Riccardo Mori

The next mini

October 24, 2017

Next mini

Prompted by a MacRumors reader on the future of the Mac mini, Tim Cook replied that the mini will be “an important part of the company’s product lineup in the future.” Remember the mini? It was last updated in October 2014, which in tech years is a long time ago. That last refresh was a bit controversial, too, because the 2014 Mac mini was a step back in terms of upgradability and versatility when compared with its 2012 predecessor (take a look at this comparison review published by Macworld UK three years ago to have an idea of the differences).

It’s a true pity that Apple has been neglecting this Mac. I still think there’s a market for a compact, affordable Mac such as the mini, though I presume the dream of Apple’s current executives is that people would all choose an iPad over a Mac mini in that price range. I think Bob Burrough (who previously worked at Apple) has best articulated Apple’s lack of interest in the mini when he tweeted: In my opinion, the Mac mini is supremely damning, because the minimum effort is just updating the CPU, RAM, storage each year. They didn’t. Like how would they answer “Why didn’t you update the Mac mini?” “We’re undergoing a difficult and sophisticated redesign.” I don’t think so.

I don’t think so, either. Still, Cook says they’re not leaving the Mac mini behind. What’s next for this machine, then? I’ve been thinking about a few possible scenarios:

The first scenario is the laziest: the mini basically maintains the same design and simply receives a speed bump with hopefully up-to-date, decent specifications. This could be ‘good enough’ from a pragmatic standpoint, but it would certainly create another round of criticism towards Apple (e.g., We waited more than three years to get a new mini that has the same design as the 2010 model, come on!).
The mini gets worse. By worse, I’m thinking of a physically smaller machine, something resembling an Apple TV, available in two configurations — ‘underpowered’ and ‘okay’ — with soldered RAM, SSD, USB‑C ports (2 or 3), and that’s it. ‘Worse’ is obviously subjective here. It would be for me, because I prefer expandable, versatile machines; but perhaps such a Mac might be a good option for people who just want something small and powerful enough on their desks.
The mini gets better. I’m thinking of a sort of ‘poor man’s Mac Pro’, a rather expandable yet compact desktop Mac, something in the spirit of the Mac mini from 2011–2012 (or even of its glorious predecessor the Power Mac G4 Cube); a machine that would be cheaper than an iMac, but not much less powerful. It could be the choice for those who don’t want a big all-in-one computer, or have already a good standalone display. And when I say ‘poor man’s Mac Pro’, I mean it from a design standpoint, too. Just like the upcoming iMac Pro shares the same design as the iMac, the next Mac mini could have a similar design as the next Mac Pro, but in a more compact package, and with less power/expandability to avoid cannibalising Mac Pro sales.

I’m not sure which of these three options is the most likely. When it comes to the Mac, Apple has been getting increasingly inscrutable in recent years. The company so far seems particularly averse to expandable, serviceable machines, so scenario №2 could be what eventually materialises. I might even be okay with a locked-down Mac mini equipped with immutable internals, but in such case I think the smart move would be to make this new mini a truly affordable Mac, something like the iPad 5 in the iPad product line. An entry-level model at $329 and a more powerful model at $429 — I believe they could be appealing to a lot of people, especially non-pro users with a limited budget.

Of course, I’d love if something like scenario №3 could come true. Imagine a Mac Pro Mini that is a bit narrower and taller than the current mini, with a hot swappable drive bay, easy access to RAM slots (and RAM expandable to 32 GB), plus one PCI expansion slot, just to give it a bit of expandability. This kind of mini would command slightly higher prices than scenario №2 above, but I’d say that keeping the current price tags ($499, $699, and $999, with the $499 model offered without the expansion slot, for example) would be quite fair.

Who knows, maybe Apple will end up surprising everybody by introducing some unexpected feature or design. Maybe the next mini will be the first Mac sporting an ARM chip. Maybe it’ll be cheap and it will come in colours. Maybe it’ll have the array of sensors and cameras of the iPhone X and come with FaceID technology. But really, the Mac mini doesn’t necessarily have to be a playground for innovation. The basic formula is still a winner — a powerful-enough, general-purpose, truly affordable desktop Mac. It’s just waiting to be refreshed to be more up-to-date. There’s really no wheel to reinvent here, Apple.

