MacInTouch Amazon link...

competition, technology, Apple management & plans

Channels
Apple, Products
Yeah, that worked out really well for Microsoft. Especially their mobile products. The more I see, the more the current Apple looks like the old Microsoft. Feh.
I would be a little careful assuming that just because developers can develop for all platforms at one time, that is the only outcome. Certain applications will only work best on a Mac or only work best on mobile. The ability to develop once is great where it makes sense - think about websites that operate both on mobile and on full-size screens with keyboards and mice. However, forcing everything into that model is shoehorning the impossible, and that's more akin to Microsoft’s problem.
 


Apple has also shown a general disregard for "pro" Mac users of all stripes in recent years; the very introduction of a "pro" MacBook with only USB-C ports was itself a slap in the face of anyone who used the Mac in any kind of media environment. Don't get us started on the trashcan Mac "Pro," which they haven't meaningfully updated in how many months [years]? Also, the disappearance of the Mac Mini server edition, and the evisceration of Mac OS X Server app, as exhibits B and C. Perhaps maybe now that we've seen "peak iPhone", Apple will start to care about people who use Macs for work, again. However I would not place a bet on this.
I've been an Apple user since the Apple ][+ and a Mac user since the first Macintosh. I use my Macs for audio production and post production. I have been in need of a new uber-powerful desktop for a while. People in my field need power and expandability. I need to be able to upgrade hard drives and memory and have lots and lots of ports for adding peripherals.

My typical setup requires USB port for my iLok, another USB port for my mixer/control surface, ethernet for my audio interface (using an audio networking protocol), a Thunderbolt port for my work drive, a Thunderbolt port (separate bus) for my Virtual Instrument drive, a Thunderbolt port for my video drive (hosting video files for playback), and either USB 3 or Thunderbolt for a backup drive. Which means, I buy a Mac (at a premium for a lower powered system than currently available in the PC world) and then spend another $1k on port multipliers/docks/etc. It's just a fiasco. And forget about a mobile system for remote recording. No current MacBook Pro allows the connectivity I need to run a location recording session.

It has come to the point where I am seriously considering a PC after 30+ years with Apple hardware. Even the manufacturer of my preferred audio interface is moving towards Windows and Linux support in the near future (after almost 20 years as a Mac-only company). When they finish their work on a Windows driver I will have little holding me in the Mac world... especially if the new Mac Pro is an overpriced, underpowered, closed system. I know Apple promised it to be "modular", but they also originally promised a 2018 release. And "modular" is a slippery word, and I'm betting it doesn't mean the same thing to Apple that it means to me.

Apple is driving creative professionals away, not because they want to move over to Windows, but because they're being left with little choice and have to find ways to keep working and remain profitable in their business. When your business revolves around technology, the end deciding factor becomes ROI - and Apple is doing everything in their power to make sure they lose that ROI argument for us creative professionals.

Here's hoping Apple turn it around (fingers crossed)...
 


The idea of a single cross-platform toolkit is a holy grail that toolkit developers have been pursuing for decades. They have become remarkably capable for desktop environments, partly because people have been working on this almost since the dawn of GUIs but also because modern GUIs today have converged to very similar behaviors. The really unusual ones (e.g. SunView and OpenLook) are not used, and the rest are close enough that most (but definitely not all) of the differences are cosmetic.

But taking it to the next step, to be portable across different classes of devices (desktops, handhelds, watches, etc.) is a completely different kind of challenge. These interfaces don't only look different, but their behavior is extremely different. And the nature of the devices are that we are not likely to ever see the UI paradigms converge the way different desktops have.

But the research will continue and the toolkits will get better, even if they never become perfect. I don't think they're ready for widespread adoption, but they might get good enough in the future.
 


... The T2 chip, which Apple markets as a security enhancement, eliminates any possibility a T2 Mac might boot from a less expensive and potentially faster industry-standard NVMe drive.
[But] the T2 does not solely limit boot to the one internal Apple drive. The default "out of the box" settings are that way, but they are also settings. The T2, in and of itself, is not that narrow.
There's no upgrading the boot SSD in a T2 Apple computer, so when buying, you’d best buy the largest SSD you can afford.
Technically, the iMac Pro could be upgraded. The flash NAND daughtercards are replaceable. The T2 controller itself isn't. So it wouldn't be a 'buy commodity parts off the shelf' solution, but it could be changed. It does mean starting over with a backup, though.

