Last night I took a Peter Pan bus from Boston to Hartford, CT. The interior was brand new, and so was the little spiel the driver gave as he pulled out of the gate.
He thanked us and gave us an ETA, directed our attention to a short safety video from Peter Pan corporate, and told us that he'd talk about mobile devices and Wi-Fi afterward.
But he forgot. His speech was never finished, the Wi-Fi was never enabled, and he also forgot to turn off the screens, so the bright DVD menu stayed on for the duration of our nighttime trip.
But I'm not complaining - it was an interesting design case study. Clearly the new bus smell and procedure had distracted the veteran driver enough to miss a couple of steps, but I don't blame him at all. I blame the design of the bus' systems instead, and whoever caused them to work that way.
The bus' Wi-Fi and screens are passenger-oriented systems that are entirely unrelated to what the bus driver should be focused on: driving the bus. I'd argue that the driver shouldn't have to worry about the state of either system, and that they should take care of themselves automatically based on the bus's speed and display input activity.
Unless I'm missing an important component here, it seems like if the bus has been traveling for a few minutes (meaning it's out of the terminal where random people can mooch off of its connection) then the Wi-Fi should turn itself on, and if the A/V system hasn't displayed anything or been controlled after a certain number of minutes, then it should turn itself off. DVD players did this many years ago, and in-bus accelerometers/GPS are far more advanced than they'd need to be for auto-Wi-Fi.
If these errors had been anticipated or spotted during usability testing and fixed, our fancy new bus ride experience could've been close to perfect. Instead, most of us were mildly annoyed by the driver's mistake without considering the actual root of the problem: poor systems design.
The Jeep Cherokee (and I'm sure others) can be completely hacked through its Sprint-powered Uconnect feature. Wipers, transmission, steering, brakes, etc. With knowledge of the car's IP address and the right timing, drivers and their passengers could be killed remotely from anywhere in the world with a burner phone.
Of course, that's not exactly the kind of language Fiat Chrysler Automobiles used in their firmware update announcement:
Similar to a smartphone or tablet, vehicle software can require updates for improved security protection to reduce the potential risk of unauthorized and unlawful access to vehicle systems. Today’s software security update, provided at no cost to customers, also includes Uconnect improvements introduced in the 2015 model year designed to enhance customer convenience and enjoyment of their vehicle. Customers can either download and install this particular update themselves or, if preferred, their dealer can complete this one-time update at no cost to customers.
Via a USB stick of course, for whatever dumb reason.
“Under no circumstances does FCA condone or believe it’s appropriate to disclose ‘how-to information’ that would potentially encourage, or help enable hackers to gain unauthorized and unlawful access to vehicle systems,” the company’s statement reads. “We appreciate the contributions of cybersecurity advocates to augment the industry’s understanding of potential vulnerabilities. However, we caution advocates that in the pursuit of improved public safety they not, in fact, compromise public safety.”
Ahem - and cost us boatloads of money from product recalls and civil lawsuits - Ahem
The two researchers say that even if their code makes it easier for malicious hackers to attack unpatched Jeeps, the release is nonetheless warranted because it allows their work to be proven through peer review. It also sends a message: Automakers need to be held accountable for their vehicles’ digital security. “If consumers don’t realize this is an issue, they should, and they should start complaining to carmakers,” Miller says. “This might be the kind of software bug most likely to kill someone.”
Something to remember as our cars become increasingly connected and automated. Chrysler is taking the heat right now, but I'm sure that other systems are just as vulnerable. Earlier this year BMW's ConnectedDrive system allowed anyone with some simple parts to wirelessly unlock doors, for example, and I'm sure we'll see plenty of other serious exploits in 2015 and beyond.
Also, putting people in danger by demonstrating the extent of the exploit on a highway with traffic seems incredibly dumb to me. What the heck?
Interesting thoughts from Brian Krogsgard on independent blogging, and in particular the need for a new container to emerge that combines the best of closed platforms like Medium or WordPress.com with the benefits of self-hosted solutions (total control, subscription cuts, etc) to make blogging more sustainable in the future.