The MacBook keyboard: I can’t say I’m surprised

October 21, 2017

by Riccardo Mori

Handpicked

Both Casey Johnston and Stephen Hackett have detailed the problems they’ve faced with their MacBook Pro’s keyboard. They’re just two prominent examples, but they’re far from being isolated cases.

It turns out that my concerns about the post-2015 MacBook and MacBook Pro keyboards — which I voiced since first trying out a 2015 retina 12-inch MacBook — were not unfounded. When that MacBook came out I was excited, and I gave it a serious thought when deciding which Mac to purchase next. But after trying it out in person, the keyboard disappointed me right away:

[T]he keyboard is really the deal-breaker here. It’s maddening. What I kept thinking was This should be the perfect computer for a writer and at the same time I was thinking I just can’t write on this MacBook all day.

In June I argued that, thanks to that keyboard, the 12-inch MacBook couldn’t be really considered a laptop for writers. Later in 2015, I added:

In the design of a keyboard, in my opinion, function and comfort should always trump æsthetics. Flattened keyboards might look cool, but may not be suitable for long writing sessions. Short key travel might reduce stress in the fingers, but in my terrible experience with the 12-inch retina MacBook’s keyboard, it also leads to striking the keys with a bit more force, which in turn is painful for your fingertips.

In early 2016, a friend of mine told me that she had to bring her 6‑month old 12-inch MacBook to an Apple Store because the ‘V’ key had stopped registering, and the spacebar was stuck. I wrote her that this was troubling, but that somehow I wasn’t surprised. Later, when she asked me for advice (“Do you think I should try selling the MacBook after they fix the keyboard? I’m bummed, but I also love it for its lightness…”), I urged her to buy AppleCare if she wanted to keep it, because I feared the problem could return in the future.

It’s October 2017. She still has that Early 2015 MacBook. She had the keyboard replaced three times. She’s grateful to have followed my advice about getting AppleCare.

I’m always wary of treating a few anecdotes as valid statistical data, but the fact is that in the last two years at least a dozen people I know, friends and acquaintances, have purchased MacBooks and MacBook Pros with that horrendously thin keyboard, and many (many more than I expected, frankly) have had some issues with it. Another friend with a 12-inch MacBook had the keyboard replaced once. An acquaintance with a 2016 13-inch MacBook Pro had to go back to the Apple Store three weeks after purchase because the Enter key had become unresponsive.

Another fellow writer got his 15-inch MacBook Pro with Touch Bar in February 2017; like me, he usually engages in long typing sessions on a daily basis, and the MacBook Pro’s keyboard never really impressed him. But he needed a portable machine, and a display that was both retina and bigger than 13 inches. He considered buying the mid-2015 model with the ‘better’ keyboard, but didn’t want to invest money in a Mac that had 2‑year old technology in it. After having to replace the keyboard twice, in May and in late September, he told me he’s seriously considering selling the MacBook Pro and getting an older one. Decades as a Mac user, and I’d never met other Mac users willing to take a step backwards this way.

Other three people I know had to bring their MacBook Pros (both 2016 and 2017 models) to have the keyboard replaced once. In one case, the whole keyboard had become unresponsive, and at least the Apple Store genius didn’t blame ‘a piece of dust’ for it, saying instead that it was ‘probably a defective connector’.

Long story short, of these twelve friends and acquaintances that got in touch with me regarding this particular matter, seven had serious keyboard issues, and the remaining five told me that their MacBook/MacBook Pro keyboard feels different now than when they purchased their Macs: some mention uneven feedback when typing, others say that certain keys — despite still working and being registered after pressed — have kind of lost what little clickiness they had at the beginning.

Casey Johnston observes:

The problem with dead keys is that, unless you can stop what you’re doing mid-paper or report or email or game and have a physical tiff with your computer, the temptation to just slam a little harder on those delicate keys to get the N or B or period you need until you reach a stopping place is high. But there is no logical at-home remedy for the consumer; when one key on a butterfly switched-keyboard becomes nonfunctional, unless you can dislodge whatever dust or crumb is messing it up without being able to physically access it, the keyboard is effectively broken. If you remove the key to try and clean under it, you stand a high chance of breaking it permanently, but if you leave it there and continue to have to pound the key to type one measly letter, you also might break it permanently. A single piece of dust can literally fuck you over.

