Subtly insane rants by
I make apps for iOS and the web. I live in Portland. Partner, political junkie, geek.
Text
Why IPv6 Isn’t Happening
Feel free to don your tin-foil hat for this one, because it’s a pet conspiracy theory. Still, I think it has some merit.
Ars Technica recently spent some time trying to figure out why IPv6 traffic isn’t increasing. While there is concern over the readiness of the hardware and software stack at both ends of the pipe, the reality is that we’re out of new IPv4 blocks. There are simply no more left to allocate. To this end, there is now evidence that a market is developing for IPv4 addresses. Not particularly surprising, but I keep looking at the situation and wonder why, exactly, we haven’t gotten any meaningful IPv6 penetration. I think the answer is mobile phone service providers.
In today’s world being connected means being reachable online and by voice. It is common enough these days that these two things are one and the same. A person choosing to rely on Skype or some other VoIP solution is choosing to take his or her voice on top of the data connection, but it’s still just data. The key to being reachable by voice is your phone number. In the grand scheme of things, when we pay our never-decreasing monthly fees, we’re largely paying for the privilege of having our devices work outside of our home networks, and the ability to be reached via a simple series of digits. This is the world in which telephones operate.
That phone number is magic in its own way. The technology makes it possible for that one number to find us wherever we happen to be. It is reliable at this task, even if the connection itself may fail over the duration of the call. If that happens, you redial, the system again works its magic.
In a sense, what we know as “IP” - the Internet Protocol - does the same thing. Given an address, it enables information to travel from a source to a destination. In the world of IPv4 this requires a complex system of sharing in order to efficiently allocate a limited resource. In the world of IPv6, this is no longer necessary.
That number is probably too large to conceptualize, but all we really need to know is that it is no longer necessary to share IP addresses. We have enough. More than enough. With IPv6, we have so many we really don’t even have to think about it for a long, long while. Really. (Have I made my point, yet?) While organization and routing will still play a crucial role in networks, it should theoretically be possible to assign each Internet-connected device its own unique IP address. The second this happens your Internet connection - any Internet connection - becomes just another dumb pipe.
There is no need for a phone number when an IPv6 address can identify your specific device. Why do you need a phone number to initiate a voice call, video call, or text-based connection between two devices? SMS data costs more to the average American than data from Hubble. Voice calls are no different. A 450-minute plan that costs $50 per month has a per-minute cost of approximately 11¢. If we’re ridiculously generous and say one minute of cell-quality phone conversation uses one megabyte in transfer, the per-gigabyte cost to the customer is well over $110!
There is little correlation between what mobile services cost to provide and what mobile operators charge for them, and they are able to do this because they provide the magic mechanism that maps a phone number to a voice. Give each connected device its own address and add DNS, and every mobile provider suddenly only gets to compete on bandwidth and price. This is not how Verizon, AT&T, et al wish to compete.
Link
The Truth about Android and iOS market share
TL;DR:
…if you factor in all of the devices that use iOS including iPhones, iPads and iPod touches, Apple reached 37.9 million people. Android, according to the data, reached 23.8 million on phones and tablets.
That’s a 59 percent lead for iOS over Android.
Text
Apple Microsoft
Today’s news of Apple’s market cap surpassing Microsoft’s makes me wonder what Microsoft might have accomplished had it maintained focus on Windows and Office. For all that Microsoft has spent on Xbox, Zune, and literally hundreds of other products — all designed to enter the market solely to ensure Microsoft had a foot in it — years of effectively nil growth is the result.
This latest Apple vs Google battle is fun to watch, but they’re both just getting warmed up.
Text
Regarding Jon Stewart*
The interwebs are abuzz over Jon Stewart calling out Apple as “The Man” for the search warrant that was executed on Brian Chen’s house. I love The Daily Show and normally agree with a lot of what I hear. On this topic, we happen to disagree.
Consider things from Apple’s perspective. Their secrecy is part of the appeal. Everybody wants to know what they’re going to do next, and the company does an excellent job of controlling the flow of information. They have a right to do this. They work within regulations and their shareholders are obviously happy.
So you’ve got this brand new iPhone in the works, and it’s impressive. It’s made of higher quality materials, build quality is going to go way up, the new CPU is much faster and more power efficient, it has a bigger battery and a higher resolution camera in addition to the new camera on the front. If you’re Apple, you want to keep these details secret. Less people want to buy this year’s model if next year’s is a couple months away. This is in contrast to cars, where the current model often becomes more attractive to a large segment of buyers, because it’s available at a discount. Apple doesn’t do discounts like that. The price on today’s model normally doesn’t change in the lead-up to new product. When new models are released, Apple will pay you the courtesy of upgrading your order (if you ask) if you placed it in the preceding 10 days, but if it’s day 11 you’re probably out of luck. ((Unless you get a nice customer rep at AppleCare)) This is also perfectly legitimate. The result is less overstock, which means less discounting after the new model is out.
Leaking the new iPhone probably cost Apple millions in revenue. While it isn’t really a secret that Apple has an annual release cycle for iPhone, the previously-described effect is so strong that Apple is selling more iPhones today than it did during the holiday quarter. Think about that. Now that the forthcoming iPhone is common knowledge, I’m expecting that we’ll see a more pronounced dip in sales over the next month or so. ((On the flip side, it’s probably fair to say that the anticipation for the new iPhone may indeed be higher than if Gizmodo hadn’t taken the actions it did.))
The evidence to date suggests quite clearly that after the device was left at the bar, none of the people who handled it afterward did the right thing. The original finder ((Now identified as 21-year old Brian Hogan of Redwood City, CA)) didn’t make enough effort to return the device directly to Apple. He took money from Gizmodo, and handed it over. Gizmodo knew what they were getting, and they didn’t return it until Apple legal counsel asked for it back.
Leaving aside the lost sales from the current model, that prototype was assuredly one of a relatively small number of units. One-offs like that are normally costly to produce, on top of untold millions in research and development.
The long and short of it is that a crime occurred, and the police are obligated to investigate. Apple could probably intervene, but at this point in the game it isn’t fair to say Apple is going too far. Better to make an example of these fools as a warning shot to anyone else who might consider acting similarly. You can guarantee whatever money Gizmodo made from ad sales premiums when the story broke is but a drop in the bucket relative to what Apple lost by Gizmodo’s irresponsible behavior.
You might think Apple’s behavior is certainly reminiscent of The Man, but Cult of Mac made a rare good point about how things work at 1 Infinite Loop that lead me to think otherwise.
On the topic of AT&T, Mr. Stewart and I are one. Despite my overall disagreement with his arguments, the bit was hilarious.
* This post is a response to a comment by my good friend Josh Bernstein, on Facebook. I decided it was a little long to stick in a comment.
Text
Batch Converting Files With FFmpeg and Ruby
I recently needed to burn some video files (h.264 MKV) to DVD such that the disc would work in a standalone player. After an hour googling for a set of commands to pass to ffmpeg to convert my files, I then started looking for batch scripts. It occurred to me shortly thereafter that Ruby could help. I had found a command that worked beautifully with a single file, so all that was necessary was to automate it.
It only took one line to accomplish the task:
It’d be a cinch to generalize for input video format.
- Older
- Page 1 of 5
