It has happened before. It will happen again. And it can happen here

Theodore Gray:

Once the tool exists, it will be abused. I really wish I could say otherwise. I wish I lived in a country where law enforcement acted with the square-jawed nobility of a comic book FBI agent. I used to think I did live in such a country, but in recent years I have been forced to realize that I don’t. I think that, in your heart, you know it too.

And if you happen to like the party in power today, just remember, every tool you give them is inherited by the next guy. Think of the worst possible outcome of the next election, the worst new president, the worst new congress you can imagine (whichever one that might be for you). Do you want those people to have the power to look and listen to anyone they please, anytime, anywhere? To search at will through recordings of our most personal moments? To use a rubber stamp warrant to gather dirt on someone who insulted them? To review the browsing history of anyone who annoys them publicly enough?

An invocation of what I call the “Worst Enemy Test” spotted in the wild.

Cartography Comparison

Justin O’Beirne did a great in-depth comparison of the different ways Google and Apple approach mapping.

From the summary of Part 1:

We looked at 54 pairs of maps across three cities (New York, San Francisco, and London) and found several significant differences:

  • Apple Maps, on average, labels more cities than Google at every zoom.

  • Google Maps, on average, labels more roads than Apple on nearly every zoom.

  • For two-thirds of zooms, both maps generally show the same number of roads. For the remaining third, Apple almost always shows more roads.

  • Both maps, on average, label a similar number of POIs [Points Of Interest] —but have only 10% of their POIs in common on an average zoom.

  • Both maps also prioritize different kinds of POIs: Google Maps heavily prioritizes transit, while Apple prioritizes landmarks. Apple also generally shows a greater number of POI categories on a given zoom—and shows twice as many restaurants and shops as Google.

When Apple Maps launched in iOS 6, I wrote a short post comparing it to Google Maps in Japan. I found that — for my location at least — Apple’s implementation was actually better in some ways at launch than Google’s was after several years.

The main reason I don’t use Apple Maps preferentially is the continuing lack of support for transit. Train scheduling and transfer information is hugely useful in Japan, so I end up using Google Maps much more often for station to station directions, while I usually default to Apple Maps for local guidance when I reach a location. It’s interesting to see that my purely intuition-based switching between them for different roles apparently has some empirical support as well.

Taking a Walk in Someone Else's Skin

This Game Is Forcing Some Players to Be Women, And They’re Freaking Out

From Motherboard:

It’s been several months since we wrote about how Steam survival hit Rust would soon be implementing a feature that randomly (and permanently) makes your character a man or a woman, and now that feature is here. Players have generally loved the way the game randomizes features for your permanent character like skin color, limb length, and (yes) penis size, but that was all back when every player avatar in the game was a male. With the latest update, though, everyone got a new model, and for some players, the possibility of playing as a woman is a little too much to bear…

Honestly, this is brilliant. I can’t think of a more fitting fate for some of the horribly sociopathic and misogynistic males, that unfortunately infest large areas of the gaming landscape, than to have to see things from a woman’s point of view.

Twitter’s Missing Manual

One of Twitter’s problems is that it’s tilted a little too far towards the vim end of the scale. It looks like a dead-simple service, but those humble 140 characters have been crammed full of features over the years, and the ways they interact aren’t always obvious. There are rules, and the rules generally make sense once you know them, but it’s also really easy to overlook them.

I’ve known about Twitter since around 2006, but I’ve only been using it — haphazardly at first — from maybe 2011. Even though I’m pretty good about figuring out unwritten rules (see: living in Japan), as a semi-newb to the service there are a lot of things about how it works that I really didn’t know. This is a fantastic resource.

TextExpander as Subscription Software

TextExpander as Subscription Software

When Should Software Be a Subscription Service?

I don’t contest the fact that developers need income. I have gladly upgraded TextExpander every time there was a new version, even if the changes weren’t important to me. It’s important to support independent developers, to ensure that the Mac and iOS ecosystems have excellent apps. In addition, the race-to-the-bottom pricing that we’re seeing with the App Store model is preventing developers from making enough profits.

This blew up while I was still recovering from some minor surgery and so I missed most of it as it happened. Kirk McElhearn’s take on the pricing part of the equation is very nearly the same as mine.

Besides the huge increase in average yearly pricing that McElhearn laid out very clearly, I have three major problems with the subscription model:


  1. If TextExpander can’t connect to a server, will the snippets disappear until it can? If there’s no offline mode, TextExpander would be basically useless to me. One of the places I work has a very spotty internet connection. Even on my iPhone, I’m occasionally completely offline, but am still doing work that I would normally use TextExpander for.