Effective & concise piece by Jesse Weaver on the recent "X-ers should learn to code" movement:
Saying designers should code creates a sense that we should all be pushing commits to production environments. Or that design teams and development teams are somehow destined to merge into one team of superhuman, full-stack internet monsters.
Let's get real here...
Designers and developers should know enough about the other discipline to be able to anticipate problems and communicate more effectively, but stepping on toes and actively trying to do the other person's job isn't recommended, as Jesse describes.
The world of MFi controllers is still a bit of a crazy one at the moment....
I would have hoped that by now any game that should support controllers would, but sadly not. Part of this is that despite a standardized protocol, you still hear of developers having issues with individual controllers.
So, developers are wary to add controller support, because it's pretty much a situation like on the PC, where there's all sorts of weird devices to potentially support. But with Apple certification, this shouldn't be happening. It scares off developers from implementing MFi controls, which leads to fewer games getting them, which leads to fewer people wanting to buy controllers, which leads to fewer games supporting them as the audience for controllers is ever smaller...
If Apple wants iOS (or the next Apple TV's OS) to have the kinds of games you'd see on dedicated game console, I'm convinced that they're going to need their own first-party controller. There has to be a standard for the platform, and an Xbox 360 controller knock-off made by Mad Catz isn't going to cut it.
Side thought: wouldn't it be interesting if, when the next Apple TV is eventually revealed, a representative from Nintendo walked on stage and announced that all existing bluetooth Wii Remotes and accessories would be compatible with games on the Apple TV and iOS 10? It's technologically possible, and if Apple wants gaming on Apple TV to thrive but doesn't want to get into the mess of creating gaming accessories itself, Nintendo's support would put it many years ahead of anything Android attempts to do.
A number of factors make this scenario unlikely, though:
Nintendo has already confirmed that they're working on another dedicated console, so enabling their controllers to work on the Apple TV would be like arming their worst enemy.
Nintendo still hasn't experimented with iOS gaming yet, even though they plan on doing so later this year. They wouldn't be comfortable with a move like this.
Both Nintendo and Apple are strong believers in controlling both hardware and software. If Apple allows Nintendo to become a (quick-to-be-dominant) controller-maker for the Apple TV, Nintendo is going to start probing about Apple's future hardware. That kind of relationship will lead to divorce unless Nintendo exits the dedicated console market (unlikely anytime soon) and signs a massive deal with Apple. I doubt that'd happen.
If you want 9 minutes of action and 31 minutes of exposition and dumb banter between hosts who don't know anything, click here. If you want 50 minutes of actual engineering and design reasoning, click here, here and here.
Greg Munson (co-creator): Robot combat how it is in traditional "BattleBots"-style needs some help. It needs a spark. It needs a bit of a game change.
Dan Danknick: I don't think anything will revitalize robot combat. We got lucky. We picked a hobby that could help build our skill set. I'm not the only systems engineer at the company I work for now, but I'm sure I'm the only one who can weld, machine, and drive a forklift. But the technology and the desire and the effort, and really a motivation for doing it, just doesn't exist right now. It was new, it was fun. I had spare time, the future seemed bright, I was still wearing shades. Would I do it now? No. There's better things I'd do now.
My guess is that the "game change" here is mostly just opportunistic timing with the recent popularity of maker culture. If geeks and makers tune in, great. If they don't, ABC is only producing 6 episodes this season anyway.
With this amount of junk and filler, I don't think this will catch on. I'd much rather watch a compiled YouTube clip of the action and subscribe to Tested instead.
Update: A few days later, those compiled action clips are now available: Episode 1 and Episode 2.
Apple has been using its considerable power in the music industry to stop the music labels from renewing Spotify’s license to stream music through its free tier. Spotify currently has 60 million listeners, but only 15 million of them are paid users. Getting the music labels to kill the freemium tiers from Spotify and others could put Apple in prime position to grab a large swath of new users when it launches its own streaming service, which is widely expected to feature a considerable amount of exclusive content. "All the way up to Tim Cook, these guys are cutthroat," one music industry source said.
