picomac

Earlier this week, Stephen Hackett revealed how he finds all those animated GIFs so quickly: with an Alfred workflow that searches Giphy.

.@ismh help i need a bespoke gif for "it's 4:50 PM on Monday and i just opened OmniFocus for the first time"

https://t.co/1YQfb5tyJ5 https://t.co/59Gh64CmZ5

The workflow itself is a shell script which takes input, does some text replacement, and opens a URL. It's just two lines, but as a longtime LaunchBar user it seemed…complicated.

If I need to automate a repeated searching task, I set up a custom search template in LaunchBar. All that's necessary is to run a search on the site in question and replace the part of the URL corresponding to your search terms with an asterisk. Cribbing from Stephen's script, the search terms come after giphy.com/tags/. The script also does some preprocessing, taking out spaces in the search terms and replacing them with hyphens. I thought this might make it impossible to create a corresponding LaunchBar template — what if Giphy really chokes on spaces in URLs? I headed over to Chrome and manually typed in a test URL. The browser handled percent-encoding the spaces and Giphy automagically converted them to hyphens! It also redirected the base of the URL from giphy.com/tags/ to giphy.com/search, adding a little time to the page load — I adjusted my template accordingly.

One benefit of the slightly more complex Alfred workflow is that it specifies a keyword. Stephen set it to 'gif', sensibly enough. But I named my LaunchBar search template 'Giphy'. Since LaunchBar finds items by letters in their names, typing G-I-F won't find it! There are two possible solutions. I could put 'gif' in the title — something like "Giphy gif search" — but that's inelegant. The other option is to assign a custom abbreviation. This is done from the main LaunchBar interface, not the index setup window. I typed 'giphy' to find my new template and then did the one thing I never think to do in LaunchBar: reach for the mouse. Clicking on the name brings up a contextual menu where I can assign the logical abbreviation.

In the end, the setup in LaunchBar, Alfred, or any other launcher app is about the same complexity; it's a matter of choosing the one that works the best for you. There are subtle differences: Alfred interprets the input "gif [search terms]" at once, while LaunchBar changes from search to text entry mode after hitting space. Alfred's documentation gives "maps" as an example: you can choose Apple or Google Maps after typing your search terms. Perhaps if you're a true GIF connoisseur you need multiple search engines at your fingertips. All it takes is a little setup in the appropriate Mac utility.

The latest major update to Hazel, the file and folder automation tool from Noodlesoft, is here and it's just as invaluable as its predecessor. If you're not familiar with Hazel, it adds abilities similar to email rules and filters — and then some — to the Mac filesystem. On the surface, Hazel 4 looks nearly identical to version 3, but it packs a lot of new power and has some welcome, if hidden, interface enhancements.

A major frustration of Hazel has been figuring out whether you're wielding its immense power properly. There are so many file attributes that can be used as rule criteria, and the number of combinations multiply quickly. Criteria can be combined with Boolean logic (all of these or any of these) and those combinations can be nested; that's just within a rule, and the precedence order of rules matters too!

For a while, Hazel's main window has had a preview mode that shows which rules will apply, but it couldn't be accessed while editing a rule. Now there's in-editor preview, although it's limited to evaluating a single file at a time. No more guess-and-check, accidentally destroying a rule's effectiveness, or having to duplicate a rule to have a "safe" sandbox environment for editing. Hazel rules are set-it-and-forget-it by their nature, and I haven't had any real use cases to create new ones in version 4, but I can tell already how much time rule preview will save.

Another potential time saver in version 4 is rule sync across multiple Macs. Automation is less effective if it needs to be manually set up in every new context. Hazel 4 smartly allows syncing of rules on a per-folder basis. Naturally, if you've got two Macs, they aren't going to have all the same files and folders, but some of the most-frequently automated areas will be on both, such as Downloads, the Desktop, or your Dropbox. This may be more useful for those who have two Macs that both operate in the same context, as I realized that all of my Downloads rules are context-dependent. My home Downloads rules mainly process personal bills, which I never put on my work machine; my work Downloads rules process files that I never access on my home computer.