  2. I have serious reservations about the security of Smile’s syncing solution. I do not have particularly sensitive information in my snippets, but I do have things like email addresses, physical addresses, and phone numbers. All of that will be stored unencrypted on their servers. They seem to be pushing toward an enterprise market, but the level of security Smile is committing to at this point is inadequate for an only moderately-concerned private user like me.

    In TextExpander’s manual they advise against using it to store anything potentially compromising, like passwords, but for many companies contact information or simply the names and positions of certain individuals in an org chart could be considered sensitive information.

    Usually, in enterprise-level software, companies want to run it on their own servers that they control, and they contract with the software provider for support service for their IT staff. I’m pretty sure most companies who might otherwise consider TextExpander would balk at Smile’s proposed syncing solution.


  3. This move offers zero benefit to me, the user. I don’t use Windows, and haven’t for about 15 years. If I did, I probably would be pretty happy about the announced support for that platform. I don’t work with teams and I don’t share snippets with a group. Having to use someone else’s shortcuts sounds like a special kind of hell to me.

    I have no incentive to “upgrade” other than the stick of eventual obsolescence when the old version breaks. While I could possibly justify the price, given how much I use TextExpander on both OS X and iOS, it’s still a significant jump. The announcement of the pricing change has prompted me to explore other options, and I’m sure I’m not the only formerly-loyal TextExpander user doing so, which can’t be good for Smile’s future sales.


One small misstep like this can drastically change people’s perceptions of your company. Apple can get away with taking away peripheral ports and eliminating optical drives because they generally get it right in the long-term, and sometimes offer temporary fixes in the transition.

Smile probably should have offered a two-tier pricing model: Basic users pay a one-time fee for major updates; “Pro” users pay an additional recurring cost for access to certain capabilities. Most people would have had no negative reaction to that model, and even basic users would have considered the pro upgrade to see if it fit their needs.

Unfortunately, Smile alienated a large section of their most loyal user base with an announcement that made it obvious that Smile’s future priorities and their customers’ priorities are definitely not the same. If it was their way of “firing” those customers in favor of a different user base, as some have speculated, Smile has definitely not put in the kind of work they need to successfully court the enterprise market, even though all of the announced features sound as if they were created for business users over individuals.

For me, ultimately, the problem isn’t the pricing. The problem is that I can’t count on Smile to continue making a product that works the way I want it to work. They’re solving problems I don’t have, and pushing features that make me distrust their future versions.

"If You’re House Is Still Available, Send Me an Email"

If You’re House Is Still Available, Send Me an Email: Personality Influences Reactions to Written Errors in Email Messages

Interestingly, the headlines overwhelmingly use negative language like “jerk”, “pedant”, “grammar Nazi”, and “snob” to describe the study findings. I guess I’d be considered disagreeable in pointing out the inherent judgment expressed by most of the headline writers referencing this article, but that’s okay because they’re probably “jerks” just like me.

People who let errors — particularly those errors called “grammos” in the article — negatively influence their rating of the person making those errors as a roommate apparently also rank highly in “disagreeableness” as rated on the Big Five personality scale.

“Disagreeability” is partially cultural/geographical. Think: southern hospitality contrasted with east-coast brusqueness. I’m, like, Californian by upbringing, dude, so I’m, like, kinda in the middle, you know.

Maybe I’m just a disagreeable pedant, but if you take a look at the questionnaire [1] there seems to be a key omission: “None of the above”, or preferably an “Aw, hells nah!” choice.

I think I would be friends with this person [sic.][2]


The writer would be a good housemate.


The writer seems a lot like me.


The writer seems friendly.


The writer seems more sophisticated than most of my friends.


The writer seems less intelligent than most of my friends.


The writer seems conscientious.


The writer seems considerate.


The writer seems likable.


The writer seems trustworthy.


This email flowed smoothly.

Participants were asked to rate these on a scale of 1 (low) to 7 (high) according to how much they agreed or disagreed with the statements.

In my not-so-humble opinion, someone who makes typos in the age of autocorrect goes beyond carelessness to willful negligence; you have to actively disable the function since it’s on by default in most OSs, desktop or mobile. Even so, most people, even “jerks” like me, are more willing to overlook typos than grammatical errors or errors that occur in the grey area between spelling and grammar.