70% of the music I listen to today was discovered through Spotify's incredibly convenient interface and free tier. Not through the tiresome (and major label-biased) Pandora + iTunes combo I used before Spotify launched in the US. "One click listen" proved to be so much better than "one click purchase" in the context of music-listening and discovery that I've been a paying member ever since they ran their (smart) $1/month promotion back in December.
Spotify is this generation's iTunes equivalent circa 2001. It offers a significant improvement over the old way of listening to music, and unlike NapsterGrooveshark its free tier is legitimately operated and licensed. The notion that Apple is trying to starve out a better product by working with the major labels that want more money (for themselves, not necessarily for Taylor Swift) is frustrating.
If you haven't yet, you should read or watch Steve Albini's thoughts on digital music distribution, particularly the last half. Here's a relevant piece:
I disagree that the old way is better. And I do not believe this sentence to be true: “We need to figure out how to make this digital distribution work for everyone.” I disagree with it because within its mundane language are tacit assumptions: the framework of an exploitative system that I have been at odds with my whole creative life. Inside that trite sentence, “We need to figure out how to make this work for everyone,” hides the skeleton of a monster.
Let’s start at the beginning. “We need to figure out”: the subject of that sentence, the first-person plural, sounds inclusive but the context defeats that presumption. Who would have the power to implement a new distribution paradigm? Who would be in the room when we discuss our plans for it? Who would do the figuring out we need to do? Industry and consumers? Consumers is a likely response, but did the consumers get a vote about how their music would be compressed or tagged or copy protected or made volatile? Did anybody? Did the consumers get a choice about whether or not Apple stuck a U2 album on their iTunes library? Of course not. These things were just done and we had to deal with them as a state of being. Consumers rebelling or complaining about things – “market pushback” – isn’t the same thing as being involved in the decision to do something. Clearly the “we” of this sentence doesn't include the listener. I believe any attempt to organize the music scene that ignores the listener is doomed.
Apple cozying up with big label "monsters" and strangling Spotify's free tier won't curtail piracy the way it did in 2001, that's for sure. Spotify may not be the greatest solution for small artists either, but it's clearly frightening the big labels a lot more.
Microsoft announced the name of its Windows 10 browser last week. While the code name, Project Spartan, was exciting and invoked visions of dominance and strength, the final name Microsoft picked is totally dull and mundane. Microsoft Edge is the successor to Internet Explorer, but the software giant had a huge opportunity to bury the bad image of IE, and it has blown it.
I don't think they blew it.
Choosing "Edge" allowed them to continue using a blue "e" for their logo in the taskbar. In the minds of IE users, that little blue "e" is the internet, and replacing it with a fancy "S" for Spartan would make them frustrated enough to start digging into the Start Menu and eventually find the old Internet Explorer.
With this name and logo, Microsoft has a prayer of converting IE die-hards to Edge without them even knowing, and that's something web developers everywhere should be thankful for. Fancy whiz-bang codenames be darned.
The "bad image of IE" that Tom refers to is only known by people like you or I who actually read about this stuff, and because we read about this stuff, we already know that Edge is supposed to be a clean break from the terribleness of IE.
Welcome to Thinkerbit v5 - the fastest, cleanest, and most functional version of my little home on the web there's ever been.
If nothing looks different to you, that's because this design has been live since October 2014, although I never announced it. The last post I wrote on Thinkertry.com (the old name and URL of this site) was "It's for a ring" back in March of 2014. From March to October, Thinkertry appeared to be completely dead because I was busy over here at Thinkerbit.com (the new name and URL) continuing to write while learning how to make the site better.
Today, after publishing 139 new links and articles at this new home without a peep, I'd like to formally acknowledge version 5 of this site and give you a tour.
I dropped WordPress and switched to Kirby
This was the primary catalyst of my hiatus from Thinkertry.
