Wired publishes a short piece on Mike Matas’ new startup: the former Apple engineer and Delicious Monster co-founder and Kimon Tsinteris, another former Apple engineer, are creating a rival to Adobe’s digital publishing platform.
There are a few quibbles I have with the article—headlining the company’s connection to Al Gore (the two were originally contracted to make an “app” version of Gore’s new book and realized they had to create a new toolset to do it) seems aimed at bringing out political trolls, and it seems a bit overstated to say Push Pop intends to “blow up the book”—but it’s worth reading. (The comments are fascinatingly pathological, starting out with the “how dare they produce this for a proprietary platform” and proceeding predictably from there.)
Sarbanes-Oxley made IPOs difficult, but Wall Street has risen to the occasion by creating “secondary markets”: just like a normal stock market, but trading pre-IPO shares held by private investors and early employees. BusinessWeek reports:
The dynamics of wealth creation in Silicon Valley have fundamentally changed. Transforming private-company stock into cash and generating potentially massive wealth—even if you briefly worked at a startup that currently makes little money—is a click away on such sites as SecondMarket and its West Coast rival SharesPost, as well as via an organized network of buyers and sellers that has sprouted up almost overnight. Whether this new liquidity is healthy for startups, good for Silicon Valley, and beneficial for the U.S. economy overall are other questions.
So we have an unregulated market trading securities that are definitionally hard to value. Twitter’s 2010 revenue was estimated around [$45M]; they’re still funded chiefly by VC money.
Accredited investors—-those worth more than $1 million and with incomes north of $200,000 a year-—have recently bid on blocks of shares in Twitter at [an implied valuation of] more than $7.3 billion.
I’m sure many of you were worried, as I was, that early-stage Silicon Valley executives and engineers could no longer become millionaires by shoveling other people’s money into a big firepit as fast as possible. Good news, everyone!
…my little-birdie-informed understanding is that consolidated.db acts as a cache for location data, and that historical data should be getting culled but isn’t, either due to a bug or, more likely, an oversight. I.e. someone wrote the code to cache location data but never wrote code to cull non-recent entries from the cache, so that a database that’s meant to serve as a cache of your recent location data is instead a persistent log of your location history.
This sounds pretty likely. I’ve examined this file from my own iPhone and noticed a couple things which seem a little underreported to me. First, it looks a lot like what’s being collected isn’t your location—it’s records of when and where cell towers were “seen” by the iPhone. A comment to my post yesterday pointed out that even if the database file doesn’t contain device-specific data, it could still identify you (“the owner of the phone lives where the phone spends most of its nights”). I agreed then, but now I don’t think the CellLocation table—which is the one everyone’s squawking about—actually can tell you that. It can, though, put you in the right neighborhood. (As Andy Ihnatko put it, “While the logfile can’t tell someone that you were at a specific house, it can obviously tell your boss that you went to the Cape on the day you called in sick.”)
Second, while this has been mentioned in passing, it’s worth noting that this isn’t the only table in consolidated.db. It’s also recording every WiFi network that you see by timestamp, latitude, longitude, and MAC address—although this log, at least on my phone, seems to have much less historical information in it. There’s also a “compass calibration” table and a fair number of other tables that, at least in my database, are entirely empty. All of the cell location tables are repeated with Cdma prefixes, which is interesting if you want to fuel rumors about dual mode iPhones.
And, as it turns out, the database does contain device-specific data: the TableInfo table contains your iPhone’s serial number.
So the big Apple news going around right now is about the “recent” discovery of consolidated.db, a database stored on the phone that seems to track the phone’s location over time. From Christopher Vance:
Every time iAds or an app that uses Location Services pings the GPS service, a new record is created in either the CellLocation or WiFiLocation table respective of what type of network was being used. There’s also a Timestamp column in each table for each record. In a recent test, I found over 8000 records of stored GPS data. By default, Location Services for all apps is enabled. Also, being enrolled in the iAds program is enabled by default. In fact, the only way to cancel your enrollment from the iAds program is to go to Apple’s website.
