triplecorrelation.com

Remember when the term “ripping” became popular? People used to “rip” songs from their CDs using itunes so that they could load the ones they wanted on their ipods.  They could then carry around just the music they wanted.

Arguably, Apple’s best idea with iTunes is not building a store, but allowing people to buy by the song. Each song cost $0.99, which is just the right price–small enough that you won’t regret an impulse buy, but large enough to make you want to listen to a preview before you bought.

Imagine if  iTunes only let you buy by the album — around $9.99/album.  You would be a lot less tempted.  Why not just buy the CD at the store? That way you could have a permanent copy, with cover art, possibly a lyric sheet, and also be able to play it in the car.  No, buying by the song made a huge difference and in my opinion accounted for iTunes’s success (and Apple’s). And it allowed electronic music to be sold for the same amount as a physical version, despite the obviously huge profit margin to Apple from not having a physical store selling physical media.

Now comes the iPad.  It’s purpose is to sell books and magazines. I don’t see how the book sales will work.  You can’t sell books by the chapter. You have to sell the whole book, and if that is sold for the same amount as a physical copy, to follow the music model–around $10 for a general book and $100 for a textbook, who would buy it?  Books already cost too much.

I’m waiting for college students to start ripping their very-expensive textbooks. They can use any of the designs shown here

http://www.diybookscanner.org/

The cost of a diy scanner is less than the cost of an iPad. It’s essentially 2 cameras, and a light source, connect to a pc with some software. If you have the patience to scan a text book, about 2-4 hours, then you can carry around the pdf coveniently on a laptop or in a book reader like the iPad.

I can’t wait for someone to build a page-turning robot arm addition.

I realized the other day that I’ve been around long enough to see at least kind of cycle in technology: the build vs buy cycle.

It used to be the fashion to buy components rather than design your own.  Computer companies like HP have long done this: intel-samsung-nvidia design and (intel - samsung) fabricate the chips, taiwanese and chinese companies assemble them into motherboards and printers and pcs. And MSFT makes the software.  HP spun off Agilent which spun off Avago, which actually designs chips.  Nobody wanted to design chips.

Apple took the build over the buy : they designed their own OS and have excellent product design.  They also used to rely on the chip companies for their components, but they bought P.A. Semi to design their own processor, used in the iPad, and rumour has it, the next iPhone.

Video compression used to be another component that people bought rather than built.  You could buy the rights to use MPEG-4 or H.264 through MPEG-LA, and deploy that technology on your encoder.  Google is taking another tack: it bought On2 , and plans to open up the source of that companies’ video compression technology to enable the “WebM” format.  Regardless of how good WebM is compared to H.264, if it is even 70% as good, I doubt anyone will notice.  One simply has to shrink the resolution to achieve the same number of bits.  3G bandwidth is more than enough for streaming video and flash storage is cheap.

I’ll wait a few years for the management fashion to swing back to buy rather than build.

If you use Google’s Chrome on a wifi network, you might notice it occasionally hangs with the message “resolving host”. It turns out that Chrome is trying to translate the website you are looking for into an IP address, which is also known as “DNS resolution” (DNS = domain name system). You can get rid of this problem and speed up surfing by turning off DNS prefetching under Options->Under the Hood.

Chrome is apparently the only browser that does DNS prefetching. The reason, according to Chromium blog, is that it speeds up DNS resolution (by a few hundred ms) by using the time otherwise wasted while you are typing in a URL to fetch the IP address.  I imagine that it works by having Chrome run a separate thread to fetch IP addresses of whatever is related to your recent visit, so that the time spent typing (which is a relatively slow activity) is concurrently spent on fetching IP addresses.

For some reason, this multi-threaded activity fails — possibly there is a livelock.  The browser hangs, with the unhelpful message “resolving host…”.

I found that turning off prefetching got rid of the hanging browser and made life much better.

I remember having once opened up the cover on an Apple IIe.  I was trying to help a friend figure out why his keyboard was not responding. It turned out to be dirt caught beneath the key.    While it was open, we could check out the pcb layout and see what chips were used.

That was then.  Today, Apple makes products which do not open or extend in any easy way. The iPhone is a great example.  It doesn’t have a flash memory card that you can save data on (like a pdf file).   As everyone knows, you can’t change the battery.   There are no screws to let you take it apart easily.  Even inserting a sim card is done through a cumbersome slot that requires you to fold open a paper clip and stick one end through a tiny hole.

Then there is Flash.  Apple does not support Flash because, in theory,  Flash lets you create an App on a web page.  That means people can create apps without needing to sell the App on the iTunes store (and paying Apple 30% of the revenue).

