Archive for May 1, 2019

Wednesday, May 1, 2019

Apple’s Q2 2019 Results

Jason Snell:

Apple’s quarterly results are in. The company posted revenue of $58 billion, down 5% from the same quarter a year ago. iPad revenue was up 22% and Services revenue was up 16%, but Mac revenue was down 5% and iPhone revenue was down 17%.

We’ve got lots of charts below, as well as a transcript of CEO Tim Cook and CFO Luca Maestri’s conference call with financial-industry analysts.

John Gruber:

At 20% of the company’s revenue, Services now accounts for more revenue than Mac and iPad combined.

I don’t see how this is good for the quality of the products or, ultimately, for customers. The continual notifications and extra screens to tap through are like the laptop stickers that Steve Jobs hated, and they’re just the smallest example of how the focus on services is shifting the company’s attention and priorities.

Tim Cook:

For our Mac business overall, we faced some processor constraints in the March quarter, leading to a 5 percent revenue decline compared to last year.

This seems like an odd comment, unless it’s just intended to lay more groundwork in Apple’s case for ARM. Are Mac buyers really that worried about processor speeds rather than, say, keyboards? And if slower processors are the problem, isn’t that mostly self-inflicted?

See also: Dave Girouard.

Previously: Apple’s Q4 2018 Results.

Update (2019-05-02): John Gruber (tweet):

I asked an Apple source last fall why it took so long for Apple to release the new MacBook Air. Their one-word answer: “Intel.”

Jeff Baxendale:

I know there’s not going to be a switch given impending ARM Macs, but would have been nice to just have Ryzen Macs instead of complaining about Intel.

They’re a way better deal, nobody buys for “Intel Inside”, and then maybe the integrated GPUs wouldn’t be total garbage 🤷‍♂️

Update (2019-06-03): Kevin Bartlett:

Just read @gruber’s post about Intel constraints on processors holding back Apple. It’s holding back all PC sales. We use Dell at work and have had a month or longer wait times on laptops because Intel can’t get chips to them either.

Microsoft’s Resurgence


Microsoft Corp. today announced the following results for the quarter ended March 31, 2019, as compared to the corresponding period of last fiscal year:

  • Revenue was $30.6 billion and increased 14%
  • Operating income was $10.3 billion and increased 25%
  • Net income was $8.8 billion and increased 19%
  • Diluted earnings per share was $1.14 and increased 20%

As someone who doesn’t follow Microsoft that closely, I’m impressed with how they have turned things around—and how quickly. It seems like they have a plan that makes sense and are executing well.

Tom Warren (Hacker News):

Microsoft has become the third US company to pass a market cap of $1 trillion. The software giant passed the milestone briefly today after a jump in stock price today following strong fiscal Q3 earnings. Microsoft joins Apple and Amazon in hitting the $1 trillion valuation ahead of rival Google.


Microsoft’s latest earnings also revealed that the three main buckets the company splits its businesses up into are all doing well and roughly contributing the same amount of revenue this quarter (around 30 percent each).

  • Office, LinkedIn, and Dynamics = $10.2 billion in revenue
  • Azure cloud, server products, and enterprise services = $9.7 billion in revenue
  • Windows, Xbox, and Surface = $10.7 billion in revenue

Bryan Beal:

Microsoft’s conversion of Office to a ubiquitous enterprise cloud service is truly impressive.

Remember when everyone said Google Apps would “take over the Enterprise”? I laughed back then. Now it’s all but impossible. Google missed the chance.

John Gruber:

One amusing side note: The press release was obviously written in Word and exported to HTML. Just look at the source.

Ben Thompson:

The critical breakthrough was three-fold, and, as it so often the case, the three break-throughs were really about the same existential question — whither Windows:

The most important factor that made all of this possible, though, is that for all of the disruption that the enterprise market has faced thanks to the rise of software-as-a-service (Saas), Microsoft was remarkably well-placed to take advantage of this new paradigm, if only they could get out of their own way.

A contrary take:

From revenues as well as profit perspective Ballmer actually grew Microsoft more than Nadella. Market just decided not to reward that. Current growth centers including Azure, Surface etc were all started and championed originally by Ballmer. It was Ballmer who was willing to pour in investment like crazy in Hololens as well Bing. On the other hand, Nadella has failed to add single new product in Microsoft's portfolio during his half decade as CEO. Microsoft today has simply no presence in home automation, smart assistants, wearable or self-driving markets that have emerged during past 5 years. Nadella has literally missed boat in every single new category that has came around during his time. All the while its rivals have moved fast and picked up top spots in these new markets. Windows revenues is still falling like crazy and obviously there is no Smarthone story from Nadella yet.

The only thing that is compensating all these troubles is cloud and that too had been hazy on actual utilization. The engineering execution in Azure begs a question about how much worse it can get. Out of all providers, Azure literally has been the least impressive from technical standpoint in everything from UX to availability to features to documentation to API design. Once cloud market saturates and race to bottom ensues, Microsoft could have big trouble in maintaining current revenues and profits due to lack of new products. It will take few quarters before stock market reacts but all of these same people praising Nadella right now would suddenly come around and brand him the worst CEO in history as soon as stock dips.

Update (2019-07-23): Charlie Bilello:

Microsoft was the largest company in the world back in 2000. Then it suffered a 70% drawdown over the next 9 years to its low in March 2009. It did not surpass its 2000 high until 2014, 14 years later (total return). Today it is once again the largest company in the world.