Back in March I started researching ways to speed up WordPress without paying a lot for a better web host. I was with Bluehost at the time (which I now urge you to stay away from at all costs, along with GoDaddy, Hostgator, Dreamhost, and the like).
In my research, I found that WordPress is actually pretty bloated in general for simple blogs like mine. Over the years it has become more of a website platform than a blogging platform - adaptable enough for everyone yet perfect for nobody.
I started thinking about moving away from WordPress, and after some research I eventually discovered Kirby CMS: a simple, flat-file content management system that doesn't use a complicated database like WordPress. I'll save most of the geeky details for another time, but these were the primary benefits of switching:
My site would be significantly faster in every way
I could move to a cheaper web host and still get great speeds
I could download the entirety of my site in just a few minutes - no database stuff involved
I could upload my site anywhere and it would just start working
I could insert custom styles and scripts for certain pages and posts (kinda like The Verge, or Medium)
I would have complete and total control of everything (which I really like)
I would be able to edit the .txt files that make up my blog posts from any of my devices
I would be able to write in Markdown natively
I would have a shiny new project to work on
Of course, there were/are some downsides and risks to Kirby as well:
Kirby is more likely to fade away eventually (WordPress is basically immortal)
Plugins for Kirby are more scarce, which means I had to learn PHP and build a lot of functionality myself
It's far less user-friendly than WordPress to set up, and to write for (using Markdown & text editors instead of a WYSIWYG editor)
Posts and content typically need to be published via an FTP client (which can be worked around with Dropbox and some know-how)
For a text-centric site like mine, the advantages still outweighed the downsides of making the switch, so I chose Kirby. The alternative, Statamic, was a little too corporate for me, and I still have a lot of trouble remembering and pronouncing its name. Kirby is friendly, and the whole darn thing can be contained in a Finder window. How cool is that?
After doing a brief experiment with Kirby and seeing how awesome it was, I began to revamp the entirety of Thinkertry.
That last reason may seem a bit silly, but it's important. TinkerTry's growth has far exceeded the expectations of my Dad and I, which is awesome, but it also reveals the lack of planning involved when choosing "Thinkertry" as the name of my own site. After my family spent a good couple of hours helping my Dad think of TinkerTry.com, the last thing I felt like doing was thinking hard about my own experimental blog's name. I added a letter and called it a day.
The funny side effect of that decision was this:
I'm not obsessed with the way this site appears in Google results, but the ghost of that decision has haunted me around the web for many years. Now that's gone.
Why the "bit"?
I've been reading stuff on the web voraciously for nearly a decade, and in that time I've grown to hate being tricked into reading crappy fluff pieces that waste my time. Articles with honest titles, concise language, clear organization and respect for their reader are what I strive to write and enjoy sharing.
This site is a place for me to practice extracting, connecting and presenting the core, central bits of the many things that I read online and the new ideas or techniques I'd like to contribute. That's been true since the beginning when I defined Thinkertry as the "bits and pieces of thought", but I think the new name reflects that idea better.
Of course, I won't always take myself so seriously here. This is also a playground where I can flex my web development and design muscles, laugh at dumb things that I think are funny, and share things that are a bit more personal. This is a blog, after all, and I plan on using it more than I do any social network.
(I also recently learned that "bits" in Information Processing Theory are what Human Factors professionals are often hired to reduce. The more bits there are to think about, the more time it takes for people to process that information, which is the basis of Hick's law. Kind of a nice tie-in with the whole KISS philosophy I'm shooting for.)
Changing the name also meant that...
I changed the URL structure (hopefully once and for all)
Over the years, a few of my articles and reviews have been linked to elsewhere on the web. When this started happening, I felt that I had an obligation to keep the URLs they were posting alive for as long as possible.
URLs may seem insignificant, but if you care about the longevity of what you publish online they're something you need to think about. Ideally, you should figure out your site's URL structure before publishing a single article. I didn't, and eventually I learned that I was paying the price.
