Saturday, October 30, 2010

Lessons from the Cloud: Nobody loves your data like you do

There's a lesson here worth sharing  in this personal story (revised from original post to correct and clarify) ...
I now have about 5000 posts without paragraphs - Blogger Help
... I have about 8,000 posts in 2-3 "blogger" managed blogs going back to 2001.
As of today thousands of them no longer have paragraphs. Google did something that changed the way their software recognizes paragraphs. Perhaps it was a side-effect of using one of Google's newer templates. (For example.)
These posts are rather difficult to read. There are far too many for me to repair.
This is the most impressive episode of content destruction I have ever encountered. I've lost many hard drives over the years, but I was always able to restore from backup.
I can't fix this...
Coincidentally, Apple did something similar to their loyal customers this week. An update to iPhoto destroyed many customers entire photo libraries.

Although Apple's sin is even greater than Google's, there's an important difference that we all need to remember.

I can manage Apple. I don't use their software until months after a release; Apple uses early adopters as beta testers. I also use a very robust (but still imperfect) backup strategy. If Apple trashes my data I can restore it.

I can't manage Google. I don't have control over my Cloud data.

Sure, there's probably somebody at the increasingly troubled land of Google who feels badly about trashing my posts. I can guarantee though, that they don't feel as badly about as I do.

That's the problem with Cloud data, and it was a problem with the "ASP" market too (application service provider, which is what we used to call "Cloud services" before that market went pear shaped).

Nobody cares about my data as much as I do.

The problem with the Cloud isn't a technical problem, it's a social problem. If Google had to pay a billion dollars every time they mangled data, they'd really care. As it is now, they only sort of care.

My blogs are the my only serious investment in the "Cloud" data -- and that investment has gone sour.

Don't trust your data to people who don't care for it the way you do.

Rally for Sanity - and The Enemy Within

The Rally to Restore Sanity was today. It went well. Emily and I would love to have attended, but most sane people our age are pretty rooted. It's no accident many of the "6 billion" attendees were around retirement age.

The theme of the Rally reminded me of one of Shatner/Kirk's better Star Trek moments -- as the divided Kirk of The Enemy Within in the fifth episode of the first(!) season. The ever unreliable transporter splits the Captain into a sane Kirk, and a sociopath Kirk.

The sane Kirk is clearly the better half, but he's not an effective leader. Like any modern CEO, he needs his inner sociopathy. When the two Kirks are reunited the Captain returns.

America needs all of its components, including some version of the GOP.

Friday, October 29, 2010

My life with Amazon

Amazon's Improve Your Recommendations list is a chronological record of purchases. The oldest item on my list of about 560 items is a book on Java and Corba (!) published in 1997 [1].

Amazon went only in 1995 and I believe I bought some books from them in 1995 or early 1996 -- so the list isn't complete. It's close though.

When I started the list we had 1 dog and no children; I worked in the early days of a startup company (Abaton.com - the domain has lapsed). Today we have 3 children, 1 (different) dog, and a fish. I work in a large publicly traded company.

Browsing the list is a strange window into the past 13 years. There were some surprises -- I bought too many (non-fiction) books I never read. I'm better about that now.

What's your list like?

[1] Oddly, there are three prior items listed that I'm reasonably sure we didn't order. You can skip the end of your list by editing the URL.

Tuesday, October 26, 2010

Can Apple be sued for false advertising?

On the Aperture site Apple claims that the move from iPhoto to Aperture is "designed to be seamless".

That's certainly untrue. Aperture has nowhere to store Album or Event comments for example.

So can Apple be sued for false advertising? Any desperate lawyers out there? (Note since I've not been injured I can't be a litigant, but I'm sure one could be found.)

Lessons in Software: Aperture 3 and iPhoto 11

The state of OS X photo management is mixed.

It’s not so bad for customers who aren’t invested in iPhoto. Much of Adobe’s OS X software is lousy, but Adobe Lightroom is a singular exception. It still, for example, uses Apple’s installer instead of Adobe’s malware installer. It’s a professional level product with a learning curve, and it shares Adobe’s uncertain future, but it’s the best choice for a new user.

