an exception this time and update the individual class separately. I think the 'tightly coupled' story still sort of works, it explains why Sun had to do things differently this time.]

Recently I bloviated about loosely coupled life and the evolution of software. It occurs to me that I might as well say something about the Daylight Savings Time mess that might reinforce that message. Or not!

The DST transition, from my perspective, is worrisome. I personally think Congress committed legislative malpractice when they mandated this change with a relatively short notice. They should have given us 10 years. I suspect I'm not the only person who feels this way, but we geeks are reluctant to say much since humanity mostly survived the Y2K transition [1].

Why is it worrisome? Well, there are a lot of reasons, and I think most corporate IT departments have already run into them. One of the more interesting examples, however, is Sun's Java runtime environment (JRE). (BTW, the same thing would be true of Microsoft's .NET, but it's less widely used, there are fewer instances in widespread use, and Microsoft learned from some of Sun's mistakes.)

The Sun JRE was an attempt to decouple the software environment from the underlying hardware and operating system. That's a seemingly admirable goal, but the implementation had a perverse consequence of different sorts of very tight coupling. Sun has updated the JRE as one large, somewhat integrated, chunk of code. It's a large and powerful collection of software services, and it all gets distributed together. In theory Sun could have updated one class at a time, but I'm guessing they found the JRE would be even more unreliable if the classes weren't synchronized. (Another example of the disappointments of classic object oriented software.). This means that all the services in the Sun JRE are tightly coupled.

The next part of problem was that, by my recollection, the Sun JRE was a huge and expensive disappointment in its heyday. Each tiny release seemed to break something different. Software that ran well on one release broke on the next one. Inevitably software distribution became dependent on a fairly specific version of the JRE [2]; the software was tightly coupled to the JRE version.

So now we have the situation of tight coupling between the classes in a specific instance of a JRE and tight coupling between a specific instance of a software product and a specific instance of the JRE. Bad enough, but along comes a significant mandated update [3] like the Daylight Savings Time update. Sun is not going to update very version of their JRE, only the most recent versions. So to get the DST update, software may have to jump several JREs to the current version. Ahh, but remember that "tight coupling" between the software and specific instance of the JRE? That's where things break. The software has to be fully tested with the new version of the JRE and it may need to be patched. Then the patched software has to be distributed ...

Well, maybe we'll all do just fine after all. Sun's JREs seem to be much better than they once were, so maybe all the software will make the transition better than they would have three years ago. We'll see!

[1] I personally think we mostly got through Y2K in good shape because thousands of people worked liked demons to fix things. In other words, on the spectrum from over-reaction to under-reaction, I think we were just a bit to the "over" side of "just right". That's pretty damn good for an ape, and it's a great example of how we do better than it seems we ought to do. Unfortunately we're still apes. Prevention is never rewarded or admired as much as recovery, so the popular perception is that the Y2K scare was grossly exaggerated. I suspect this has led a lot of people to understate the DST risks, but I know folks are still working hard to avoid problems. So we'll probably get through, though maybe a bit on the "under" side of "just right". It does help that Y2K eliminated a lot of software that would have been hard to patch for DST.

[2] Lots of bad things happen when you have multiple JREs on a single machine, but that's another story.

[3] Getting the rules around DST right is non-trivial.

Thursday, March 08, 2007

HD Photo - can anyone trust Microsoft?

Microsoft may submit "HD Photo" to a standards body...
HD Photo: Microsoft's next standard?

Last November, Microsoft renamed its JPEG competitor from Windows Media Photo to HD Photo... Ars has learned that the company plans to announce that it will begin seeking standardization for the HD Photo format, essentially bringing it one step closer to becoming the next JPEG.

HD Photo's feature set includes fixed or floating point high dynamic range, wide gamut image encoding; more efficient compression compared to JPEG; lossless or high-quality lossy compression; the ability to store 16 or 32 bits of data per color; and a design intended for use by digital cameras. HD Photo also supports CMYK, RGB, and monochrome as well as embedded ICC color profiles.