That really has little to do with the T2 itself. The last 2-3 iterations of Apple's blade SSDs had Apple SSD controllers on them. For the Mac models that are on the path of ever-shrinking z-heights, Apple was still on the path to direct mounting. Whether Apple enhanced the security of the boot media or not, that logic board mounting was being driven by other higher priority factors. Apple would still be in the SSD subsystem-making business if they didn't do a feature "superset" of the T2.
Rather a shame, as SSD capabilities are rising, and prices falling.
The major problem with Apple's approach is that they are disconnected from that latter point. Apple's SSD capacity prices are for the most part detached from the market prices.
Again this is for the most part independent of the T2. If Apple was using a 3rd-party SSD controller with Apple-tweaked firmware, then for the z-height limited systems it would still be probably soldered to the logic board to save height.
It takes effort to design a T2, and while there may be some security benefit over and above an encrypted boot SSD,
T2 doesn't only protect the default boot SSD. It protects the firmware also, and it protects user biometric data - both of those in ways a completely removable subsystem cannot match in security. The "may be some benefit" isn't really a "may". It is much more than just a simple SSD controller.
The financial benefit to Apple from non-upgradeable Macs is substantial... and will continue years down the road by blocking parts from being salvaged.
A large 3rd-party market of salvaged parts never was Apple's or the vast majority of users' primary objective. Most folks want a system that works and doesn't fail.

We'll see. The repair costs for Apple during the span that the systems are covered by AppleCare is also higher. That may be one of the factors for them goosing NAND prices so hyper-inflated over market levels to recoup that higher cost. As more users exit AppleCare and are then hit with very high repair costs, they aren't particularly going to be "vey high satisfaction" users either. Over time Apple will lose revenue on this - the negative outcome feedback loop isn't short.

More integrated (more multi-function chips doing more in less space) is a path that computers have been on for a long time and probably won't stop for many years to come. That systems will have less discrete parts in the future is just going to continue. Might as well grumble at the sun coming up every morning... it is coming.

Where Apple is failing down a bit is that this should be guided by getting to a net increase in reliability. It is a balance where their OCD on thinness gets in the way.
 


... The T2 chip, which Apple markets as a security enhancement, eliminates any possibility a T2 Mac might boot from a less expensive and potentially faster industry-standard NVMe drive. There's no upgrading the boot SSD in a T2 Apple computer, so when buying, you’d best buy the largest SSD you can afford. Rather a shame, as SSD capabilities are rising, and prices falling....
[But] I'm happily booting my bought-deliberately-with-a-puny-128GB-internal-flash-drive MacBook Pro 13" from an external (OWC) RAID 1 flash drive (Mercury Elite Pro Dual with two 1TB flash drives).

External interconnect is now fast enough that one doesn't need to worry too much about performance of external flash. And who'd want to rely on the internal flash not just wearing out too much so that the machine's life was limited? Of course, it's another <medieval expletive> box to carry around, but the machine can operate from the internal disk when I don't need all the high value cruft in the external (a client visit, for example).
 


Ric Ford

MacInTouch
Samsung is giving Apple a little competition in technology, announcement events, user interface innovation, pricing... etc.
Wired said:
The Entire Galaxy of Hardware Samsung Announced Today
It's the 10-year anniversary of Samsung's Galaxy phone, a line of devices that made phablets cool, survived baptism by fire, and turned Samsung into a leader in the mobile phone market. Today, at an event in San Francisco, the company showed how it plans to push its smartphones into the next decade—and into another dimension. The vision includes 5G, folding displays, and a brand new line of Galaxy phones filled with flashy features.

For Samsung, this isn't just another product showcase. It's the company's pitch for "the next era of mobile innovation," as D. J. Koh, CEO of Samsung's mobile division, put it onstage today. Samsung wants you to buy its latest hardware, but it also needs you to buy into its vision for the future, so you'll keep returning to the ecosystem of Galaxy products for years to come.