And the reason, in case you didn’t pay attention, is that:

If Apple decides to replace the keyboard, it sends out the computer to replace the entire top case; there is no such thing as replacing an individual key or just the keyboard. On a Macbook Pro, the top case retails for $700, but the computers haven’t been around long enough for anyone to be out of warranty yet. In regular MacBooks, which were first available in the spring of 2015, Apple has quoted as much $330 to replace a top case out of warranty. The path from “a piece of dust” to “$700 repair” is terrifyingly short.

This is bad design, simple as that. The butterfly switch mechanism is a design solution dictated by compromise. The compromise here is Apple painting themselves in a corner after making laptops that are thinner and thinner, and thinner. Two years ago it took me three days of intensive testing to understand where this was going — forgive this little bout of immodesty, for once — and if I remember correctly, Apple has machines in place to conduct this kind of stress test. Perhaps they simply concluded that the keyboards were robust enough to withstand all typing styles; but really, nobody had a clue that by introducing dirt and dust in the equation, things would go awry sooner rather than later?

To conclude, another thing concerns me — that Apple’s way of fixing the problem (they created) in next MacBook iterations is to introduce an even lower-travel keyboard with Force Touch-like sensors. As MacRumors noted two years ago, Apple was granted a patent for it, so it’s a plausible path. Who knows — maybe it’ll end up being a better keyboard than the current one; maybe it’ll break much less frequently. Yet I’m guessing that, should something go wrong, it will still be a matter of replacing the whole top case of the computer. As for the typing experience, I doubt it’ll be as pleasant as a keyboard with physical switches. This is all hypothetical of course, but somehow I’m not convinced Apple is going to address this problem by making thicker laptops with more thoughtfully designed keyboards.

Website refresh

October 16, 2017

by Riccardo Mori

Briefly

A few months back, when WordPress launched version 4.8, I updated right away rather impulsively. Of course I always keep a backup copy of my materials, but it’s always a bit of a shock when things go unexpectedly, spectacularly wrong on a generally stable platform that never gave me a problem since perhaps version 2. WordPress acknowledged that new code and features underlying version 4.8 may break some sites, and indeed, after the update, I was presented with a blank page. Even attempting to log into the admin panel was futile. Blank pages, blank pages everywhere.

After a long, painful troubleshooting session where I tried every suggestion outlined in the support forum, I deduced that the problem had to be related to the WordPress theme I was using. After another long session — mainly spent watching paint dry as folders and hundreds of files were re-uploaded on my server via FTP — I finally reverted to version 4.7.x and the site was back and working again.

The march of software updates never stops, though, and I realised the site with that theme was living on borrowed time. Today, WordPress is at version 4.8.2, and I noticed in the admin panel the news that there’s already a version 4.9 Beta 2 available. My concern is security updates: WordPress may still release security updates for version 4.7.x while 4.8 is around, but perhaps when 4.9 launches, 4.7 will be considered out of date. So, while I theoretically could check every PHP file in the problematic theme and try to fix things through a trial-and-error method, the kind of trial-and-error method involved here would have been absurdly time-consuming.

This is why I chose to look for a more modern and supported theme and (begrudgingly) opted for a website design refresh.

In reorganising some of the site elements, I tried to follow an important statistic: where people click when they visit my site. I may not care about the number of visitors I get on a daily basis, but what they look at and what they click once they’re here, that’s an intriguing indicator. And what I have noticed is that visitors are rarely interested in whatever is located outside the boundaries of the article they’re reading. The previous design featured a main menu on top of the page, a few widgets on the right sidebar, and an elegant, yet somewhat cluttered footer. By looking at the statistics for the past few months, I saw that the footer elements got less than five clicks against hundreds of visits. It was time to simplify things a bit.

Some widgets are gone now. Some of the information they displayed has been discarded, while other things have moved elsewhere. There are two new pages in the main menu on top: Archives lets you navigate the site’s archives with more obvious fine-grained options. You can directly access the last 30 articles, or browse the contents by month, category, tag. I think it’s more useful and friendlier than a simple pull-down menu on the sidebar. Contact & RSS Feeds is another short, self-explanatory page. I think that putting contact information and the links to the various feeds in a standalone page was better than cramming the same information in two footer widgets, so there you have it.