Apple’s professional product, Aperture, used to compete with Lightroom, but it lost that battle. Aperture has been unreliable, slow, and buggy – which professionals and prosumers can’t tolerate. Aperture’s only future is to be an upgrade path for iPhoto. That’s where things start to get grim. Apple claims there’s a smooth migration path from iPhoto to Aperture, but Apple lies (yes, they do). Aperture has nowhere to store much of iPhoto’s event and album annotations

So what about iPhoto? Ahh, that’s where the really bad news starts. The latest release, iPhoto 11, has a “bug” – it can delete images. Tens of thousands of images. This would be bad even if everyone had backups – but backup is an unsolved technical and social problem. It’s likely that iPhoto 11 also deletes images in a less obvious fashion, which, if you think about it, is worse than deleting all images all at once.

Image deletion is a bit of a nasty bug, but it doesn’t affect me personally. I know Apple. I never use their software until the beta testers early adopters are done. I am, however, impacted by Apple’s product direction. They are making iPhoto more of a true consumer product, removing functionality and foreclosing features I want (like detached library management).

iPhoto power users are on a sinking ship, and the Aperture life raft comes with mandatory limb removal. We’re left in the sad situation of hoping Adobe will bail us out with a Lightroom migration path, but I suspect they no longer have the resources to build one.

Yech. This sort of thing happens much too often these days. I’m in a similar situation with my Google hosted blogs. So, what can I (we?) learn from this? Here are my take home lessons:

  1. It’s very hard to live between markets. This is true across many domains, from power tools to software. There’s a stable market for costly professional products, and a market for lowest denominator consumers. The in-between “prosumer” market is unstable.
  2. The “consumer” market isn’t a good place to be. Consumers have short memories – they upgrade to Apple’s new products despite a long history of major data destroying bugs. Consumers, by and large, don’t care enough about their data.
  3. Apple doesn’t have a culture of quality because their customers don’t demand quality. They do have a culture of design. If you’re like me, and you love both design and quality, you’re in trouble. There’s no easy answer, but don’t forget the tradeoff.
  4. There’s no fundamental reason Aperture couldn’t be changed to support more of iPhoto’s metadata…

The last one, to me, is the most interesting question.

Why doesn’t Aperture support more iPhoto metadata? It’s in Apple’s business interest to migrate iPhoto users to Aperture, why not make that work properly?

I don’t know, but I think this is one aspect of a general problem with software. There are a million good paths to take in software development, but you can really only take one. If you take ‘em all, you get a symphony composed by committee. Software development requires a tyrant, but it takes a long time to do good software. It takes about 5-10 years.

Ten years is a career. It’s too long a tenure for modern business structures; it doesn’t match career or business models. Tyrants can’t last in large businesses – unless the tyrant owns the show (Jobs). It costs too much to devote very talented tyrants to maintaining and building something like Aperture when iOS development returns far more value.

I’m hopeful we’ll eventually figure out a solution for problem 4. I’m hopeful that the OS X App Store and its FairPlay DRM will make small company software more viable. If that happens Apple could sell Aperture to a company that could make a good profit migrating iPhoto customers – and Apple would still earn 1/3 of the revenue from App Store sales. That could be a win-win for everyone.

In the meantime, OS X iPhoto users need to stay with iPhoto 9 and wait for a solution.

Monday, October 25, 2010

Google's computational thinking curricula for grades 6 and up

A new way to torture your grade 6-12 child ...
Google: Exploring Computational Thinking
... Use a CT approach to develop an algorithm for calculating percentages using mental math. This example is based on released questions from the California Standards Test, 6th Grade...
I looked over the examples, but they they felt strained. The curriculum is based on teaching Python in 6th grade.

Sunday, October 24, 2010

Why we can't raise the retirement age to 70

We live longer, work is less physical, and so we can raise the retirement age. That's the usual argument.