I put “recent” in quotes because Vance’s article is from September 2010. This has been a subject of discussion in digital forensic circles for a while.
Some people have rushed to condemn this as another sign of Draconian Apple Policy™—I find it wryly amusing that this was retweeted by a local Android-using friend who refers to iPhone uses as “fanboys” even while his Twitter avatar icon is based on the Android robot. And around the web, the typical suspects like ReadWriteWeb are doing the headless chicken dance.
Others have rolled their eyes. What’s being recorded is what we were told was going to be sent to Apple anonymously in the iOS Terms of Service—the database records have nothing personally identifiable. If I handed you a dozen consolidated.db files, you’d be able to plot the movements of the phones associated with each file, but you couldn’t connect those files back to the phone.
Even so, the file doesn’t seem programmatically necessary. Applications receive the location data that will be recorded in the most recent consolidated.db update as an event once, when the location changes. There’s no API call for “oops, I wasn’t paying attention, give me the last 20 updates again, please.” It’s possible that for behind-the-scenes reasons the location event queue needs longer persistence, but it doesn’t need to stick around when there are no applications running that are subscribing to it. And even in the most generous interpretation, there’s no reason for the file to stick around between phone restarts.
What’s grating about this reporting is, as usual, the emphasis being placed on the fact that it’s an Apple product. The headlines are all variants of “Apple is recording your every move!” In fact, there’s no indication that Apple has any way to access this data at all—the only way to get it is to have access to your iPhone or to your iPhone backups, and to know whose iPhone the file belongs to.
The forest that’s being missed for the Apple trees here? Go back to the observation I made about where this has been discussed: digital forensic circles. I don’t want to claim that consolidated.db exists to aid forensics investigations, but it’s digital manna from heaven for law enforcement (and hackers). Yet any phone you use stores information locally—and if it’s a smartphone, that can be a lot of information, from your calendar to your browsing history. Call me a bleeding heart if you will, but the amount of “digital fingerprints” we leave has increased exponentially over the last two decades, and that trend shows no signs of slowing down. Search and seizure tools are keeping up with technology—but laws and regulations intended to keep searches and seizures reasonable are lagging far behind.
We introduce a new forensic technique that allows to collect users’ past locations on most current Android phones, within a few seconds. It becomes possible to tell where the user was at a given time, or where a phone call took place over the last few hours or days.
While I don’t like to engage in political navel-gazing here—Daring Fireball has the “Apple-centric tech blog with left-leaning political links” market cornered—it’s nice to see the occasional reminder that positive things can happen when politicians stop bloviating about how The Other Side wants to Destroy Everything America Stands For. Privacy advocates have noted that this legislation isn’t all that they hoped it would be, and it isn’t. But I think we—by which I mean everyone—need to get over the idea that compromises are effectively losses. They aren’t. Getting nothing is effectively a loss.
As part of the company’s comprehensive plan to align its operations, Cisco will […] integrate Cisco umi into the company’s Business TelePresence product line and operate through an enterprise and service provider go-to-market model, consistent with existing business TelePresence efforts.
I’ve got to wonder what this actually means. The “ūmi” service, as Cisco styles it, is basically televideo conferencing for the living room, pitched at families. Think of, say, Apple’s FaceTime, except you can’t talk to mobile devices, you can only talk to other people who’ve ponied up $400 for Umi boxes and are paying $99 a year for the service. Puzzlingly, people haven’t been beating down the doors for this service.
The curious thing about the realignment memo isn’t that it mentions Umi, it’s that it doesn’t mention killing Umi, a product which PC Mag described charitably as “probably the grossest miscalculation in Cisco’s history.” (The charitable part was the “probably.”) What does it mean to integrate it into “existing business TelePresence efforts?” Sure, we all know that if there’s one thing that the videoconferencing market has been crying out for, it’s expensive equipment with cutesy names, but even so.
Calling Creative Suite 5.5 a “mid-cycle product release,” Adobe said it focused on three main areas: Proliferation of Devices; Design and Interactivity; and Monetization.
In announcing this mid-cycle release, Adobe said it was changing the schedule for its major announcements. The company will go from an 18 month development cycle to a 24 month schedule, but will include mid-cycle releases in between.