I've waited years for JPEG2000 to come to widespread use, but fears of patent vulnerabilities on the underlying math have kept it mostly in the labs. (Adobe Acrobat can use it for image compression, but it's an almost-secret option.)

Microsoft is, needless to say, tough enough and rich enough to fight the patent battles. So this is all about how they handle the patents, and the risks of future patents. Will they, for example, contractually commit themselves not to strike a side-deal with a future patent claimant that would leave competitors out to roast?

If Adobe and/or Apple support HD Photo we'll have a real alternative to JPEG. It's way past time to retire that old warhorse. I'd love to have something like HD Photo as a storage format for my images.

Health care financing: the 80/20 questions are the only questions

I still remember the burnout from studying health care (healthcare? health-care?) financing in the first year of the blessed rule of Clinton. The papers were full of dense analytic essays, until even the specialists were exhausted.

I fear to return to those days, but it's hard to dodge the topic. Like many I've been expecting the train to wreck for years, but it still chugs on, albeit with more and more bits falling off the side. Gradually the topic is returning, so I'll write a bit about it. On occasion.

This post was triggered by a long essay I only scanned: Why does Health Care in the USA cost so much? Over-utilization is an important factor by Walter Bradley.

Well, yes. Except one person's over-utilization is another's "excellence". So here are two examples of the "80/20" question -- the only question that matters. I will tell you how the question(s) will be answered -- but not when.
Suppose I offered you a choice between a test that cost $1,000 but caught 90% of colon cancer, or a test that cost $100 but caught only 70% of colon cancer. Which would you choose - the first or the second?

or

I have two treatments for your heart disease. One costs $200,000 and will completely restore your cardiac output. The other costs $20,000 and will restore 60% of your cardiac output. Which would you choose - the first or the second?
These are the only questions that matter. I will tell you how they will be resolved:
  1. Everyone residing in American will have the second option. Always. I call this the "HMO from Heck Solution".

  2. The first (class) option will be available in a number of ways. Some will get it via risk sharing plans. Some will pay cash. Some will buy it on the gray market ... or the black market. I call this the "Libertarian Solution".

  3. Five years after this choice is available, after development costs have been recovered and competition has arisen, the specifics of the choices will change. The "first class" choice will now become the guaranteed "second class" choice and it will be "cheap". There will, however, be a new, better, very expensive, "first class" choice.

  4. There will be a huge amount of spending on luxury experiences associated with health care and on "alternative" therapies -- none of which will have any impact on outcome.

  5. Innovation, invention, chaos and harm will be far greater in the Libertarian world than in the HMO from Heck world. It is the Libertarian world that will crush costs and convert the "First Class" option into the cheap and universal "Second Class" option.

  6. NIH research funding will shift to favor development of solutions that provide 80% of value for 20% of the cost -- rather than the current disposition to the "best possible" solution.

  7. Once people wrap their heads around this, and that will take a while, they'll decide having the "HMO from Heck" isn't the worst thing in the world.
Of course I can't tell you when this will happen. I'm sure the road will be exciting, highly profitable for some, and very painful for many.

Wednesday, March 07, 2007

Google's DRM: two licenses for Google Earth Plus

Google Earth Plus, I've discovered, is licensed for only two machines per user account.

I found this out when it stopped working -- on every machine! I'd been using it on my two machines at work and my 3 machines at home. Evidently Google doesn't expect people to have more than one machine, even though I was the only user on each machine.

Google, surprisingly, responded to my email after a delay of merely a few days. They supposedly reset my account, but it didn't work.

I've asked them for my money back -- I'll stick to the free version for now. It's their right to enforce their license of course, but I'd grown used to getting away with the old Borland license (anything goes as long as the registered user is using it). I'll have to start paying attention to license restrictions now, and putting products like Google Earth Plus on my black list.

It's only a matter of time before some helpful person puts together a list of products like GE Plus that we all need to avoid ...

