Mike Wuerthele, Apple Insider:
Following anecdotal reports of a keyboard more prone to failure than in previous years, AppleInsider has collected service data for the first year of release of the 2014, 2015, and 2016 MacBook Pros, with an additional slightly shorter data set for the 2017 model year given that it hasn’t been available for a year yet.
Not including any Touch Bar failures, the 2016 MacBook Pro keyboard is failing twice as often in the first year of use as the 2014 or 2015 MacBook Pro models, and the 2017 is better, but not by a lot.
First things first, this is some nice, boots-on-the-ground reporting. The numbers are relatively small, but seems a good, if not precise, indicator of the problem.
Apple has a second-generation MacBook Pro keyboard. It is in the 2017 MacBook Pro, and repaired 2016 models. The repair percentages on those are up from the 2014 and 2015 keyboards as well, but not nearly as much as the 2016.
Note that all Apple Store MacBook Pro stock has the new second-generation butterfly keyboard. You can’t walk into an Apple Store and walk out with the old MacBook Pro keyboard. So the data that matters going forward is the repair data on machines with that new keyboard.
I’d like to see a bigger survey of machines with that new keyboard, to get a surer sense of the success Apple has had in solving the problem. The sense I get in talking with people at my local Apple Store is that the new keyboard design made a world of difference in terms of reliability. This just words? Is the dust problem still there?
One thing that I know is true: There’s no way for an average buyer to take their machine apart and swap out the keyboard if they do encounter this problem. As Mike says, it’s not an easy repair, requiring complete disassembly of the machine.
This headline and analysis is completely wrong. The data doesn’t show that they are failing twice as frequently at all. It shows that the percentage of repairs that were for the keyboard doubled. This is so unrelated to the percentage of sold Macbook Pros that have experienced a keyboard failure that nothing can be concluded about that from this. Gruber already had to rewrite his post about this because of the same error https://daringfireball.net/linked/2018/04/30/appleinsider-macbook-keyboards
My friend is on his third keyboard. Hopefully this last one is the 2nd generation version. But he still hates typing on it. I tried it and it was horrible.
It’s worth teasing apart the *feel* of the keys from the reliability. The feel is polarizing; some of us think it’s the best-feeling keyboard ever made, and others think it’s like typing on a slab of glass. And that’s a trade-off Apple makes, just like they make a bunch of other tradeoffs.
Reliability is a deeper issue, though, and if it’s an issue, is far less justifiable.
I don’t see much upside to them denying false stories, though. As soon as they do, the narrative switches from “Why is Apple keeping quiet about this problem?” to “Why is Apple lying about this problem?”
This is a good theory. Also, you may get more keyboard repair requests from people who tried prying off a new key, only to discover they broke it.
Are you saying you don’t think there isn’t a reliability problem with these keyboards?
Ahhh… double negatives.
Should be: Are you saying you don’t think there is a reliability problem with these keyboards?
What would satisfy you as proof?
The anecdotal evidence is pretty strong.
I don’t think this is the same as “iPhone X sales are weak” (which is this year’s variant of the quarterly “Apple is doomed story” we’ve seen since forever); or even stuff like “bendgate”.
Those are sensationalist “general media” stories. The keyboard stories are “Mac media” stories.
The only problem is that’s not what the numbers say. That is NOT how to read the data they’ve collected.
AppleInsider, and now Dave, say that the numbers show the keyboards are failing at twice the rate of prior MBP keyboards. Maybe they are, but that’s not what can be deduced from what they’ve reported.
The numbers, as presented, show that IF a MBP is brought in for repair to these particular Apple Stores, it’s twice as likely to be keyboard related than in prior models. It does not present enough data to draw any conclusions about total failure rates. You’d need total sales figure, etc. to draw those kinds of conclusions.
It isn’t even enough to draw any conclusions about the keyboard durability itself. If the rest of the MacBook Pro is holding up much better than in prior models, and the keyboard failed at about the same rate, then you’d still see the same type of increase.
That’s right. Percentage fallacies aside, it could also work both ways in terms of the keyboard itself—for example, more awareness might mean more brought in whether or not the keyboard is an issue, or maybe it’s in fact failing even more and people just put up with semi-stuck keys.
This is where Apple’s silence causes a narrative to be fueled, whether true or not. I understand Apple speaking up may not help in many cases, but I still wonder if there would be a better way to manage this, especially since a number of high-profile, notable folks in the Apple space have commented.
My non-Touchbar MBP13 was eligible for a battery repair so I took it in and got that done which mean they also replaced the top of the machine including the keyboard. This keyboard is a little better in terms of feel (very slightly more travel) and the keys feel… spongier isn’t right but there’s more of a physical feel to them. We’ll see.
I think one of the understated reasons for the increased complaints isn’t only the failure rate which might be higher, it’s that you can’t easily clean under a key. Get a crumb under a key? In prior versions you could carefully pry the key plate off and clear that out. I would bet this led to a lot of self-repair issues that now, with the less repairable keys, result in a trip to the store.
Anecdote: I’ve had to use canned air once on my 2016 MBP and I’m very clean with my computers, so I think the low tolerance of the new keyboard is a potentially severe design flaw.
The question is, how is this realized in practice? What do we define as a reliability problem? What’s an acceptable failure rate? If a normal keyboard has a 1% failure rate in X years, would a 2% failure rate count? A 10% failure rate?
Right now, we have a ton of anecdata which is tremendously worrying and strongly suggests something’s different. However, we don’t know how much this is affecting consumers broadly in practice. Absent guidance from Apple, we’re guessing.
No. I’m saying it hasn’t been proven that there is. It might be likely, but it’s not established fact.
Mentioning the effect of denying these stories is based on previous rounds of (now) proven bullshit. I can’t even remember the last one (maybe it was about weak iPhone X sales their FIRST quarter?), but the narrative definitely went from “Why is Apple silent on this?” to “Why is Tim Cook lying in an earnings call? Conspiracy to defraud everyone!” (Hint: He wasn’t.)
If Apple decides the numbers indicate a serious problem, they should (and likely will) recall them or offer free repairs. In the meantime (before they’re convinced and have a plan in place) denying the story would be bad and acknowledging it wouldn’t be especially helpful.
Sorry about edits, took a while to say what I wanted. 🙂
It’s worth noting that Apple *does* have that data. And there are issues, something is done eventually.
Without real numbers it absolutely is. I am personally very tired of pundits turning their pet peeve of a problem into a sensational story of malice and corruption.
I’ll grant you this isn’t all over the media, but that just allows the Mac media to behave worse.
Does the control key have a ^ above it (which is the symbol for “control”)? If so, it’s the 2017 version. If not, it’s the 2016 version.