Switching to Kirby may have been the catalyst of my hiatus, but this URL problem was the reason I couldn't turn back to Thinkertry. I didn't want to continue publishing things at URLs that wouldn't be around in the future, and would contribute to even more link rot on the web.
It took many weeks of thinking in the shower and a bunch of comparative analysis, but I think I figured out the best structure for this site and I'm sticking with it. I have enough research material and analysis to write an entirely separate article eventually, so I'll stick to the gist for now.
The homepage displays articles of every type (links, articles & reviews), and each type has its own dedicated page. There are no longer any dates in my URLs (for many reasons) and the single subdirectory gives some indication of what's located there - a review, a link, a longer article, and maybe tweets eventually.
When I switched the site to Kirby and changed the structure for good, I could finally get back to writing without worrying about people linking to my stuff. That's what I did (silently) in October.
Once the new name and structure were settled, I started focusing on a variety of annoyances I had with the site.
I improved a lot of little details
A better, filterable archive
The new Archive page is significantly cleaner, simpler, and more powerful than it used to be.
The old design focused on the time that I published things. You could click a year and see everything I published that year, or a month and see everything I published that month.
But honestly, who really cares? I never used those links, and if I didn't find them useful then nobody else did either.
The old design also de-emphasized my articles and reviews by lumping them together with link posts. It was impossible to find my long-form stuff, which I typically spend a lot more time on and is (hopefully) more valuable in the long-term.
The new Archive page fixes both of these issues. It de-emphasizes the date of each item and condenses the titles into an easily-skimmable list. Those titles can be filtered by type with one click, and that's it.
You'll also notice that the "filters" are actually just links to each post type's directory. For example, thinkerbit.com/articles shows all the articles I've ever written in an archive-style format.
Eventually I may add a search field to these archive pages, but Cmd/Ctrl+F has been working pretty dandily so far.
A 'Dark Mode'
Reading websites at night is a pain in the eyes. Black text on a white background may be optimal for daytime readability, but it can cause a lot of eye strain when reading in dark environments and screw up your sleepy brain. Trust me, I know.
Dark-themed mobile apps like Instapaper, Tweetbot, Reeder and Press have spoiled me the point at which I hardly ever visit websites directly anymore (including my own), but I realize that not everyone has discovered the luxury of RSS. For those people, I've added a Dark Mode to the site.
Clicking or tapping "Dark Mode" on the top right changes the style of Thinkerbit site-wide. Whites turn to dark greys, reds turns to DSLR-inspired oranges, images become darker, and a super tiny on/off cookie is used to keep track of your preference across pages. The cookie deletes itself immediately when you close your browser - there's no tracking involved at all.
Dark Mode isn't perfect yet - you may still see an annoying flash of white light when switching pages - but I think it's a unique improvement. I'll keep working on it.
A design archive
The first design of Thinkertry.com mysteriously went missing back in 2011 and hasn't responded to my emails, but versions 2, 3 and 4 are all available in the archive for historical purposes. I'm probably the only one who cares, but that's alright. Looking back at them keeps me humble.
At the time of publishing it looks like v4 is having some video embed issues, but whatever, I'll fix it later.
Crap-free YouTube embeds
Out-of-the-box YouTube embeds are pretty terrible. They don't resize well to small screens, they put ugly suggested video thumbnails at the end of each video, and because they're an iframe (like an embedded version of a YouTube webpage) they force your web browser to download a whole bunch of unnecessary files before you even click play. 12 files, to be exact, and I only need to send you 3.
I fixed that.
Now, only the preview image of a video is downloaded from YouTube's server, and I add a YouTube-style play button over it with a clean hover effect. Once you click play, the entire iframe is downloaded and the video starts playing. On iOS a second tap is required to initiate the iframe once it's loaded, but I think that inconvenience is worth the speed improvement on mobile.
Here's an example.