If you take a photo with the iPhone 3G, the photo is stored in JPEG format which doesn’t contain EXIF information (exposure time, f#).   It is the only phone or camera that I know of which is so secretive.

It’s well known that the iPad doesn’t support Flash either.  There are predictions that Flash will die out because of the advent of HTML5. If it turns out to be possible to write web-hosted Apps using HTML5 ,don’t look to Apple to support it.  Control is everything to the most secretive company.

I bought an iPhone in the US without a contract, which requires paying full price. It is unfortunately still locked to AT&T, which means I cannot use it in Singapore. However, if I unlock it, I can use it on any network. Unlocking for personal use in the USA is not illegal, according to the exemption given to the DCMA. Since I live in Singapore, it is not clear what law applies. Anyway, if I am doing for personal use and have already paid Apple the full price of the phone, I can do what I want.

I asked a friend who has experience with this.  The process involves installing Cydia.

Anyway, thanks to my friend, my iPhone now works beautifully in Singapore. I am able to use any SIM card I like. I must say the audio quality is better than any phone I have used previously. However, I cannot update my firmware without going through the unlocking process again. I can still install apps–in fact I just installed Skype.

One annoying thing about Apple’s phone keyboard is that all phone numbers are displayed in (3)-3-4 US format.  Here in Singapore, the number are in 4-4 format, and Apple’s reformatting throws me off.  Strangely, a simple option to display numbers in non-US format is not available in any of the menus I have seen.

The White House Flickr stream keeps getting more interesting all the time. Here is a link to one of the best pictures Situation Room photo It’s a photo taken in the Situation room of the White House, where the President meets to plan strategy. The high resolution photograph is great. You can see, for example, the title of the document next to Robert Gates, the Defense Secretary. It says “Afghanistan Strategy : Brief to the POTUS”. POTUS is the acronym for President of the United States.  Here it is.

Paper in front of Defense Secretary Robert Gates

I attended a great talk by Professor Eugene Fiume on Digital Media today.  He showed examples of technically brilliant computer graphics, which as he put it, still lacks “irony”. He also showed examples of great art where there is a lot of irony despite the lack of detail or technical brilliance (a clip from the Disney movie “Dumbo” was really moving).

The talk made me think about Flickr, the photosharing site.  There’s a lot of technically brilliant photos on Flickr, which are excellent in exposure, focus, etc, but they are not that interesting.    On the other hand, YouTube has a lot of poorly lit, grainy, shaky, technically crappy videos which are very compelling.  The YouTube videos have “irony” in the Fiume sense.

Why is that? Why do YouTube contributors “get it”, while Flickr contributors don’t? Is it video vs photographs, or is it YouTube vs Flickr (the way they are organized) that matters?

I started reading this after being taken in at the bookstore by the first page, which, as in other Gibson books, is memorable.   Here is one sentence on that page: “Nothing at all in the German fridge, so new that it’s interior smells only of cold and long-chain monomers.”

The story is about the search for the maker of “the footage”, a collection of snippets from a movie which is being released anonymously on the internet.  The snippets are released in random order.  Fans of the footage reassemble the snippets to form their own movie, applying their own pattern recognition.  This is the only book of fiction that I know of to discuss watermarking of video.

The rest of the book doesn’t live up to the first page.  In fact, it drags and is often tedious.  It would have been good to cut about 50-100 pages.    Gibson’s style is not particularly easy to follow.  It’s a lot like reading emails from a busy, self-absorbed geek.

I’ve read glowing reviews of this book and its meaning on the internet.  The book would be worth those reviews had it been better written and motivated.

My Samsung Innov8 phone has a feature called “DLNA”. This is supposed to let you wirelesslessly connect your TV, cellphone, and pretty much everything else that may have wifi, together. Presumably, you can show pictures or video from your phone on your wifi-equipped TV. In theory. Of course, these grand connectivity schemes never work (bluetooth?).

My DLNA client is located in the cellphone’s “Applications” folder. When activated, it says version 1.0. That should tell you what you to expect. However, the DLNA logo is prominently displayed on the back of the phone, and on the box it came in. So I decided to try it. When launched, the client asks you for your wifi network name. After 3 tries, it managed to log in. The screen showed a status message that said “server active”. Nothing else happened. None of the other menu options suggested a way to use the client. Nor was there a brochure that came with the phone to explain what to do next.  I wonder how much longer DLNA will survive as a labelled feature on cellphones: 1 more year, 2 years (my bet), or 5 years.