Even with these enhancements, I feel that setting up rules in Hazel will never be maximally efficient as long as it remains a System Preference Pane. This has long forced the rule editing interface to reside in a sheet and hide subsidiary views in popovers, including impractically small popovers for editing embedded scripts. I long for the day that Hazel becomes its own app, granting access to multiple file preview, rule lists, and rule editing at once — and gaining basic menubar features like proper undo. I've paid my upgrade fee for Hazel 4, and if a standalone app is the future of Hazel 5, I'll be waiting with money in hand.

I'm the kind of digital packrat who preserves everything possible. On the desktop, with each upgrade, storage size gets closer to infinite and storage cost gets closer to zero. Not so on iOS, where storage sizes have slowly crept upward and the relatively flat folder structure limits how much stuff you can squirrel away out of view.

I've maintained data continuity on my iPhones going all the way back to the 3G by doing an encrypted iTunes backup and restore with every new device. I've done a couple large purges, especially when space became tight on my 16GB iPhone 4, but years and years of unused apps have accumulated since. It was well past time to clean house again. I made another full backup (of course) and entered wiggle mode. Just a few minutes later, my phone was 55 apps and over 2GB lighter, and in day-to-day usage I won't miss a single one of them.

Nice and tidy now.

There were a couple trends in what stayed and what went:

My biggest gain (nearly a gigabyte) was from dumping Pages and Numbers. I never attempt to edit documents on a sub-5" screen, and iWork documents can be previewed natively without the apps installed. Contrast the Google Docs and Sheets apps (almost 300MB combined) which are necessary for practical document preview, despite their roots in the browser. Keynote also got a reprieve for its remote features, which I use infrequently but find indispensable.

I consume most of my media on iOS in a couple apps. An app just for my hometown NPR station? Obviated by TuneIn, which I now mostly play on the Amazon Echo. The iTunes Festival app? Probably dead, subsumed into Apple Music and a better experience on the Apple TV anyway. A few big-platform apps made the cut, like NPR One, SoundCloud, and the Bandcamp app, all of which surpass their web counterparts in some way.

Many beautiful weather apps sprung up when forecast.io made their data freely available. Unfortunately, I find that data unreliable where I live, so I only kept first-party Dark Sky, and only for its minimalist widget. I still need a folder for what's left, thanks in part to Weather Underground and Storm's split personality and the undeletable Apple offering.

Games used to be the biggest space hog on iOS, but with many apps nearing 100MB, they're closer to average now. Still, I parted with all but a few favorites. Some I rarely play but view as classics: Tilt to Live, Puzzlejuice, Threes. Some are discontinued, so I have to preserve them: FlightControl and Burning Monkey Puzzle Lab are gone even from the Purchased tab in the App Store. I do look askance at games that wouldn't fit on the Apple TV and eat battery for breakfast: hello recent favorite, INKS. I should make a note to cull them — and, honestly, all of my apps — more frequently.

Complex automation with email can be difficult, especially on iOS. Mail clients and services can provide the ability to create intricate rules, but typically their application is limited to moving, tagging, or otherwise organizing messages within a mail system. Heavy users of automation point to the conspicuous absence of the share sheet in Apple Mail. Fortunately, there are good third party mail alternatives on iOS. One of them, Airmail, offers multiple customizable actions, including access to the share sheet. Getting to the share sheet is really the magic step because it opens up access to action extensions from the Workflow app, making the data truly portable.

Airmail's sharing abilities plus the new URL parameters in OmniFocus 2.14 let me speed up a common work task. As part of my job I review content from freelance authors, and I get my review assignments by email. Each review has several required tasks that I've saved as a template project in OmniFocus. The old way was to copy the template project and manually fill in placeholders: assignment ID, author, and combinations of the two. It was quick, but manual. OmniFocus' new ability to accept the TaskPaper format meant that my project template, including contexts and due dates, can now be represented as plain text, and "plain" text in Workflow can also include variables.

Connecting the dots from there was relatively simple: extract the ID and author from the (generally) predictable email subject line with a regular expression. Pop those into the TaskPaper formatted project template. OmniFocus date math in the template gets parsed when the Workflow runs — these projects are to be completed within a week, so I just set a @due(1w) parameter. (More complex date math could be done with Workflow actions and an additional variable.) The whole assembled project gets copied to the clipboard and run with the new omnifocus:///paste trigger. Total it's four taps from incoming message to filled-out project: action menu in Airmail, Share, Run Workflow in the share sheet, and finally select the workflow itself.