A reader-friendly design
When you Google for something and click on a link, you care about the following things:
1) Finding and reading the content you're looking for
2) Making sure the site you're reading is legit and can be trusted
My priorities as a writer and a website-maker are therefore:
1) Ensuring that the article content is apparent, loads fast, and looks good on every device
2) Ensuring that the site looks legit, trustworthy and authoritative
So far I think I've done the first bullet pretty well. The second bullet I still need to work on, and I'm thinking the homepage (typically a user's second click) will be the place to do that. The large title area of v4 was a bit more than necessary, but I may do something similar.
I dropped Bluehost in favor of DigitalOcean
Bluehost, GoDaddy, Dreamhost and others are a complete ripoff for 80% of the people who need to host websites - myself included. I didn't know that back when I made TinkerTry.com for my Dad and plopped this tiny site onto his server for free, but I sure as heck know it now.
Up until about a month ago this new site was hosted on NearlyFreeSpeech for (theoretically) about a dollar per month. For comparison, Bluehost was costing my Dad and I $10/month.
Something so cheap may sound like a scam, but it wasn't - I only paid for the bandwidth and storage I used, and because NearlyFreeSpeech doesn't offer any free customer support (figuring that the only people who use their servers are geeky people who know what they're doing) they keep costs extremely low for everyone. Outages were minimal, speeds were way more consistent, and after testing my site with them for over half a year I was pretty pleased with them overall.
Except, I wasn't quite spending that little per month. This is what I spent in 2014:
And 2015 so far:
This data is a bit muddled because I added two additional (small) sites and domains during the time I was with them, but the gradual price creep is pretty evident, and February 2015's crazy $12.20 bandwidth charge was the result of a sudden spike in spam traffic from Russia. I wasn't too fond of paying for crazy Russian servers to jam my site with fake requests and downloading my images repeatedly, so I had to leave.
I'm glad I did. DigitalOcean has been awesome so far for a flat $35/yr. It's still very geeky - I had to learn my way around a Linux terminal to set it up - but the speeds are ridiculously good compared to anything else I've used.
I'm currently seeing under 200ms to load a page over ethernet. Adding Google Analytics' tracking code makes the actual load time a bit longer, but the perceived load time is still awesome. Even simulating a 2G (250 Kbps) connection on a phone, navigating between pages is still faster than most sites - especially text-only posts or pages with just a single YouTube embed.
The sheer speed of Thinkerbit v5 is probably the thing I like most about it. It wouldn't have been possible with WordPress without paying some serious bucks, and I'm doing it for $5/mo.
I still need to fix things
I've learned a lot about web design and development in the last year or so. Way more than I knew before switching to Kirby, and 100X more than I did when I made Thinkertry v1 back in 2011.
Although I'm finally starting to feel good about this site's performance and underlying code, there are still some niggling things that are annoying enough to pop into my head from time to time, but not important enough to delay this launch post before fixing.
For example, the front page currently has the full text of my most recent posts. That's great, but when I post a review with 20+ images that can significantly increase the front page's weight. I need to make certain posts, like reviews, truncated to prevent that.
When you're on the About, Archive or Follow pages, their respective menu item at the top should indicate that you're already on that page. Right now it's unclear.
The "MENU" button that appears on small screens is very rough right now. I think it's better than many of the junky, over-animated hamburger menus on other sites, but it's still not what I'd like.
Having a search field in the archive page might be useful, but I want to get it right before implementing it. I'd like to see if Kirby's search is powerful enough to replace Google.
The source code and CSS/JS files of this site have some cruft and old code that needs to be removed and optimized eventually. That will probably be true for the next decade, so I've stopped worrying about it.
Clicking or tapping on images brings you to their native file location instead of zooming in or displaying them in a nicer format. This is a tricky problem to fix across platforms, but I'm close.
Right now I think the design of the site lacks personality. Particularly the front page. Nothing visually distinguishes links vs articles vs reviews, and I think I can present things a little more creatively without adding unnecessary cruft.
At the moment, embedded videos don't appear in Thinkerbit's RSS feed because of the crap-free YouTube trick I implemented. This is my biggest annoyance right now, and the first thing I'll try to fix.