If you missed the event, you can watch the entire thing here—or read on for the TL;DR of everything Samsung announced today.
Reuters said:
Blown away by innovation or price? Samsung's foldable phone opens new frontier
The South Korean tech giant’s Galaxy Fold resembles a conventional smartphone but opens like a book to reveal a second display the size of a small tablet at 7.3 inches (18.5 cm). It will go on sale from April 26.

At its launch event in San Francisco on Wednesday, Samsung upped the surprise factor by briefing analysts and journalists on widely anticipated aspects ahead of time, such as 5G versions of its existing top-end Galaxy S phones.

The subsequent unveiling of the foldable device came as a shock to many in the auditorium.
 


Upgrading the 128GB boot drive in the base 2018 Mac Mini to 2 TB adds $1,600 to the bill, and Apple gets to keep the 128 GB...

The Samsung 970 EVO 2TB NVMe PCIe M.2 2280 SSD is $550 on Amazon, leaving $1,050 to pay for a $456 Intel NUC 8 (NUC8i7BEH) with 8th Generation Intel Core i7-8559U processor, Intel Iris Plus 655 Graphics, room for an internal second (SATA) SSD or hard drive, and Thunderbolt 3, if you want to boot externally, or use other benefits of Thunderbolt.

NUC $456, 2TB SSD $550, 16 GB Patriot DDR4 2400 $90 = $1,096, $504 less than Apple's charge for just a 2TB SSD.

The roughly comparable Core i7 2TB Mac Mini with lesser Intel graphics is $2,899. If its 2TB T2/SSD combo dies, you'll probably not get your data off, so do back up. If it dies post-AppleCare, you'll have a very valuable door stop.
 


Apple is apparently pushing forward with its iOS-macOS integration strategy:
From a new article up on ZDNet
Liam Tung said:
iPhone, iPad, Mac apps merging? Apple SDK will create iOS apps that work on Mac
Apple could release a new software development kit as early as its June WWDC annual developer conference that would initially let developers port iPad apps to Mac.

The SDK would mean developers don't need to double up on code for the same app. However, they would at this point still need to submit different versions of the app to each App Store.
I can see where that makes sense from a management and marketing viewpoint. Whether it makes sense from a coding viewpoint is another question.
 


Ric Ford

MacInTouch
As Apple continues to push its "services" strategy:
Reuters said:
Apple, Goldman Sachs to jointly launch credit card paired with iPhone: WSJ
Goldman Sachs Group Inc has teamed up with Apple Inc to issue credit cards that will be paired with iPhones and will help users manage their money, the Wall Street Journal reported here on Wednesday, citing people familiar with the matter.

The card, which will be linked with Apple’s Wallet app, will allow users to set spending goals, track rewards, and manage balances, WSJ said.

The new cards will be rolled out to employees for testing in next few weeks and will be launched later this year.
 


Samsung is giving Apple a little competition in technology, announcement events, user interface innovation, pricing... etc.
For those who had problems with the price of the iPhone X ($1000+), I wonder what the reaction to the cost of the Galaxy folding phone (~$2k) would be? I realize it fits in a pocket, but it does seem to be on the outrageous side.
 


For those who had problems with the price of the iPhone X ($1000+), I wonder what the reaction to the cost of the Galaxy folding phone (~$2k) would be? I realize it fits in a pocket, but it does seem to be on the outrageous side.
If it is couched as "just a phone", it lands in the 'outrageous'. However, if couched as an uber-light laptop computer that fits in your pocket, that's not quite as much.

The 'Fold' has 12 GB of RAM and 512 GB of SSD, ~17 Wh battery, and USB Type-C port. Apple's "one port wonder" MacBook with 16 GB and 512GB SSD ( ~40Wh battery) is $1,799. it is the same ball park price-wise, as well as in capacity.

I don't think they are positioning this as trying to be the "biggest screen" phone, but more as phone, tablet, and super-portable computer. Or at least the iPad (Pro) in your pocket.

One of the narrow targets probably is "always on the move" folks who mainly use office productivity apps and communication apps. For a laptop budget, you also get a phone and a tablet with "mobile computer".

It doesn't appear to work with Samsung DeX, but that could arrive later (DeX works with the Galaxy Tab S4 and the new '10' series). It wouldn't be surprising if they didn't need the next version of Android to make more apps do 'three modes'. Even without DeX, it should pair with a keyboard. If the Type-C port could mirror/output the "tablet" display, then we could have something similar to DeX when in tablet mode.
 