It took about 30 minutes to put together this workflow (including pecking out regular expressions on the iPhone keyboard) and I'll likely run it 30 times before the end of the year, with about a one-minute savings each time. In the process I also learned the new OmniFocus text format and URL scheme, which is almost endlessly flexible. (One caveat is that new projects can't be auto-filed in folders yet, but Omni says that's coming soon.) I should also note that this workflow would be harder to set up on the Mac; it would probably require a mix of two or more scripting languages and a service to get it all working. This is my serious work — my livelihood — and thanks to these three apps, I'll be turning to iOS to do it faster than ever before.

It's May, and that means it's officially WWDC prediction season. While some will be eagerly awaiting revised MacBook Pros or to find out just how Apple is going to capitalize "Mac OS", I'm primarily waiting to see what iOS 10 will look like on the iPad.

I'm one of the people who is personally ruining Apple by not buying a new iPad every year. In fact, I've only ever owned a single iPad, the third generation. I held out for Retina because I planned to use it primarily for reading PDF articles in grad school. At about the time that I was due for an upgrade, I graduated and stopped using it daily, so I simply haven't bothered. In the meantime, a lot has happened with the iPad line, including the introduction of not one but two Pro models.

There's almost no question that I'm in the market for an iPad Pro. Lured again by a Retina screen, last fall I decided not to get a new Mac laptop, instead opting for the 4K iMac. It remains in a fixed location in my home office and is my canonical computer; even as things migrate to the cloud, it's my digital hub. In the past, I sacrificed some portability by choosing a MacBook Pro, a canonical computer that could travel with me. Now I can choose the ultimate in portability, as long as it can act as a window to home.

The greater screen real estate on the large iPad Pro would make a VNC session back to the iMac more practical, but the Baby Pro is more portable. This is where software will determine my next hardware purchase. The 9.7" Pro comfortably fits with iOS, while the 12.9" often looks like it's wearing ill-fitting clothes. I'm not excited about the prospect of spending close to $1000 on an iPad that makes me cringe every time I return to Springboard. OS X scales rather gracefully all the way from 11" to 30" screens, by virtue of its design. The spread of the 4" to 13" range for iOS is comparable, but isn't adequately covered by the OS. It's reasonable to ask whether the introduction of the 9.7" Pro was an admission that iOS just can't go that big. Personally, even though it would cost me a few hundred dollars more, I hope that next month's WWDC announcements prove that's not the case.

I'm on a streak of discovering old technology. This time it's Spotify Connect, thanks in part to the Amazon Echo. I've just signed up for Spotify (and a Premium trial) for the first time, to broaden the music collection that I can access with Alexa. I had avoided Spotify for some time, when it was only accessible via Facebook login, so it's all brand new to me. That includes the premium Connect feature, which lets you choose your audio output device from any instance of the Spotify client.

Connect is a revelation to anyone who has only used AirPlay to get music from point A to point B. The major differences are that Connect is 1) not directional and 2) not constrained to the local network. AirPlay relies exclusively on a directional model, with some devices acting as sources and others acting as receivers. For example, there's no way to beam audio or video from an Apple TV out to a Mac, or from any device into an iPhone. On Connect, all devices are created equal, and unlike AirPlay sessions that can only be initiated from a source device, the source is the cloud. That means that I can switch output from my Echo to the speakers attached to my Mac using my iPhone, something thoroughly impossible with AirPlay.

Choose wisely.

With great flexibility comes great responsibility, especially since Spotify Connect operates over the internet, not local Wi-Fi. Yesterday morning I tried to listen to Spotify on my phone at work, but I didn't realize that playback was set to my Echo at home. Suddenly some pretty gnarly noisepop chiptunes were playing in my living room at home, much to the surprise of the person sitting a few feet away from it. (I stopped it pretty quickly, at least!)

All of my marveling and missteps aside, Connect is rock solid. As I swapped between earbuds on my phone and speakers on my Mac, I barely noticed the crossfade between them; certainly there was no stuttering or buffering. This is a feature I've wanted from Apple since the iPod. In college, I had an iPod dock plugged into my PowerBook G4 in my dorm room. I wanted to be able to listen to music on my walk from campus, arrive home, dock the iPod, and have iTunes pick up right where I left off. Today, with iCloud everywhere, that seamless experience should be part of the Apple ecosystem, and with no 30-pin connectors required. In the meantime, two competitors — Amazon and Spotify — are the closest to granting my wish from 2004.