If these “mid-cycle releases” are paid upgrades—which CS 5.5 is—I’m fairly sure that makes this what people on our planet call a 12 month schedule. To be fair, Adobe did say that one of the three main areas they were focusing on was “monetization.”
In 2005, LiveJournal was bought by Six Apart, the Movable Type guys, as near as I can tell so they could pilfer through its ideas and try to launch a better version called Vox. LJ itself was then sold to SUP in 2008, a Russian media company—it turns out LJ was big in Russia and getting bigger. I wrote an article back then on my own LiveJournal in which I estimated that 30% of the extant LiveJournals were Russian.
LJ is an interesting beast—in some ways it almost seems like an illustration of the dangers of being too early to market. It was never taken seriously as a hosted blogging platform by the digiterati, being dismissed as a place for angsty teens. It is a place for angsty teens, but it’s also been a place for a lot of SF/fantasy types—I’ve been amazed at how many professional authors journal there—and if we’re honest, we have to admit it was doing “social networking” years before that became a buzzword. The LJ “friends list” is the first place many of us saw the social graph that defines Facebook and Twitter. And at least for me, it was that mechanism that kept me from moving to other blogging platforms for a long time; the attempts to replicate that with “trackbacks” and “pingbacks” never worked well (although Disqus may have finally cracked that nut), and up until Tumblr, none of the other hosted blogging platforms put much effort into the whole “follow/like” dynamic. (Other than Vox, of course, which despite being pretty successful on a technical level never got any traction.)
But LJ has kept growing in Russia; despite a fellow nitpicking with my numbers back in 2008—my fault, perhaps, for not explaining that the 30% figure I’d gotten came from SUP’s press releases, not from the stats page he’d found that contradicted it—the Financial Timesreports that over 50% of LJ’s users are from Russian-speaking countries.
I bring this all up because LJ users have noticed LiveJournal misbehaving over the last week—apparently, that’s because it was subject to a denial of service attack aimed at Russian bloggers which, since LJ’s servers are still located in the US, took it out for everyone. Prominent LJ bloggers include president Dmitry Medvedev himself, along with several prominent Kremlin critics. This has led—as the FT article reported—to the curious spectacle of Medvedev calling for an investigation of the attacks on LJ. Even so, it’s hard not to notice that the other sites that were targeted were anti-Kremlin.
There are a few things that fascinated me about this. One of them is just how important LJ is to online political discussion in Russia and other former Soviet states; there’s no one site in the West that I can think of that’s analogous. Nobody is going to target Twitter or Facebook in an effort to suppress what passes for political commentary there.
Another is SUP’s deliberate choice to leave the servers in the United States so they’re under our data protection laws. A lot of pundits here of both the professional and armchair variety toss around phrases like surveillance state and government takeover in a way which seems very cavalier to me. Russians understand what those really mean. For the most part, we don’t.
Amdahl is the gentleman who coined the acronym “FUD,” after he left IBM to start a rival mainframe company and accused IBM salesmen of spreading fear, uncertainty and doubt about his products to keep people from buying them.
The idea of “FUD” has always been tied up with fights against competitors: anti-Apple FUD, anti-Linux FUD, whatever. There’s somebody benefiting from that fear, uncertainty and doubt—it’s a sales tactic. But who’s out there arguing that Google “tightening control” over Android is a reason to stop using it? If open source developers turn against Android, it’s not as if they’re going to all flock to iOS and Windows Phone 7. If Rubin hadn’t invoked Amdahl, this would just be another vapid exercise in mild damage control, but the touch of paranoia elevates it to slightly bizarre.
The interesting bit about this article isn’t in the headline, it’s in this bit:
YouTube is looking to introduce 20 or so “premium channels” that would feature five to 10 hours of professionally-produced original programming a week. It is planning to spend as much as $100M to commission the creation of original content for the premium channels, the people familiar with the matter said.
YouTube is talking with big talent agencies and production companies, as well as considering “tastemaker” channels, which sound like curated lists of existing content.