Update 3/15/07: I've canceled my subscription. To Google's credit they will refund a prorated portion of my subscription fee. On the other hand the form requires me to give a reason for canceling -- and none of the allowed options apply.

Update 3/16/07: I deleted Google Earth, downloaded the latest version, and still got the notice that I was suspended for using Google Earth Plus on too many machines. The trick, which Google might have mentioned, is simply to delete GE's plist (OS X). I searched for 'Google Earth' with Spotlight, found the file and deleted it. Now I'm a regular GE user again. seems a bit odd to me. Th

Homeland security: radiation detector overload

The New Yorker reviews the state of the art in American nuclear weapon defense ...
The New Yorker: Can the United States be made safe from nuclear terrorism?

...The federal government has distributed more than fifteen hundred radiation detectors to overseas ports and border crossings, as well as to America’s northern and southern borders, domestic seaports, Coast Guard ships, airports, railways, mail facilities, and even some highway truck stops. More detectors are being distributed each month... In the United States alone, the sensors generate more than a thousand radiation alarms on an average day, all of which must be investigated.

The world, it turns out, is awash in uncontrolled radioactive materials. Most are harmless, but a few are dangerous, and many detectors are still too crude to distinguish among different types of radiation; they ring just as loudly if they locate nuclear-bomb material or contaminated steel or, for that matter, bananas, which emit radiation from the isotope potassium-40. So far, the result has been a cacophony of false alarms, which, in most cases, are caused by naturally occurring radiation that has found its way from soil or rock into manufactured products such as ceramic tiles. In addition, people who have recently received medical treatments with radioactive isotopes such as thorium can set off the detectors. At baseball’s All-Star Game in Detroit in 2005, unobserved NEST scientists screened tens of thousands of fans entering the stadium, and their sensors rang just once—reacting to the former Secretary of Energy Spencer Abraham, who was radioactive from a recent doctor’s visit...
This is more than the boy who called wolf. This is the boy's choir calling wolf -- continuously.

It's a fascinating article. I suspect there's no real defense against a technologically competent enemy, but fortunately al Qaeda appears to be quite hostile to geeks and intellectuals. It reminds me a bit of Germany's attitude towards its Jewish scientists, who subsequently joined the Manhattan project. There may be defenses against duller enemies.

Of all the defenses we might have, I suspect the best is at the retail end of the nuclear weapons and waste industry. There may be "legitimate" vendors out there (fewer now that Pakistan is keeping a lower profile), but they are probably outnumbered 100:1 by undercover security agents.

Women and combat: the real problem?

The real issue with "women in the military" is alleged to be their male comrades.

We need independent high quality research, but it would not shock me if the data showed that military women in combat settings are exposed to an intolerably high risk of sexual assault from their male peers and superiors. The next question is then -- can this be changed? I would also not be surprised if we were to discover that this male behavior is extremely difficult to change, particularly when, as in Iraq, things are going badly.

Assuming research showed the risk was high, and that the probability of changing male behavior was low, then there are only two options -- strict segregation of combat forces by gender or removal of one gender from combat settings.

Tuesday, March 06, 2007

Outsourcing the military: Another Bush/Cheney mind-boggler

Tim Spicer, a once notorious mercenary, commanded the second largest armed force in Iraq.
WIRED Blogs: Danger Room (quoting Vanity Fair)

But then, somehow, two months later, Spicer's company, known as Aegis Defence Services, landed a $293 million Pentagon contract to coordinate security for reconstruction projects, as well as support for other private military companies, in Iraq. This effectively put him in command of the second-largest foreign armed force in the country—behind America's but ahead of Britain's. These men aren't officially part of the Coalition of the Willing, because they're all paid contractors—the Coalition of the Billing, you might call it—but they're a crucial part of the coalition's forces nonetheless...
and in Vanity Fair
...As I walked back to Victoria Station, I couldn't help wondering how Spicer had ascended so quickly from notorious mercenary to corporate titan. What had he done to wangle that fat Iraq contract from the Pentagon? Serving 20 years with the British military in the toughest parts of the world was certainly one qualification. So was being smart, connected, and personable. But how had he overcome the taint of Sierra Leone and Papua New Guinea, two scandals indelibly attached to his name? Apparently the Pentagon had decided that an Africa hand could do in Iraq what the American military couldn't: subdue the most xenophobic and violent people in the Middle East...