All the way back in the first week of Picomac, I observed that the Apple leather case on my iPhone was nearing the end of its life. That was over two months ago, and it only continued to deteriorate. With the volume and power buttons almost completely inaccessible, it was time to find a replacement.

Like a good nerd, I went to The Wirecutter for advice. With direct button access being my primary criterion, I was lead to their "ultrathin" pick, the Caudabē Veil. Although they viewed it as a downside, they mentioned that the Veil XT — a slightly different model — has no lip around the front glass. My two usability problems solved in a case that's 0.35mm thick? I had to give it a shot. I ordered the Frost variation to match my silver iPhone 6.

When the Veil arrived, I took it out of its box and initially mistook the case itself for a piece of packing material. It seemed so flimsy as to be completely useless. In fact, the Veil flips the usual approach for iPhone cases on its head: instead of being the structure that reinforces the naked robotic core of the phone, the Veil relies on the sturdiness of the iPhone itself. It wouldn't work if it left even the hint of a gap between metal and plastic. But Caudabē delivers on their claim that it's "engineered with precision": there are six separate holes in the Veil for the speaker grille and they align perfectly. The case nearly vanishes when on the phone, simply substituting plastic for the iPhone's ultra-slick aluminum back.

Another advantage of the Veil's vanishing act is that it shows off the design of the iPhone. When I bought my phone, I slapped it into the Apple leather case before I even left the store. I didn't so much forget about the iPhone 6's design as never learn it. It seems foolish now to say "Wow, those antenna lines look great!" but I've never had the chance to appreciate them. The minimality of the Veil also means no tacky branding; Caudabē only embosses their company name, white on white, right under the regulatory information; it blends in with the other inoffensive fine print.

Beyond looks, ditching the larger case makes me feel like I have a brand new phone. I've never used an iPhone this thin; it slides in and out of my pocket far more easily; and, yes, the buttons actually work now. And while I can't drop it onto concrete, I can slide it around my desk or carefully place it in a bag without the fears I had for naked phones. The Veil XT is perfect for me, someone who never wanted a case in the first place.

Although Wall Street is unimpressed, I continue to be amazed at the sheer number of iPhones produced by Apple. (After all, I was a Mac user when Apple was doomed — properly doomed, that is.) 51.2 million is in the range of numbers that are more or less incomprehensible in everyday terms. I remember when I was a kid I pored over a picture book that tried to explain the concept of a million anything, trying to really grasp it.

It's impossible to visualize 50 million phones once they've been dispersed to pockets and purses all around the globe. But what if they could all be brought together in one place for a grand Q2 2016 iPhone reunion? Just how much sheer volume would they take up? Let's assume that all those phones, on average, were iPhone 6Ses; they weren't of course, but perhaps the 6S Pluses canceled out the 5Ses still being sold. Let's also assume that whoever is in charge of this magical summoning of the iPhones is a real obsessive neat freak and stacks them in a perfect grid arrangement.

Now, ironically, we can turn to Google to do some handy unit conversions for us. The total volume of the assembled iPhones is some 3.3 million liters — another meaningless number, without context. We can compare it to the volume of another impressive piece of Apple construction: the glass cube that sits atop their flagship store on 5th Avenue in New York City. The cube measures 30 feet on each side. If we tried to fit all of the Q2 iPhones in there, we'd have many left over. In fact, if they were stacked with a 30 by 30 foot base, the tower would reach 132 feet (40 meters) tall — not exactly a skyscraper, but a good size office building of solid phones. It would weigh over 8000 tons (7300 metric tons, or 7.3 gigagrams — aren't SI units fun?).

It's a simply remarkable quantity. It also reinforces the urgency that Apple is placing on recycling — all those devices have a shelf life of about two years, and if they were all destined for the landfill they would fill a lot of land indeed. But until they meet their maker, those magical pocket computers are in the hands of 50 million people. Just try to wrap your mind around it.