Other advantages brought by the new theme:

The ability to post bigger images. Don’t get me wrong, I’m not a fan of those blogs where you always get this huge intro photo no matter what the article is about and how long it is. It’s that every so often I like to discuss design and UI elements, and it’s better to be able to display bigger images in the first place, instead of having to post small images you have to click to enlarge.
This theme does a much better job at being responsive and mobile-friendly than the previous one.
Thankfully, this theme proved to be much easier to customise than previous ones, and was almost perfect for my needs right out of the box. The amount of tweaks I had to perform in the CSS have been really minimal.
This theme has many tricks up its sleeve, and through shortcodes it lets me easily do some text stylings (like asides, and left or right pull-quotes) that required a lot of CSS noodling in past iterations of this site.

But most of all, I was able to finally update to WordPress 4.8.x without issues.

There are still a couple of things I may change down the road, some information in the main pages to be updated or rewritten, and I can’t guarantee you won’t find broken links or layout glitches in past articles. But at the time of writing I deemed the site to be ready for prime time and to be put out of maintenance mode. If you encounter strange things while navigating, and they persist after clearing your browser’s cache, let me know.

iPod shuffle (3rd generation) — a post-review

October 11, 2017

by Riccardo Mori

Tech Life

IPod shuffle 3GII

I’m calling this a post-review because I don’t like the term post-mortem, which perhaps is technically more correct.

The other day I made an impulsive purchase. In the window of a local second-hand shop, I noticed this little buddy at €15. I know this iPod is possibly one of the quirkiest Apple products, but at that price I thought it was foolish not to seize the opportunity. Sadly, the iPod didn’t came with its original box or packaging. I was simply given the unit and its connection cable — and that I consider one of the coolest dongles Apple has made. Look at it:

iPod USB cable

Anyway, this is a third-generation iPod shuffle, first introduced in March 2009, and discontinued in September 2010. It is the model with the shortest lifespan of all the iPod shuffles. Even the first-generation (the white stick) lasted a couple of months longer. But it’s also the model with the most daring design of all. For how it looks and how it works.

Apple made a bold move at the time. They retired the successful second-generation iPod shuffle (the second from the left in this image) and introduced a new iPod shuffle that lacked physical controls entirely. The only one is the ON/In order/Shuffle three-way switch on top of the unit:

At a glance

For all media controls, you relied on the included Apple Earphones with Remote. The iPod ‘revolutionary’ feature was VoiceOver: after enabling it in iTunes, the iPod could speak song names, artist names, album names, playlist contents, and even battery status in 20 different languages. This iPod shuffle was also smaller and lighter than its predecessor, and it’s also the smallest and lightest (10.7 grams) product Apple ever made.

The third-generation iPod shuffle was introduced in two series. The first came in March 2009, with 4 GB models in silver and black. The second came in September 2009, and came in colours — black, silver, blue, green, pink models available in 2 GB and 4 GB capacities, plus a more expensive 4 GB stainless steel model, exclusively available through the Apple online Store. My recent purchase is one of these third-generation Late 2009 models, a silver 2 GB iPod.

From Apple.com homepage, March 2009

From Apple.com homepage, September 2009

The stainless steel Special Edition model

Using the iPod

While I appreciated Apple’s will to innovate, at the time this iPod shuffle was presented I thought that moving all controls to the earphones was too drastic a move, and that having only the three physical controls on the earphones’ remote was a step back usability-wise, in that you had to memorise additional gestures to control media playback. Finally being able to use this iPod eight years later, I can say that:

1. Having all the essential media controls on the earphones and not on the device was not a good choice on Apple’s part. Of all the earphones and headphones I own, only the Apple Earphones with Remote work with this iPod. The current Apple EarPods work only partially: the central button is recognised, but the Volume +/- buttons are not. A no-brand, third-party pair of earphones with a similar remote aren’t recognised at all (you can only listen to music; no play/pause, no volume up/down). If you acquire a third-generation iPod shuffle today, make sure you have a pair of Apple Earphones with Remote. According to this paragraph on the iPod shuffle Wikipedia page, Several months after the third generation release, several third-party companies, including Belkin and Scosche, released adaptors which can be used to add the controls to standard headphones. I believe they’re an essential addition if you want to use the iPod shuffle with your favorite earphones. The Belkin adapter mentioned before looks like this:

2. For basic playback control, things aren’t so bad. The volume buttons on the earphones are self-explanatory. Then, all you have to remember is to click the Center button once to play or pause, double-click it to play the next track, and triple-click it to play the previous track. If you want to hear artist and song title, press and hold the Center button. The iPod manual gives you the basics:

Basics

Navigating playlists with the VoiceOver feature feels more complicated. From the manual:

To choose an item from the playlist menu:

Click and hold the Center button on the remote.

Continue holding after you hear the current song announcement, until you hear a tone.

Release the Center button at the tone. You hear the names of your playlists.
When you’re listening to the playlist menu, you can click the Volume Up (+) or Volume Down (-) button to move forward or backward in the playlist menu.

When you hear the name of the playlist you want, click the Center button to select it. You hear a tone, and then the first item in your playlist plays.
To restart a playlist, follow these steps to select the playlist you want.

All this dance of clicks, pauses, double-clicks is a bit confusing. This happens when you have only three buttons to accomplish several tasks of varying complexity. That’s why Apple produced a Guided Tour video, explaining the interactions in the clearest possible way. I remember some people back then complaining about the lack of intuitiveness of this iPod’s controls — they felt the navigation to be awkward and frustrating. I tend to agree, although I want to point out that, if you keep things simple (play/pause/skip and volume up/down), the third-generation iPod shuffle is rather intuitive to use.

3. When it debuted in 2009, I thought the VoiceOver feature was essentially a gimmick, but its usefulness was quickly apparent to me the first day of use. The night before, I had filled the iPod shuffle with random selections from my sizeable iTunes library, and when I was listening to the music the day after, there were some songs I didn’t immediately recognise (probably picked from the least played in my collection), so hearing artist and title via VoiceOver was helpful.

4. It is no surprise that the aspect I find most amazing about this iPod is just how small and lightweight it is. I own other ‘wearable’ iPods — the 2nd-generation shuffle and the 6th-generation nano — which aren’t exactly heavy devices (15.5 and 21.1 grams respectively), but this shuffle truly disappears after you clip it on what you’re wearing. And you can easily fit earphones and iPod in a small pocket of your jacket, trousers or backpack when you’re not using it. Imagine if Apple had added Bluetooth capabilities to this minute iPod. Coupled with wireless earphones, this solution would feel even more invisible.

Conclusion

The third-generation iPod shuffle is without doubt one of the most peculiar and unique iPods. It seems rather obvious that its success at the time was limited: the fact that Apple went back to the design of the second-generation iPod shuffle when they introduced the fourth-generation model is a clear indicator that people were not happy with the user interface and interaction of the third-generation model. Some didn’t like the limited choice of headphones; some didn’t like the revolutionised controls; some didn’t even like its size, too minuscule for their taste. I admit I was sceptical, too, back in 2009. After a few days of use, however, I’ve definitely warmed up to the little guy, finding it quite practical and inconspicuous when out and about, its interface less weird than anticipated. At €15, it was an irresistible purchase (this particular model was €55 new), but I think I’ll have to look for one of those aforementioned adapters if I want to be able to use the iPod with more than just one pair of earphones…

Do iOS devices get slower over time? No and yes.

October 9, 2017

by Riccardo Mori

Software

Futuremark, in a recent post:

Last week, a story went viral that claimed Apple was intentionally slowing down older iPhones to push people to buy its latest models.

The claim was based on data which shows Google searches for “iPhone slow” spiking dramatically with the release of each new model.

And while plenty of reputable sites debunked the logic of that claim, no one looked at actual performance data to tell the true story.

Fortunately, we have plenty of real-world data we can use. Since 2016, we have collected more than a hundred thousand benchmark results for seven different iPhone models across three different versions of iOS.

These benchmark results provide a unique insight into the everyday performance of each iPhone model over time. And, as you’ll see, there are no signs of a conspiracy.