Ric Ford

MacInTouch
Apple is apparently pushing forward with its iOS-macOS integration strategy:
From a new article up on ZDNet
Liam Tung said:
iPhone, iPad, Mac apps merging? Apple SDK will create iOS apps that work on Mac
Apple could release a new software development kit as early as its June WWDC annual developer conference that would initially let developers port iPad apps to Mac. The SDK would mean developers don't need to double up on code for the same app. However, they would at this point still need to submit different versions of the app to each App Store.
And another article:
Cult of Mac said:
First Macs with Apple chips could mean tumultuous 2020
Moving macOS computers from Intel processors to ones Apple has created itself seems to be on schedule. At least, that’s what Intel thinks, according to a recent report.

This is likely a part of bringing all the software that runs on iPhone, iPad and Mac together.

Unnamed sources at Intel told Axios that Apple is definitely making the move to custom-designed ARM processors, possibly as early as next year.

The first report of this project came from Bloomberg last spring, with a rollout expected in 2020 or 2021. Noted analyst Ming-Chi Kuo said last fall that the first Mac with an Apple chip could be out this year or next.
 



Did quick-and-dirty preliminary benchmarks of 2013 Lenovo M92p for comparison with Apple systems - faster than a 2017 MacBook Air for CPU but extremely slow for graphics:
(It has a single x16 PCIe low-profile slot plus two PCI slots and an x1 PCIe slot.)
The integrated Intel HD Graphics 2500 is indeed the weakest link. Depending on need, you have the ability to install a bus powered graphics card, such as:
 


Ric Ford

MacInTouch
Dell's XPS 13 competes head-to-head with Apple's laptops. Here's a review of the latest model, which is tiny but includes two Thunderbolt 3 ports, plus a USB-C port and an SD card slot (and a quiet, non-butterfly keyboard). A 4K touchscreen is optional ($300). A mid-range version with quad-core i5, 8 GB RAM, and 256GB M.2 NVMe drive is $1209.99. In contrast to Apple locked-up laptops, the XPS 13 internals are accessible and documented for upgrades and repairs.
Ars Technica said:
XPS 13 2019 review: One small move made Dell’s best laptop even better
Key fixes in this year's model mean the $899 XPS 13 has few things holding it back.

The Good
  • Compact, attractive design with large screen area.
  • Dolby Vision support.
  • A practically placed, FHD webcam.
  • Fingerprint sensor embedded into power button.
  • Stays fairly cool.
  • Solid performance.
  • Good battery life.
The Bad
  • 16:9 aspect ratio will bother some.
  • No one-touch power or login feature.
  • Fans can be noisy.
  • No LTE option.
The Ugly
  • No USB-A ports.
The XPS 13 Developer Edition, with Ubuntu, is $939 with quad-core i5, 4 GB RAM and 128GB SSD. Unfortunately, like Apple laptops, the XPS 13 has soldered-in RAM:
Dell said:
Note: The memory modules are integrated on the system board. If the memory modules are malfunctioning and need to be replaced, a replacement of the system board is necessary.
 


Dell's XPS 13 competes head-to-head with Apple's laptops. Here's a review of the latest model, which is tiny but includes two Thunderbolt 3 ports, plus a USB-C port and an SD card slot (and a quiet, non-butterfly keyboard). A 4K touchscreen is optional ($300). A mid-range version with quad-core i5, 8 GB RAM, and 256GB M.2 NVMe drive is $1209.99. In contrast to Apple locked-up laptops, the XPS 13 internals are accessible and documented for upgrades and repairs.
I appreciate these XPS updates.

I'm hoping by this summer or fall I can pick up a Dell XPS 15.6" i7, 32GB RAM, 1TB SSD, 4K laptop. I would then install Linux Mint on it and use it as a work computer (email, web browsing, address book, calendar, LibreOffice, etc.) I'll stick with my older Macs for Logic X and FCP X with no need for an Internet connection.

Since a discussion here about a year ago of good deals at Costco for an extended warranty when purchased with one of their credit cards (I believe it ends up being a total of 3 or 4 years), I've been keeping an eye on what they offer. I just checked their website and see what I'm looking for at about $1900.
 