Then there's this. Gassee is quoting from a May 2007 USA Today story ...
“Fatality rates for drivers begin to climb after age 65, according to a recent study by Carnegie Mellon University in Pittsburgh and the AAA Foundation for Traffic Safety, based on data from 1999-2004. From ages 75 to 84, the rate of about three deaths per 100 million miles driven is equal to the death rate of teenage drivers. For drivers 85 and older, the fatality rate skyrockets to nearly four times higher than that for teens.”
Driving a car is cognitively demanding. Performance drops off for most people at age 65 -- traditional retirement age.

We're not going to be able to extend the knowledge worker's employment life until we can slow the decline of the human brain.

Of course this is before we factor in the effects of epidemic Diabetes.

Lessons from Allais Paradox - making right choices

DeLong refers us to a summary of the Allais paradox and its relationship to the 2002 Economics Nobel.

The bottom line is that humans fear losses more than they love gains. This leads to consistently illogical personal and business decisions. The 'rational economic animal' is an illusion.

Among other things, this explains we dislike recognizing sunk costs (see resolution 242). To recognize a sunk cost is to recognize loss.

How should we use this knowledge about the way our minds work?

First of all we should formally express the potential losses and gains of significant decisions, and recognize that our emotions will lead to consistently inferior economic choices.

Secondly we can also use this knowledge to bias decisions others make. For example:
... When asked whether they would choose surgery in a hypothetical medical emergency, twice as many people opted to go under the knife when the chance of survival was given as 80 percent than when the chance of death was given as 20 percent...
If you want a certain decision made, give the probability of the gain, not the probability of the loss.

Surprisingly, however, we may still want to make the "wrong" choice. We are not, after all, creatures of pure reason. If our goal is happiness, we will be happier if we experience fewer losses at the cost of smaller gains -- because that's the way we are built. The "right" choice depends on whether the goal of the decision is our personal happiness or economic success or maximal lifespan.

Advertising

The kids like FM pop radio.

I don't mind the music, but I have to listen to the ads. That hurts, because half the ads are for goods and services that are largely fraudulent.

In Ayn Rand's world the libertarian elite were (warped) nobles, in the real world they're predators feeding on the weak.

Incidentally, nobody advertises anything Emily or I want - anywhere.  That's a market failure. We do spend money. Why don't I see ads for family nordic ski resorts?

Thursday, October 21, 2010

The rational basis for climate change denialism

I consider this a respectable and rational basis for denying that the earth's climate is being significantly altered by human greenhouse gas emissions ...
Global Warning Skepticism in Tea Party - NYTimes.com
...A rain of boos showered Mr. Hill, including a hearty growl from Norman Dennison, a 50-year-old electrician and founder of the Corydon Tea Party.
“It’s a flat-out lie,” Mr. Dennison said in an interview after the debate, adding that he had based his view on the preaching of Rush Limbaugh and the teaching of Scripture. “I read my Bible,” Mr. Dennison said. “He made this earth for us to utilize.”...
I like this response. It's much less painful than reading right wing pseudo-science.

Mr. Dennison holds a set of religious beliefs. That belief set includes the understanding that God gave Man a planet to use as Man wishes, and He designed the planet so Man could not damage it. Therefore the scientific consensus on climate change is a fraud.

His reasoning is absolutely internally consistent. His conclusions follow directly from his premises. There is no response save to criticize his religious beliefs -- which is a rather sensitive topic.

I wish more Denialists were as honest as Mr. Dennison. I'm not being sarcastic. I think, at the core, this is what most Denialists believe but refuse to say.

Wednesday, October 20, 2010

Good news for Obama -- the GOP will take the House and Senate

Gwynne Dyer makes the case -- it's all sunshine for Barack.

Contrarian? Certainly. Persuasive? Pretty much.

The disruptive telecom play - the 3G iPod touch

Telecom CEOs can be ... tiresome. Yes, they are evil, but it's a dull, brutish, boring sort of evil. Sometimes they forget their place, and ask Jobs to do something he doesn't care to do.

That's when he pulls out his toy. It looks like an iPod Touch, only a bit fatter. Jobs strokes it. The Telecom CEO's shut up.