In this hypothetical situation, the person is sending a communication to ask to be a roommate. I had many, many roommates between moving out and the end of university. At my last place, there were 5 of us sharing a house. I was on good terms with most of the guys — especially during the last couple of years when there was less turnover — and hung out on a regular basis with 3 or 4 of them. Through sheer experience, I’m pretty good about figuring out who would be a good roommate, and who I can be friends with. These are sometimes very different people.

Agreeableness is not high on my list for a good roommate. Contentiousness is. I could easily be friends with someone I would never, ever want to live with.

It’s been a very long time since I lived with a roommate (lovers, spouses, and offspring are in a totally different category) so I would evaluate these survey responses the same way I would have when I was a college student.

Even misanthropes can follow social conventions and be tolerable company in short bursts. People who leave rotting food on a plate that ends up buried behind couch cushions are never going to be tolerable roommates. Someone who is casually friendly when encountered in the house, but spends most of his time studying and almost never goes out with the others is a good housemate. The guy who acts like your best friend 30 seconds after he meets you and comes back stumbling, roaring drunk at 3:00 AM to fuck his on-again, off-again, oh-gods-they’re-back-together-again girlfriend in marathon wall-banging, operatically screaming orgasmic sessions is not a good housemate.[3]

Which type do you think is more likely to write, “If you’re house is still available”?

I think I would be friends with this person.

I’ve never met this (fictional) person, so have no idea if I would actually be friends with them based solely on an email. I need to actually meet them to make that kind of decision. Tell you what, though, I could consider whether I am capable of overlooking errors in communication in someone I choose to consider a friend.[4]

The writer would be a good housemate.

I’ll tell you what’s a big fat red flag when you’re looking at a potential roommate: attention to detail. You know, the little things, like what day rent and utilities are due, whether Tuesday or Thursday is garbage pick-up, how many weeks take-out has been mouldering in the back of the fridge, remembering that occasionally you need to scrape clothes off the floor before they grow mold and you wonder whether it would be better to bleach the rug, or tear it out and burn it (and maybe some of the affected floorboards too) to get rid of the stench.

Someone who doesn’t bother to check an email for “typos” and “grammos” in a first communication is someone who won’t bother to think about all the myriad details living with other people requires. Being nice and getting along well with other people doesn’t help much when you never pay for anything on time and your lack of room hygiene makes the whole house stink.

You could be an über-bubbly supermodel who performs oral sex that makes angels weep with joy, but if you clog the toilet with a wad of toilet paper that a firehose couldn’t force down the drain, and keep forgetting to lock the door so we get robbed on a weekly basis, I’m probably not going to want to live with you. And you probably use too many emoji punctuating messages like, “hay, its not mi prolblem, it s they’re problim!”

The writer seems a lot like me.

You know who seems a lot like me? Someone who proofreads messages before hitting the “Send” button.

The writer seems friendly.

Friendly ≠ ignorant and careless. Like I said earlier, I’ll completely accept silly stuff from friends. This is not that situation. The person is a suppliant. They are asking you to consider them for a place in your home, as someone who shares space with you every single day. This communication is supposedly representative of what they are like at their most conscientious. If they are lax here, how bad will they be after they move in, make your life hell, and you legally have to enter eviction proceedings to get rid of them?

The writer seems more sophisticated than most of my friends.

<snort> That is not a high bar. Hell, more sophisticated than me is not a high bar. Know what’s sophisticated AF? Correctness. And being able to belch the Sumerian alphabet after chugging a seasonal artisanal brew sourced from my buddy’s basement.

The writer seems less intelligent than most of my friends.

Not necessarily less intelligent, but certainly more careless than I’d like for a roommate. As far as I’m concerned, it’s an irrelevant response since I wouldn’t care what the person is like in relation to my friends. There is, unfortunately, no “not applicable” option on the survey, which is a shame since it would cut the necessary responses to about three.

The writer seems conscientious.

if error_count=0  
    then applicant=conscientious  
    elseif count_errors  
        select error_count (*) from error_count_table;  
        where error_count=count_errors  
endif

error_count_table;  
+-----------+--------------------------------------+  
|error_count|response                              |  
+-----------+--------------------------------------+  
|1          |"OK"                                  |  
|2          |"Meh"                                 |  
|3          |"Hmm"                                 |  
|4          |"Really?"                             |  
|5          |"Ugh"                                 |  
|6          |"You’ve got to be fucking kidding me!"|  
|7          |"Lost…will…to read"                   |  
+-----------+--------------------------------------+

The writer seems considerate.