Since a discussion here about a year ago of good deals at Costco for an extended warranty when purchased with one of their credit cards (I believe it ends up being a total of 3 or 4 years), I've been keeping an eye on what they offer. I just checked their website and see what I'm looking for at about $1900.
The program for the Costco credit card additional 2-year warranty extension, three years total, is no longer advertised, as it was previously, but it has not gone away. At least I no longer see it listed on the product pages for the affected products. There used to be an small banner noting the program. Instead, you can go to the following URL for details:

Costco Anywhere Visa® Card by Citi

A standard second year of warranty still applies to all such purchases, no matter how you pay for it directly from Costco.
 


Ric Ford

MacInTouch
So, Samsung beat Apple to the market with a foldable phone, and now there's this amazingly innovative camera system from Nokia:
Ars Technica said:
Meet the Nokia 9: Five cameras bring a different approach to phone photography
Many cameras combine to make a single, higher-quality photo.

... You press the shutter button, and the Nokia 9 takes a photo with each lens, the software mashes all five photos together, and outputs a single, better photo. This is called "image stacking," and it's the same technique that Google has been doing on the Pixel and Nexus cameras for some time, albeit with a single lens.

HMD's hardware-based approach to image stacking comes with some interesting implementation details. You get five lenses—three 12MP monochrome cameras and two 12MP RGB cameras.

... With the Light chipset, the Nokia 9 takes five simultaneous pictures—60 megapixels worth of data—and then downmixes them into a 12-megapixel photo with more dynamic range and better light performance than you could get from a single photo. HMD says the five sensors can collect "up to 10-times the amount of light than a single sensor of the same type."
 


Steven Sinofsky just shared some interesting thoughts in a series of tweets about Apple's rumored porting of macOS to ARM. (Sinofsky ran Microsoft's Windows Division for a few years. He's perhaps not always on the same page as the Apple community, but he's often quite astute.)

Visit Steven Sinofsky on Twitter and scroll down for the full list of tweets.
 


Ric Ford

MacInTouch
New operating systems are rare, but Google's building one:
Business Insider said:
It's an open secret that Google is quietly working on an entirely new operating system that's different than Android — here's everything we know about 'Google Fuchsia'

... What really differentiates Fuchsia from Chrome OS and Android is its core, which is not based on Linux but on a new kernel called Zircon. What this means is that Fuchsia has been developed as a system intended to work on a several platforms, not just phones and laptops.

... The big difference between Fuchsia's home screen and those of more traditional operating systems is the lack of apps. There's no dock, no desktop icons, no launcher. What is there, though, is Google's famous search bar — and in this alpha version of Fuchsia it doesn't search the web but rather the computer itself, including apps.
 


Apple is apparently pushing forward with its iOS-macOS integration strategy:
This heading might confuse some people. Marzipan is all about helping developers maximise code reuse when porting from iOS to macOS – it isn't about merging or integrating the operating systems. This issue was directly spoken to at the 2018 WWDC – during the keynote Craig Federighi said, "Are you merging iOS and macOS? Let me address that! NO!"… and displayed a huge "NO" on the screen for anyone who wasn't sure.

iOS and macOS have always sat on the same foundations, but some API design issues have meant far more work porting an app from iOS to macOS than you might expect. Marzipan is an effort to clean that up and minimise the amount of OS-specific coding/tweaking. But it's still a porting exercise, not simply a case of 'build once, run everywhere.'
 



I appreciate these XPS updates.
I'm hoping by this summer or fall I can pick up a Dell XPS 15.6" i7, 32GB RAM, 1TB SSD, 4K laptop. I would then install Linux Mint on it and use it as a work computer (email, web browsing, address book, calendar, LibreOffice, etc.)
I appreciate these XPS updates as well. Been using an XPS13 Developer Edition as my primary laptop for about 2.5 years now, and have generally been happy. I was a bit disappointed by Dell's decision to remove the Mini DisplayPort (although I guess Mini DisplayPort connections are hard to find) and go the Apple dongle route (which I hate).