It happened. I'm the latest in a long line of Apple enthusiasts to buy into the Amazon Echo and give Siri some competition in my life. Just one day in, it's far too early for me to give a full review, but I can say a little about the setup process: it was flawless.

Until a couple months ago, I never owned any Amazon hardware; now I have two devices, the Echo and a Kindle Paperwhite. For both, Amazon has tried to shorten setup as much as possible. With the Kindle, there's even the option to pre-load your Amazon account credentials onto the device at the warehouse. That raised minor security concerns for me, but I was so intrigued by the potential convenience that I opted for it. As promised, it arrived at my doorstep two days later, and I didn't have to transcribe any information from 1Password to a non-capacitive touchscreen keyboard. What a contrast to the awful initial password entry on the Apple TV's letter-line.

The Echo was nearly as simple, although it doesn't come preconfigured. Its in-box documentation lists just two steps: plug it in and download the companion app. You're not quite done at this point: the Echo needs to get online. So how do you get your Wi-Fi credentials into the keyboardless cylinder? Not by clunkily dictating SSID and password one letter at a time, as tvOS 9.2 might have you do, but by setting up an ad hoc network with your phone, which acts as a suitable input device.

These experiences shouldn't be remarkable, but compared to current Apple hardware, they seem revolutionary. Apple prided itself on simple setup in the past: "There's no step three! There's no step three!" While that was an oversimplification for the original iMac, it was a slight one, just as it is for the Echo. For today's iPhones, not only is there a step three, but forty more to follow.

no one raindrop believes they are to blame for the flood. pic.twitter.com/cXQsP0RngD

There's no doubt that the iPhone is a more complex device than the Kindle, but I would love my next iPhone to set up just as easily. Especially for enrollees in the new iPhone Upgrade Program, why shouldn't it be a two-step process? A Genius slides your new phone across the Bar, you verify your iCloud password (username pre-filled) and set up Touch ID. And forget re-downloading all your apps and settings over the store Wi-Fi; that could be done for you in advance. That would seem like magic to some users, and at least great service to the rest. That doesn't just drive "customer sat" — it gets people to go beyond buying devices to welcoming them home.

Welcome home, Alexa. So fun, and by far the best speaker in our house. Highly recommend (after 1 hour, anyway). pic.twitter.com/6uqOPCphsa

I said yesterday that the main faults of Airmail for iOS are actually the faults of email itself. Despite consistent development and improvement, I can't totally say the same of Airmail for Mac. Interestingly, though, I find that two of Airmail for Mac's major shortcomings — search and threading — illuminate the larger problems with email as a whole. (I almost said "email as a communication medium", but not truly being used for communication is one of the problems.)

Everyone with an email account has at one point or another struggled with email organization. The "modern" approach to the problem is to punt, only performing coarse organization (sometimes as basic as archived vs. not archived) and trust search to handle the rest. This is very much The Gmail Way, but given its Gmail roots, Airmail's search is surprisingly lacking. Want to search the current label or folder? You'll have to type its name and hit return to generate a search token, and then begin your search. Found the message, but not the relevant term in several paragraphs of text? Type your keywords again as you enter battle against Airmail's recalcitrant intra-message search.

Often my email searches target a particular piece of information: a flight confirmation number, a receipt, the time and location of an event. Tools like Google's Inbox app attempt to make these, rather than messy messages, the units of email. Yet a proper solution isn't to extract meaning out of chaos, but to begin with structured data. Convincing corporations who use email as a one-way street to adopt such a model is a big ask, but one that the likes of Google, Facebook, or even Apple could take on.

Back to Airmail, my greatest frustration with it is woefully poor threading. Like most modern mail clients, it tries to collapse earlier quoted messages, but unlike even Apple Mail, it fails more often than not. The problem here is that, especially in a work environment like mine, people treat email like extended chat. It would be pathological to repeat every previous message in a chat, but it's the default for email. My coworkers are wedded to Exchange email, and I'm still more likely to convince them to move to a chat-centric solution like Slack than to alter their quoting behavior.

Nobody can singlehandedly fight the entropic decline of email from a useful communication platform into a point-to-point transfer of unwieldy blobs of text. Bloop, the makers of Airmail, are doing their part to make those blobs easier on the eyes and more versatile. They are striving to build a better band-aid, while modern email continues to be death by a thousand cuts.