...

Finding the right personnel can pose a problem. Hart Security, a private military company with roots in South Africa, recruited many of its contractors from the ranks of the apartheid-era South African army, among the most ruthless counter-insurgency forces ever known. One of Hart's men was Gray Branfield, a former covert South African operative who spent years assassinating leaders of the African National Congress. ..

The private military company Erinys also had a South Africa problem. In 2004 an Erinys subcontractor, François Strydom, was killed by Iraqi insurgents. It turned out that Strydom was a former member of the notorious Koevoet, an arm of apartheid South Africa's counter-insurgency campaign in what is now Namibia. There have been press reports of a link between Erinys Iraq and Ahmad Chalabi (the onetime head of the Iraqi National Congress, which was a conduit for the fabricated intelligence used to justify the Iraq war), which both Erinys Iraq and Chalabi deny. After securing an $80 million contract to guard Iraq's oil infrastructure in 2003, Erinys did hire many of the soldiers from Chalabi's U.S.-trained Free Iraqi Forces as guards. Chalabi himself eventually became acting oil minister. ..

...In an odd but lethal twist, it came out last November that the rogue K.G.B. agent Alexander V. Litvinenko had visited the London office of Erinys shortly before his death, by means of radiation poisoning, leaving behind traces of polonium 210...

It constantly amazes me what gets left out of the New York Times. This is the first I remember hearing of Spicer and his army. Bush/Cheney just love outsourcing (they outsourced Walter Reed apparently, with typical results), and they seem to have a fetish for men like Spicer.

Bush is such a multidimensional disaster. I'm sure he's a deep KGB plant.

Update 3/14/07: DynCorp is the US military in Somalia
The State Department has hired a major military contractor to help equip and provide logistical support to international peacekeepers in Somalia, giving the United States a significant role in the critical mission without assigning combat forces.

DynCorp International, which also has U.S. contracts in Iraq, Bosnia, Afghanistan and Iraq, will be paid $10 million to help the first peacekeeping mission in Somalia in more than 10 years.
This is so much like the Heinlein/Dickson(Dorsai) science fiction I read in high school that my Deja Vu has Deja Vu. Typing was one of my most important high school classes, and my trash reading was my best preparation for the future. Who knew?

Loosely coupled life and the evolution of software

Natural selection has selected for designs that facilitate adaptation - and natural selection ...
A Toast to Evolvability and Its Promise of Surprise - New York Times

.... In their recently published book, “The Plausibility of Life,” Dr. Kirschner and Dr. John C. Gerhart of the University of California, Berkeley, offer a fresh look at the origins of novelty. They argue that many of the basic components and systems of the body possess the quality of what they call “evolvability” — that is, the components can be altered without wreaking havoc on the parts and systems that connect to them, and can even produce a reasonably functional organ or body part in their modified configuration. For example, if a genetic mutation ends up lengthening a limb bone, said Dr. Kirschner, the other parts that attach to and interact with that bone needn’t also be genetically altered in order to yield a perfectly serviceable limb. The nerves, muscles, blood vessels, ligaments and skin are all inherently plastic and adaptable enough to stretch and accommodate the longer bone during embryogenesis and thus, as a team, develop into a notably, even globally, transformed limb with just a single mutation at its base. And if, with that lengthened leg, the lucky recipient gets a jump on its competitors, well, g’day to you, baby kangaroo.