I find the offerings from System 76 (Galago Pro) and Lenovo (the upcoming X390 and X1 Carbon) intriguing. The S76 offers two storage bays (one M.2 and one 2.5-inch, I believe, so you can configure one speedy M.2 startup drive and a slower SSD data drive), and the option for 32 GB of RAM. The new X390 will add the option of 32 GB RAM, as well, all in a package about the size of the older XPS 13's. I like that the offerings from S76 and Lenovo also offer a built-in HDMI port. No more dongles!

(I found that the Dell-branded USB-C to HDMI/USB-A adapter didn't work well with Linux. Oddly, the Apple adapter worked just fine.)
 


I find the offerings from System 76 (Galago Pro) and Lenovo (the upcoming X390 and X1 Carbon) intriguing.
I've recently set up two Galago Pros for two of my users. Very nice machines overall, though the trackpad button does feel a little cheap, and I don't care for the placement of the power button on the left edge of the laptop. Love the huge amount of connectivity, too, though I wonder how the funky spring-loaded Ethernet port will hold up over time.

Overall, their stuff keeps getting better and better... definitely worth a look!
 


Ric Ford

MacInTouch
Thinking about Apple's promised 2019 Mac Pro, I started imagining a modular design that would be tied together via short, hard Thunderbolt 3 connections. Shortly afterwards (not having discussed anything about this idea), a friend sent me this remarkable story of a similar Apple design from decades ago. (Caveat: dates given in the article seem inconsistent/erroneous.)
Stories of Apple said:
The Jonathan Computer
... Inspired by the Apple II “open” architecture mindset, Fitch proposed a modular approach. He designed a simple hardware “backbone” carrying basic operations and I/O on which the user could add a series of “book” modules, carrying hardware for running Apple II, Mac, UNIX and DOS software, plus other modules with disk drives or networking capabilities.

Thus beginners, mid-level and high-end customers could all use the same basic hardware but could configure and enhance their systems over time.

Apple could manufacture at low cost and publish (or license) the specs to the “spine” and ensure wide acceptance. Most importantly, Jonathan could offer both Macintosh and DOS compatibility, giving PC users an opportunity to “cross over”, sample the Macintosh and stay with it, or use both on the same system.

Fitch named his concept “Jonathan” and asked Helmut Esslinger, Guido and the frogdesign team to help him design a physical model to present to management. Hesslinger was initially skeptical but later adopted and implemented the “books on a shelf” metaphor, understanding that, unlike conventional computers, Jonathan could look not only strikingly different but also more impressive as its performance increased.
 


Ric Ford

MacInTouch
Here's a little update on Apple's secret car project. (You can do your own math to estimate how much money Apple is pouring into this vs. its existing products and services.)
Reuters said:
Apple self-driving car layoffs give hints to division's direction
... While the iPhone maker has acknowledged its interest in self-driving cars in broad terms, it has never detailed precisely which technologies it is working on and whether it seeks to build a whole vehicle or the sensors, computer system and software to control one.

The public documents filed with regulators provide some previously undisclosed clues.
Among those laid off were at least two dozen software engineers, including a machine learning engineer, and 40 hardware engineers, according to a letter sent by Apple to California employment regulators earlier this month.

... Apple operates the car project on a “need-to-know” basis, with only about 5,000 of Apple’s 140,000 full-time workers included, according to court documents in a theft of trade secrets criminal case filed this year against an ex-Apple employee.
 


Ric Ford

MacInTouch
... and now there's this amazingly innovative camera system from Nokia...
Below is a hands-on report about this radically innovative smartphone camera, complete with sample images. The Nokia 9 PureView is due next week for purchasing, with a $100 discount from March 3 to March 11, but its production will be limited.
GSM Arena said:
Nokia 9 PureView hands-on review

... All the cameras lenses have the same fixed focal length of 28mm. This camera doesn't have a wide angle camera, it doesn't have telephoto cameras, and it doesn't have any fancy shooting modes - this is a bit of a turn off as people expected the phone to have a set of lenses of different lengths - not unlike the Light L16 camera.

Instead, all five cameras are 12MP each and have f/1.8 aperture lenses: two of the cameras are RGB and the other three are monochrome. The images from all five (sometimes even a few photos per camera) are all stacked to produce a single image with a spectacular dynamic range - up to 12.4 stops of difference in light which is as much as a real large sensor camera.

