Deconstructing Carrier IQ’s Press Release

By | December 1, 2011

I couldn’t find this press release on their website, and it’s a couple of weeks old, but I thought it worth deconstructing anyway. My comments in quotes. The rest is from the release. I don’t pretend to have got anything right here, but these might be the starting points for deeper questions.

Carrier IQ Says Measuring Mobile User Experience Does Matter! – MarketWatch:

MOUNTAIN VIEW, Calif., Nov 16, 2011 (BUSINESS WIRE) — Carrier IQ would like to clarify some recent press on how our product is used and the information that is gathered from smartphones and mobile devices.

Carrier IQ delivers Mobile Intelligence on the performance of mobile devices and networks to assist operators and device manufacturers in delivering high quality products and services to their customers. We do this by counting and measuring operational information in mobile devices — feature phones, smartphones and tablets.

operational information is a very vague term. And it’s clear from this comment that it’s not just smart phones that have the software installed. Feature phones and tablets also have it.

This information is used by our customers as a mission critical tool to improve the quality of the network, understand device issues and ultimately improve the user experience. Our software is embedded by device manufacturers along with other diagnostic tools and software prior to shipment.

It calls it a diagnostic tool, but most people’s understanding of a diagnostic tool is one that runs in diagnostic mode. This doesn’t. It runs all the time–even on WiFi and airplane mode. But this comment also hints that there are other tools and software installed by manufacturers too.

While we look at many aspects of a device’s performance, we are counting and summarizing performance, not recording keystrokes or providing tracking tools.

‘Recording’ keystrokes could be as it looks, or it could be weasel language, given the fact that keystrokes are definitely logged. Logging could be considered different to recording in this context.

The metrics and tools we derive are not designed to deliver such information, nor do we have any intention of developing such tools.

But they clearly do, so is that a bug? Is the word deliver here key, as in not designed to deliver such information to certain parties?

The information gathered by Carrier IQ is done so for the exclusive use of that customer, and Carrier IQ does not sell personal subscriber information to 3rd parties.

This doesn’t really help. Not only was it not really the issue that Carrier IQ was selling the data–it was assumed the carrier would be, if anyone was–and the term personal subscriber information is quite possibly a weasel term, as personal has tended to mean to include the actual subscriber’s name. But we know now that even anonymized data can be mined so it is quickly connected to a specific person.

The information derived from devices is encrypted and secured within our customer’s network or in our audited and customer-approved facilities.

I don’t know enough about this, but I’m guessing these are weasel words too. The key word is within. It seems pretty clear that most if not all of the Carrier IQ data is in plain text, so presumably the encryption and securing is only when that data reaches the customer’s network (i.e. this doesn’t include the external network, but the customer’s own computer network.) It also makes clear that the data, whether encrypted or not, also resides within Carrier IQ’s systems.

Our customers have stringent policies and obligations on data collection and retention. Each customer is different and our technology is customized to their exacting needs and legal requirements.

Except that at  no point was any customer, as far as we know, actually asked whether they approved this data being collected about them. In fact, we don’t even know who those customers are in order to be able to verify this.

Carrier IQ enables a measurable impact on improving the quality and experience of our customer’s mobile networks and devices. Our business model and technology aligns exclusively with this goal.

Don’t get me started on the word ‘experience.’ It covers a multitude of sins and can mean more or less anything. My experience of call dropouts? Yes, sure, fix that. My experience of what services I use, how many times I enter my password, whether I’m buying something in Starbucks or Coffee Bean, how many people are in my address book etc. No. Not what I want you to log.

I think there’s another element at play here. Clearly the device manufacturers have allowed this to happen since the software is installed at the point of manufacturer. A carrier can use the service because whatever device their customer uses, they can be pretty confident that the Carrier IQ software is embedded. So one has to ask what data are being shared between carrier, Carrier IQ and manufacturer? And how does this work?

SOURCE: Carrier IQ

Quaintness in Salt Lake

By | November 27, 2011

(This is the script for a piece I did for the BBC World Service. Posted here by request. Podcast here.)

Something rather quaint is going on in a Salt Lake City courtroom. A company called Novell, who you’d be forgiven for not having heard of, is suing Microsoft over a product called WordPerfect, which you also may not have heard of, which it says was hobbled from running on something called Windows 95 to protect its own product, called Microsoft Word.

To be honest, you don’t need to know the ins and outs of this Microsoft law suit; nor do you really need to know much about Novell—once a giant in word processing software, and now a subsidiary of a company called The Attachmate Group, which I had never even heard of. Or, for that matter Windows 95—except that once upon a time people used to stay up all night to buy copies. Sound familiar, iPad and iPhone lovers?