Dr. Kirschner also observes that cells and bodies are extremely modular, and parts can be moved around with ease. A relatively simple molecular switch that in one setting allows a cell to respond to sugar can, in a different context, help guide the maturation of a nerve cell. In each case, the activation of the switch initiates a tumbling cascade of complex events with a very distinctive outcome, yet the switch itself is just your basic on-off protein device. By all appearances, evolution has flipped and shuffled and retrofitted and duct-taped together a comparatively small set of starter parts to build a dazzling variety of botanic and bestial bodies.

Living organisms are "loosely coupled" from the macro to micro levels, enabling adaptation at many levels. Arguably the invention of sentience is the taking this "loose coupling" to another qualitative level ...

There are lessons, obviously, for software. Object-oriented programming was supposed to facilitate this kind of modular extensibility, but it did not completely succeed. So-called "web 2.0" mashups [1] and even service-oriented architectures are another stab at building change-tolerant software. I think we'll make progress on this, but it will take some time to learn all the lessons of evolved systems.

[1] I want credit for my age-old technique of turning usenet into a blog by way of tagging.

Monday, March 05, 2007

Understanding software: object oriented programming begins to fade

There are some noteworthy items I read that are a bit too geeky for Gordon's Notes, but are to much opinion or insight to fit into Gordon's Tech. This is one of them, and I'll explain why I put it here ... (note, the original post has several excellent links worth following)
Coding Horror: Your Code: OOP or POO?

... I'm not a fan of object orientation for the sake of object orientation. Often the proper OO way of doing things ends up being a productivity tax. Sure, objects are the backbone of any modern programming language, but sometimes I can't help feeling that slavish adherence to objects is making my life a lot more difficult. I've always found inheritance hierarchies to be brittle and unstable, and then there's the massive object-relational divide to contend with. OO seems to bring at least as many problems to the table as it solves...
I'm not a developer, but I hang around a bunch of old pros. It's not that OO programming has vanished, it's more that it's become a part of the toolkit rather than the entire solution. Service oriented architectures and model-based programming are somewhat in vogue, but really it feels like there's no current "silver bullet".

Inheritance, in particular, seems to have fallen on hard times. It didn't seem to deliver the reusability and extensability most of us had expected ...

Update: As an added bonus, one of the links was to an article on the problems with object-relational data storage. I ignored the distracting comparisons to Vietnam and read through the text. I can vouch for most of the problems the author discusses, I've seen them in action! The political issues are at least as intractable as the political issues.

Sunday, March 04, 2007

Restoring American civilization: 12 steps

The NYT provides a work list congress can follow to attempt to rebuild American civilization:
The Must-Do List - New York Times
  1. Restore Habeas Corpus
  2. Stop Illegal Spying
  3. Ban Torture, Really
  4. Close the C.I.A. Prisons
  5. Account for ‘Ghost Prisoners'
  6. Ban Extraordinary Rendition
  7. Tighten the Definition of Combatant
  8. Screen Prisoners Fairly and Effectively
  9. Ban Tainted Evidence
  10. Ban Secret Evidence
  11. Better Define ‘Classified’ Evidence
  12. Respect the Right to Counsel
...Beyond all these huge tasks, Congress should halt the federal government’s race to classify documents to avoid public scrutiny — 15.6 million in 2005, nearly double the 2001 number. It should also reverse the grievous harm this administration has done to the Freedom of Information Act by encouraging agencies to reject requests for documents whenever possible. Congress should curtail F.B.I. spying on nonviolent antiwar groups and revisit parts of the Patriot Act that allow this practice.

The United States should apologize to a Canadian citizen and a German citizen, both innocent, who were kidnapped and tortured by American agents.

Oh yes, and it is time to close the Guantánamo camp. It is a despicable symbol of the abuses committed by this administration (with Congress’s complicity) in the name of fighting terrorism.
I agree with everything on this list. That said, nothing on the list helps elect a sane president or helps keep the GOP out of power.

It's the old, cursed, choice. Do we choose immoral actions (the status quo) that may avert a greater evil, or do we choose moral actions that might return an unreformed GOP to power? The choice would be easier if I held a higher opinion of the American voter, but the reelection of George Bush Jr cured me of those illusions.