This won't affect most people, but it does mean that image previews on Facebook or Google+ may get screwed up down the line when I try to remove the /content/ directory from these URLs. This is one of the last Kirby-related issues I'm struggling with.
I'm still reading, learning, and improving
During Thinkertry's first year I learned that attempting to cover the consumer tech news cycle is pretty much suicide. Last year a promising site called Tab Dump did a pretty great job at it, but shortly before burning out, Stefan Constantine was hired away.
... I found myself simply not caring about what was going on in the tech space. That's not to say I don't find technology absolutely fascinating, I do, it's the never ending coverage of every trivial detail that turns me off. I enjoy reading a 10,000 word AnandTech article as much as the next soldering iron and multimeter packing geek, but the needless drama around Bash, iOS 8.0.1, and whatever new quote that leaves Eric Schmidt's mouth is frankly tiring and mentally exhausting.
The day that Thinkertry became a thing, the way I read articles from Engadget and other sources completely changed. Every article that passed through my RSS reader was something that I could potentially write about, and the rate at which I consumed articles was dramatically slowed down.
After deleting more 1,000+ word draft posts than I care to admit and repeatedly slashing through my Instapaper queue over the years, I like to think that I'm better at spotting what's worthwhile today than I was back when I started writing. These days the articles that get stuck in my feed reader are more often related to design or web development than consumer technology, simply because I'm less able to distinguish between the noise and signal related to those topics. Eventually I'll get better at filtering those too, and I think that's one of the hidden benefits of having a personal blog like this one - I'm paying more attention and learning far more than I used to as a casual consumer of information.
Thinkerbit is frequently a personal exercise in web design and web programming, sure, but its primary purpose is to help me write better. I'd like to eventually optimize, or design, my own unique form of writing by talking about a variety of topics here, and I think I need to focus on that a bit more now that the infrastructure is mostly figured out.
I imagine Apple did some strong-arming to make this happen - Best Buy sells Apple products, after all - but with one CurrentC partner switching teams, I wouldn't be surprised to see CVS, Rite Aid, Walmart and Sears cave this year as well. Walmart will probably be the final holdout, since they're spearheading the whole thing and have the most to lose from CurrentC's failure.
OS X Yosemite has been out for many months now, but I hadn't tried to use a couple of its new 'Continuity' features - namely, Handoff and Instant Hotspot - until recently.
I say "tried to" because I couldn't actually use either of those features on my mid-2011 Mac mini. I turned on my iPhone's bluetooth (which I usually keep off), connected it to my Mac, and because I knew my Mac had Bluetooth LE-compatible hardware, I assumed it would work.
Notice two incompatible models in particular; the mid-2011 Mac mini and MacBook Air. With the help of this simple utility, both of these Macs can use all of Continuity's features without any additional dongles because they already have the requisite hardware. It only takes a flip of a yes/no switch in the system's configuration to allow the magic to happen.
So why not make those two models compatible? You could speculate that there are some weird bugs that crop up on the 2011 hardware (which I would re-speculate isn't the case, or wouldn't be insurmountable), but my best guess is that it:
A) wouldn't make for as clean a cut (it's easier to say "every Mac since 2012 is compatible") and B) Apple decided to use Continuity as an incentive for users to upgrade from their older 3+ year-old hardware
I don't like either of these reasons. Maybe my Windows upbringing has warped my perception of how long hardware should last, but three years seems too early to start withholding compatible features and nudging people to think about upgrading - especially when those users already have the necessary hardware built into their system.
I know that Apple and Microsoft are very different companies with very different approaches, but I still can't help from being a bit surprised by the chart above. 3 years (or two hardware iterations, whichever comes first) seems to be when Apple will start to subtly encourage Mac users to upgrade to the latest hardware, technical compatibility be darned.
I honestly don't even care enough about Handoff to install the utility - I just find it helpful to know that Apple thinks my Mac is ready to be replaced.