Considerate is providing clear and efficient communication that does not waste my fucking time. I’m screening people looking for a roommate who won’t puke in the front hall and fall asleep face-down on my bed with their shoes on my pillow at 2:00 PM on a Wednesday, when I’ve got 20 whole minutes to change, dump study materials, and get to my job after class. You must convince me that you’re both “agreeable” and not a waste of oxygen. You’ve got one email … GO!

The writer seems likable.

Honestly, I don’t care. I probably won’t see you for more than a few minutes a day while we live together because people usually have wildly incompatible schedules when they’re at the stage of their lives when they’re living with roommates.

What I do care about is not having problems. It’s like a teacher once said about parent-teacher conferences, “If I know little Timmy’s name within the first couple of weeks, nothing good has been happening. The fact that I can’t immediately pick your kid from a list of 30 other students is a good thing.”

I’m human. I’ll cut people more slack if they’re nice and fun to be with. But if I haven’t even met you and you’re coming across as friendly but ditzy, I’m swiping to “Archive” and moving on to the next email.

The writer seems trustworthy.

“LOL so i’mm lik sooooo sarri but ill toadally get the $2u buy 2morr!!!11 🤑”

🙄 Um, yeah, sure. Completely trustworthy individual.

This email flowed smoothly.

(Must. Resist. Poop joke.)

If I ever have to pause in reading an email to determine meaning, no, it did not “flow smoothly” by any reasonable definition of “flow” or “smoothly”. You can make writing flow smoothly by mostly using clear, declarative sentences. Spell correctly. Use punctuation.

It’s not particularly complicated. It doesn’t even have to be perfect. I’m not a super-pedant. Obvious errors demonstrate that you have a complete lack of fucks to give. You didn’t care enough to check it, read it back to yourself, or fix typing mistakes. If you don’t care enough to put that tiny bit of effort into your message, why should I care enough to read it?

I guess that just makes me a disagreeable snobby jerk grammar Nazi pedant.





  1. Quoted below because the link doesn’t properly resolve to Table 1, even though it looks like a direct link. Don’t look at me, I didn’t code their page or the CMS.  ↩

  2. Yes, I went there. The first response in their survey is missing a period. Ironic, that, since I’m assuming it wasn’t deliberate.  ↩

  3. These are composites of actual roommates I have had. There are many reasons I lived alone as soon as I could afford to, despite the good roommates I met and have stayed friends with for years afterward.  ↩

  4. Has your mom ever responded to, “Can you bring me the remote control, please?” with, “I can”, and no action? No? It was just my mom who taught the difference between can and will through demonstration? “Okay, mom, will you bring me the remote control. Please. With sugar on top?”  ↩


Apple Pay in Japan

A month with Apple Watch as my wallet

Tom Warren for the Verge:

It all started when I left my house without my wallet. I had ventured out to grab some lunch with no cash or cards to pay for it, only my iPhone or Apple Watch. I’ve used Apple Pay on my iPhone and Watch before, but this was the first time it was actually useful. I paid for a sandwich at my local store with my Watch, and thought nothing more of it.

via: The Loop

Meanwhile, in Japan:

Apple (日本) - Apple Press Info - Apple Pay、モバイルペイメントの変革に向けて10月20日より運用開始

The line at the top 「米国報道発表資料抄訳※—2014年10月17日」 translates as “Selected translation of report released in the US on Oct. 17, 2014”. Like it says on the tin, it’s basically the same text as the original English-language Apple Pay press release. That’s the first and last official news from Apple about Apple Pay that has been released in Japan.

Despite the lack of stated support for Apple Pay in Japan, Touch Lab, an Apple enthusiast site, tested out Apple Pay in February of 2015 to see if it would work with existing NFC terminals and payment systems. Sometimes features aren’t officially supported in other countries, but work anyway. This wasn’t one of those times.

今回試した中では3件中1件で使用できましたが、そもそもEMVコンタクトレスの端末を置いているところがほとんどなく、結局「現状Apple Payは国内でほぼ使えない」ということを再確認しただけでした。

国内では既にFeliCaが普及していることもあり、国際規格であるEMVコンタクトレスの普及が他国より遅れ、日本だけApple Payが使えない状況にならないか心配になります。

5年後のオリンピックでは海外からの旅行客が急増することが予想されるため、その準備として端末が増えることを期待したいところです。

My translation:

At one of the three sites tested at this time it was able to be used, but there are almost no EMV Contactless terminals, so in the end the statement, “Apple Pay can’t be used inside Japan”, has to be reaffirmed.