It’s weird this case is going on, and I won’t bore you with why. But it’s a useful starting point to look at how the landscape has changed in some ways, and in others not at all. Microsoft is still big, of course, but no-one queues up for their offerings anymore: Indeed nobody even bought Vista, as far as I can work out. But back then, nearly every computer you would ever use ran Windows and you would use Microsoft Office to do your stuff. You couldn’t leave because you probably didn’t have a modem and the Internet was a place where weird hackers lived.

Now, consider this landscape: Apple make most of their money from phones and tablets. Google, which wasn’t around when Windows 95 was, now dominate search, but also own a phone manufacturer, have built an operating system. Amazon, which back then was starting out as a bookseller, is now selling tablets at cost as a kind of access terminal to books, movies, magazines and other things digital. Facebook, which wasn’t even a glint in Mark Zuckerberg’s 11 year old eye at the time, is now the world’s biggest social network, but is really a vast walled garden where everything you do—from what you read, what you listen to, as well as how well you slept and who you had dinner with—is measured and sold to advertisers.

All these companies kind of look different, but they’re actually the same. Back in 1995 the PC was everything, and so therefore was the operating system and the software that ran on it. The web was barely a year old. Phones were big and clunky. So Microsoft used its power to dominate to sell us what made the most money: software.

Now, 15 or 16 years on, look how different it all is. Who cares about the operating system? Or the word processor? Or the PC? Everything is now mobile, hand-held, connected, shared, and what was expensive is now free, more or less. Instead, most of these companies now make their money through eyeballs, and gathering data about our habits, along with micropayments from data plans and apps, online games and magazines.

And to do this they all have to play the same game Microsoft played so well: Dominate the chain: Everything we do, within a Hotel California-like walled garden we won’t ever leave. So my predictions for next year, most of which  have been proved true in recent days : A Facebook phone which does nothing except through Facebook, an Amazon phone which brings everything from Amazon to your eyes and ears, but nothing else, an Apple-controlled telco that drops calls unless they’re on Apple devices. Google will push all its users into a social network, probably called Google+ and will punish those who don’t want to by giving them misleading search results. Oh, and Microsoft. I’m not sure about them. Maybe we’ll find out in Salt Lake City.

Podcast: Quaintness in Salt Lake

By | July 28, 2020

The BBC World Service Business Daily version of my piece on my predictions for next year  (The Business Daily podcast is here.)

Loose Wireless 110913

To listen to Business Daily on the radio, tune into BBC World Service at the following times, or click here.

Australasia: Mon-Fri 0141*, 0741

East Asia: Mon-Fri 0041, 1441
South Asia: Tue-Fri 0141*, Mon-Fri 0741
East Africa: Mon-Fri 1941
West Africa: Mon-Fri 1541*
Middle East: Mon-Fri 0141*, 1141*
Europe: Mon-Fri 0741, 2132
Americas: Tue-Fri 0141*, Mon-Fri 0741, 1041, 2132

Thanks to the BBC for allowing me to reproduce it as a podcast.
 

China’s Mystery Patterns

By | November 22, 2011

This has absolutely nothing to do with what I should be working on but this piece in Gizmodo caught my eye: a number of weird lines and structures in the middle of the Gobi Desert in China’s western reaches. Like this one:

image

They don’t seem to make much sense, despite some quite ingeniuous explanations by some of the commenters.

I’ve put all the locations in one Google Map here. I don’t claim to have the answers but I’ve found some clues.

While it’s true that they seem to have some military connection, they are not close enough to Lop Nur to be part of the nuclear weapons testing that took place there.

A book by John Wilson Lewis and Litai Xue called China Builds the Bomb says that Dunhuang, the nearest town, became the temporary base for a PLA unit in 1958 assigned to find the country’s first nuclear test base. Although they quickly moved further west (settling for Lop Nur), the Soviet advisors had come up with a site some 140 km northwest of Dunhuang, relatively close to where all the weird patterns are.

Part of the explanation can be found on an Australian military buff’s website.  It doesn’t give sources, but describes the patterns which most resemble airfields to be mock airstrips along with concrete pads that serve as targets for missile testing (the piece was written in 2005.) This would seem to suggest that the other patterns are also targets, although they’re not mentioned in the piece.

资料图:在2006年珠海航展上亮相的国产月球车。.

Another clue is in this machine-translated piece about China’s lunar ambitions. It says that Chinese researchers are based about 200 km from Dunhuang where the country’s version of the Mars Rover is undergoing testing in conditions “closest to the moon.” It says they have  built a “a board room, five generators…and a huge indoor stadium.” I can’t see anything like that but given what is out there in that desert I wouldn’t be surprised to find several.

Scammers Scam Gmail Scam Filters

By | November 22, 2011

This amused me. A scam message got through Gmail’s eagle-eyed scam filters telling me to update my account details. That’s not unusual. But was it because the scammers added their own assurance that they had already done the filtering?

image

It says:

**************************************************************************
This footnote confirms that this email message has been scanned by New Google Mail-SeCure for the presence of malicious code, vandals & computer viruses.
**************************************************************************

Well that’s alright then.