TextEdit Deletes Original File Even When You Cancel

Chris Hamady:

Unbelievable...why in heaven’s name does Apple allow TextEdit to modify files EVEN IF the user CANCELS the save command. This video was just made on Yosemite, but I’ve also confirmed that TextEdit on Mojave does this as well:  … Notice html changes to rtf.

I’m interested in your take on this. Should any app on Mac OS have the ability to change a file format/type without a user saving the change?

CM Harrington:

This whole thing is probably an extension of autosave… and why it takes forever to close a file without changes you’ve made even if you have autosave off (I think it’s deleting all the SQLite rows).

Ben Szymanski:

It’s really hard to trust the system auto save functionality with these UX oversights. It was rough in Lion and it’s gotten maybe(?) marginally better since then.

I think it’s gotten worse with Preview.

Howard Oakley:

The real bug is in older macOS, where that second dialog doesn’t do what it claims. But the whole behaviour is unnecessary: TextEdit is going out of its way to delete your original HTML document when there’s no need at all. It does that because whoever implemented this behaviour didn’t understand macOS (or iOS for that matter).


For many years, Apple used TextEdit as exemplary code for macOS developers to see how it’s done. The last time it did this was seven years ago, in early 2012, since when I can only presume that TextEdit has gone steadily downhill and is now too embarrassing to release in source form. Like so many other standard tools in macOS, TextEdit is another festering sore on the rump of Apple’s engineering indolence.


Folder Access and Inconsistent App Review

Matthias Gansrigler:

It’s the typical case.

A feature that was approved with a previous build suddenly is not allowed anymore.

I bet if I remove this feature as requested, I’ll get rejected again, because Yoink is able to install a (macOS Mojave) Quick Action (which was also approved before).


2.4.5 Apps distributed via the Mac App Store have some additional requirements to keep in mind:

(i) They must be appropriately sandboxed, and follow macOS File System Documentation. They should also only use the appropriate macOS APIs for modifying user data stored by other Apps (e.g. bookmarks, Address Book, or Calendar entries).

All he’s trying to do is let the user click a button to install a PDF service. You’d think that would be allowed, since the app is sandboxed, the installation is happening due to explicit user request, and it prompts to allow access to the PDF Services folder. As far as I can see, the documentation doesn’t say not to do this. There is no API for installing PDF services, so the only alternative is to put up some instructions and ask the user to do it manually (which is what I had to do for EagleFiler). What’s the point of having the entitlement if you aren’t allowed to actually use it? Or, rather, how can something so basic not be handled consistently by App Review?

I’ve heard similar stories from other developers. And when I’ve asked whether certain designs/features would be allowed for my apps, I could only get perfunctory responses saying that I needed to follow the guidelines and would have to develop the feature and submit the app before finding out whether it would be acceptable.


Plans for Third-Party Marzipan Apps

Steve Troughton-Smith:

Are any iOS developers ready to commit to bringing their apps to the Mac via UIKit publicly? Any Mac/iOS developers preparing to sunset their existing Mac app for its iOS variant? I aim to go all-in on Marzipan with at least four apps on day one, one replacing a Mac version

Marco Arment:

It’s hard to say with 100% certainty since we don’t have concrete info and released tools yet, but I plan to ship Overcast as a Marzipan app on day one if I can.

Kyle Howells:

I intend to port several of my personal utilities and 2 as yet unreleased in progress iOS apps to the Mac if marzipan is good enough.

Only one of those was ever likely to see an AppKit version.

Erik Schwiebert:

The really interesting thing will be what the LargeCo’s do with their apps. I don’t yet know what we (MSFT) will do with Office; there’s a lot of code shared between Mac and iOS but much is very different. Multiple windows, file management, app/suite integration, sandboxing, etc.

Peter Steinberger:

Pretty likely that we bring @pdfviewerapp over to the Mac. After all the prototype last year already worked great and there’s definitely a market.

Greg Pierce:

Regardless of what technical aspects you have uncovered that make you bullish on Marzipan, it would be reckless for anyone who does this for a living to commit to shipping on it until we hear how the business end of this transition is going to work.

Mike Piontek:

I’ve spent 2 years working on updates that include big changes to my Mac app, so I’m not in a rush to throw that out… but If I have to for features like Siri Shortcuts I’ll consider it. I was previously assuming I’d want to wait a couple years, since this will require 10.15+.

Pádraig Kennedy:

I’d love to bring Castro to Mac some time, but I’m pessimistic about Marzipan so if I was doing it this year, I think I could make something more Mac-like using AppKit 😬

Wooji Juice:

Not going to commit to something sight-unseen, but strongly interested in creating a Mac version of Ferrite (and maybe 1-2 other apps) using Marzipan. Will have to see what gets released, how stable it is, & what issues there are making quality non-shovelware Mac apps with it


We’re excited to bring proper manual controls to MacBook and iMac webcams and intend to ship early April next year

Christian Selig:

I’d love to bring Apollo to the Mac, if only to scratch my own itch.

Markus Müller-Simhofer:

It really depends on what Apple is doing. If Apple uses Marzipan for it‘s productivity apps (Mail, iWork) we will also switch. I always disliked how Carbon apps felt. I don‘t want to be one of the apps that „feels“ different than all the other apps.