In Japan, FeliCa has already penetrated. The international standard EMV Contactless is penetrating much more slowly than in other countries, so it is worrying that possibly only in Japan, Apple Pay may not be able to be used.

At the Olympics to be held in five years, it is expected that there will be an increase in visitors from overseas. In order to prepare, hopefully terminals [that can process foreign payments] will be increased.

In other words, while elsewhere Apple Pay has been hailed as possibly the best thing about the Apple Watch, a game-changing implementation that might actually bring NFC payments into the mainstream, in Japan it’s a nearly-useless non-feature. Apple has not been able to either work out deals with market incumbents, or push the adoption of systems that integrate with Apple Pay in the year and a half since its US launch and worldwide announcement.

This is not surprising. Despite Japan’s image overseas as a country that is at the forefront of technological progress, it is an often hostile market for foreign products, particularly those that compete with any incumbent technology, and is bewilderingly behind the times in some ways. There are still many places even in Tokyo where you must pay cash; no credit or debit cards are accepted.

The apparent head start Japan had by implementing contactless payment systems, starting with train passes over a decade ago, has stultified, as all too many products developed for the Japanese market do. Japan has a tendency to be first and best with a technology that is quickly superseded, and then are unable to change course due to massive investment, or create an overspecialized version suited only for the peculiarities of the domestic market.

Desiging for the the Japanese market first and foremost has even led to the coining of a phrase, gara-kei (ガラケー) or Galapagos handset, which I mentioned in Why Apple isn’t Japanese nearly four years ago, and explained in more detail in Mobile Internet in Japan a couple of years before that.

While the phrase was first applied to Japan’s overloaded “feature” phones that quickly seemed inferior to “smart” phones, the idea has been applied to other technology. FeliCa is, in my opinion, one of these technological cul-de-sacs that is well-suited to the domestic market, but unlikely to be adopted overseas, and actively interferes with any adoption of international standards. Technology Nazi[1] says, NO APPLE PAY FOR YOU!


  1. Soup Nazi reference provided because I’m old and can’t trust that whippersnappers will get an allusion to a comedy that went off the air around the time they were being conceived.  ↩

Encryption, Privacy Are Larger Issues Than Fighting Terrorism

From the transcript of the NPR interview between David Greene and former NSA official Richard Clarke:

CLARKE: … If I were in the job now, I would have simply told the FBI to call Fort Meade, the headquarters of the National Security Agency, and NSA would have solved this problem for them. They're not as interested in solving the problem as they are in getting a legal precedent.

GREENE: Wow, that sounds like quite a charge. You're suggesting they could have just gone to the NSA to crack this iPhone but they're presenting this case because they want to set a precedent to be able to do it in the future?

CLARKE: Every expert I know believes that NSA could crack this phone. They want the precedent that the government can compel a computer device manufacturer to allow the government in.

As suspected, there is probably a physical way around iPhone security features. One of the truisms in computer security is: if someone has physical access to the device, there is virtually no security that cannot be bypassed.

Apple, FBI, and the Burden of Forensic Methodology

The best overview of the technical aspects of what the FBI is asking Apple to do is at Zdziarski’s blog starting on his February 18 post “Apple, FBI, and the Burden of Forensic Methodology” (linked above) and subsequent follow-up posts. The most frightening section there was:

FBI has asked to do this wirelessly (possibly remotely), which also means transit encryption, validation, certificate revocation, and so on.

I have seen virtually no commentary about this point, which I think is a big, big issue. With previous data extraction cases, Apple took extensive precautions, including requiring investigators to physically transport the iPhone to the Apple facility, and isolating the unit within a faraday cage. In other words, law enforcement had to have physical possession of the device. As many security researchers have pointed out in the past, with physical access it is almost guaranteed that the attacker will find some way to read some or all of the data stored on a device.

With an over-the-air attack tool, anyone who finds a way to bypass the supposed safeguards of the tool could target anyone at any time; they would not need physical access to the device. That makes it significantly easier for an attacker to bypass the security features and unlock the targeted iPhone. And once that happens they can do just about anything they want, including load malware, wipe the device, or do a data dump. With a sufficiently sophisticated tool paired with an over-the-air attack, the person might not even know that their iPhone has been hacked.