This iPod Touch, they know, is special. It has a prototype TMobile 3G CDMA chip, a $15/month 10GB data plan, and FaceTime. There's no voice capability, it's a pure data phone. The voice stream is VOIP.

TMobile has nothing to lose. They'll play this game.

This is a disruptive device. It's a computer with a net connection. It works with an external keyboard. One day it might drive an external monitor.

It's coming one day. If not from Apple, then from someone else. (Not from HP though. They're as dead as Dell). When it comes, the Telecom market blows up.

Whether it comes in 4 years or in 1 year is entirely up to those Telecom CEOs. They just have to keep Jobs happy ...

US 2003 and China 2010 – Nations on meth

The NYT is reporting that China is limiting the export of “rare earth” minerals to the US and Europe (and Canada too I’d assume). China did the same thing to Japan a few weeks ago. This is alleged to be a response to threats of tariffs on imported goods made in China.

This is a parenting moment. This is the time when the 16 yo steals the keys, goes for a ride, and puts a nice dent in the front fender. Time to count to ten, and repeat ten times. Then remember what is a stake, and what the long game is.

If you were a reasonably mature and sane nation in 2003, such as Norway, you’d look at 2003 America and draw some conclusions. You’d decide that America was in much worse emotional shape than you’d thought in the Clinton years. You might guess that America had an unsuspected substance problem. The best response would be calm engagement, while also making wise preparations for American derailment.

That’s what we should be thinking about China. China is a critical component of world civilization, and, if this story is accurate, she’s in much worse emotional shape than we thought. We thought she was a smart but rowdy striver, but now we learn she’s snorting meth. This is a problem for China, and for the family of nations. We’re all in this together.

In the meanwhile there are obvious things to do. We have a strategic petroleum reserve. We will tax all rare earth import and production (not just China’s) to build a rare earth strategic reserve. All nations will do the same thing. That will give time for nations like Canada, Australia and throughout Africa to increase production.

More importantly, we need to think differently about China, just as all nations had to think differently about the US after 2003. The world needs a healthy, stable, rational China. We’re going to have to do a lot of counting to 10, even as China works through her own issues, and even as we prepare for the worst.

Maybe the US and China can join a rehab program together.

Monday, October 18, 2010

Tools of the NYT

I liked this NYT reference on favored home tools, especially their product endorsements ...

  • STANLEY FATMAX RATCHETING MULTI-BIT SCREWDRIVER $10, (sigh)
  • DeWalt 1/2-inch, 18-volt Cordless Compact Hammerdrill kit (hammer, not just drill)
  • BOSCH 5-AMP JIGSAW (rather than rotary, sigh)

Saturday, October 16, 2010

Oddly silent dogs: The digital TV conversion

Canada still hasn't converted, but the US went to all digital TV on June 12, 2009.

We were supposed to convert in Feb 2009, but the switchover was delayed to allow more consumers to buy D/A converters or digital TVs. My family stayed with rabbit ears; they work better with our subsidized D/A boxes than with old school analog broadcasts. It is true, however, that nobody here watches TV on a TV any more. (The kids see their cartoons streamed to the laptop during their earned computer time, we also stream Netflix to iMac or Wii or we try to watch Netflix's frequently defective DVDs).

Which brings me to something noteworthy. The dog didn't bark. I expected some screaming by on June 13th -- didn't happen. Nothing happened. Cheapskates like us bought subsidized converters, others bought new TVs, cable subscribers weren't affected anyway. As best as I can tell, everyone who wanted a subsidized converter got one.

Most of the analog frequencies were sold off for vast sums (dwarfing the A/D converter subsidies), but a critical component has been made a public good with potentially enormous implications.

This was a huge change in American life, and it was created and executed by government. The six month switchover delay and extra converter subsidies happened under the Obama administration, but this was a long transition that started in Clinton years and went through Bush II.

A longterm, bipartisan, massive technological transition entirely driven and controlled by the Federal government -- that worked.

Nobody talks about it.

Isn't that a bit weird?