As Shawn King remarks at The Loop, This is a charge that has been leveled at Apple since they released the second iPhone.

While logic and data have demonstrated that there’s no one at Apple with a ‘remote Slow Down switch’ aimed at older iPhones and iPads, like other people I have observed two main phenomena related to performance and older iOS devices:

After installing the latest iOS version on an older, yet still supported device, there is usually a perceived slowdown in the responsiveness of the interface. Certain actions and UI animations appear slower, and the device appears to be struggling more when carrying out certain tasks.
Performance of older devices that aren’t getting system updates anymore appears to degrade over time anyway.

The most likely explanation for №1 is that the latest version of iOS is especially tailored to work at its best on the latest hardware, and while I’m sure there are optimisations in place to make it work just as fine on older, supported devices, the user may still notice the occasional slowdown or delay here and there during everyday use. Except for extreme cases, users simply get accustomed to the new ‘feel’ of the latest iOS on their devices, and it all becomes a non-issue over time. It happened for me in the past after installing iOS 4 on the iPhone 3G, or after installing iOS 7 on the iPhone 4, to make just a couple of the most classic examples.

№2 is worse, as I’ve observed personally. It’s worse because, over time, you notice it more, not less. Especially if you’re still using iOS devices with a 32-bit architecture (all iPads that came before the iPad Air; the first iPad Mini; all iPhones that came before the iPhone 5s; all iPod touch models that came before the 6th-generation iPod touch).

What happens, I think, is that as third-party apps get updated, they are generally optimised for the most current iOS version; new features are added to work with the latest iOS version; compatibility and bug fixes are tuned so that the app can work at its best with, say, iOS 11. But when the system requirements of that same app are, e.g., “iOS 9.0 and higher”, your devices that are still running iOS 9.3.5 or 10.3.3 get that update too, and from personal experience in most cases that updated app won’t be as optimised for older devices as it was before. Sometimes in the release notes for an app update you notice entries like Fixed crash on launch under iOS 9, or Fixed crash issues on older iOS versions, which I find somewhat telling of the optimisation process behind the scenes.

Mind you, I’m quite understanding towards developers. I can’t even begin to imagine the amount of work and debugging they have to do constantly to prevent anything from breaking on the many different iOS devices out there. Still, sometimes I wish I could revert to a previous version of an app, because it felt more stable and responsive before an update which was meant to improve things on newer iOS versions anyway.

Another factor that has impacted performance on older iOS devices more and more noticeably over time is Web advertising. Content blockers, which appeared in iOS 9 for the first time, don’t work on 32-bit iOS devices. Browsing the Web on older iPads and iPhones has progressively become a pain because the browser has to render everything and execute all ad-related scripts. I’ve had my iPad 3 since 2012: all Web-related tasks have slowed down considerably in five years. My workaround is to use Brave, the only browser with ad-blocking and script-blocking capabilities built in (i.e. they don’t depend on system content blockers, so the browser blocks ads even on 32-bit iOS devices — read this past article for more information).

The issue, however, persists with all apps with an internal Web browser using the system’s UIWebView to render Web pages. So, even if an app is still usable in iOS 9 on an older device such as my iPad 3 — like Flipboard, for example — whenever I need to access one or more Web pages from inside the app, and such Web pages are riddled with ads as it happens so often now, the app’s performance is noticeably impacted. Sometimes this even leads to freezes and hangs where the device stops responding entirely and a force-reboot is needed. In extreme cases I witnessed the device self-rebooting, even.

The most unfortunate device in this position is the iPad 2 running iOS 9.3.5, whose overall performance has taken a huge hit due to both these factors (poorly optimised apps, and Web advertising slowing down Web browsing). Every time I pick up my wife’s iPad 2, I keep thinking that Apple should have never allowed this device to be updated to iOS 9 in the first place.

To conclude: as Futuremark demonstrated through all the data they’ve gathered, no, hardware performance in and of itself doesn’t degrade over time. CPUs and GPUs still perform today as they performed when the device was new. It’s the software that, update after update, becomes more demanding and impacts performance more and more severely. Sometimes the drop is just limited to specific areas or apps. In other cases, like with the iPad 2, the snowball effect is such that the whole device becomes barely usable.