Aside from letting 2.8x more light in due to the lack of a Bayer filter, the job of the B&W sensors is also to allow for a true monochromatic shooting mode rather than a mere desaturation of already color images.

The reason Nokia decided against telephoto and wide-angle lenses is because it allows to keep a slim profile of only 8mm for the phone without any camera bump. They also want photographers to treat this smartphone as if it was a real camera with a fixed prime lens. A prime lens is one that offers great image quality but a fixed focal length so there is no zooming in or out - to get the desired composition, you either move closer or further away.
 


Ric Ford

MacInTouch
iOS and macOS have always sat on the same foundations, but some API design issues have meant far more work porting an app from iOS to macOS than you might expect. Marzipan is an effort to clean that up and minimise the amount of OS-specific coding/tweaking. But it's still a porting exercise, not simply a case of 'build once, run everywhere.'
So, is that different from Qt, for example?
The Qt Company said:
Qt Documentation
Using Qt, you can write GUI applications once and deploy them across desktop, mobile and embedded operating systems without rewriting the source code.
 


Thinking about Apple's promised 2019 Mac Pro, I started imagining a modular design that would be tied together via short, hard Thunderbolt 3 connections. Shortly afterwards (not having discussed anything about this idea), a friend sent me this remarkable story of a similar Apple design from decades ago. (Caveat: dates given in the article seem inconsistent/erroneous.)
For 33+ years I've been writing for our local El Paso Mac User Group newsletter. I came up with the idea of using a mercury switch for a rotating display a year before Radius released theirs. (I didn't patent it, so my bad.)

Prior to the 2013 Mac Pro's introduction, I wrote about my idea for the next Mac Pro. Essentially, it would be a larger, stackable Mac Mini with a beefy GPU; more modules would provide additional CPU/GPU/Storage/RAM. The "interconnection" method would have been some sort of relatively direct pipe that Apple would certainly be able to accomplish. In this manner, Apple can sell one or two basic modules to which the user may add additional modules.

If the interconnection pipe is wide/fast enough, how many of these modules might be stacked? Updating this for 2019, maybe four modules - each with, let's say, a Core i7 - would provide 16 cores, four video cards (or one connection with four times the power?), 8 TB of SSD, and maybe 512 GB of RAM. Sell the module for, oh, $1600? Let's call it Apple's Pro-level NUC. But it's the interconnectivity of the modules that will provide the difference.
 


So, is that different from Qt, for example?
With Qt, you write once and get the same app on multiple platforms. I assume that you end up with a mediocre interface for everything, which is noticeably not native. For the mythical Marzipan, I think its goal is to make it easier for someone who writes iOS apps to rewrite them for the macOS. For media apps, such as NetFlix, I think very little work would be needed on the Mac side. For apps that are not full screen apps, we are probably looking at more work, but not as much work as using the current macOS APIs.
 


I believe that "Marzipan" will be the way that Apple gets rid of the Mac.

It will make it simpler/easier for iOS-centric developers to write for/port to macOS, and since the development tools will need to cater to the "least common denominator" (iOS), it means eventually anything macOS-specific will no longer be supported/eventually disappear.

So Federighi was right when he said "No" to the macOS-iOS merge question, since the answer is likely that macOS will slowly get deprecated/disappear altogether.
 


So, is that different from Qt, for example?
Yes and no. We use Qt for cross-platform development but targeting macOS and Windows (not iOS). It enables you to write code that works on multiple platforms, but we still have to make separate builds for each platform, and there are some platform-specific tweaks (particularly for the UI).

The objective for both Marzipan and Qt is essentially the same – minimise the coding specific for each platform. Qt is trying to make it possible to do this with no code differences, but I don't believe you will achieve an entirely satisfactory result that way. I expect the divergence would be larger if we wanted to target iOS, too.

The difference with Marzipan is that this is entirely Apple's baby and, consequently, they are very focussed on making their platform look good and work well. I also believe this is largely focussed on getting apps developed on iOS to easily run in macOS, rather than making both platforms developmentally identical. There are many small apps on iOS that could rapidly be ported to macOS. They're all functionally tiny compared to most macOS applications, so porting has to be quick/easy to be viable. It brings a lot of useful functionality cheaply and quickly to macOS without harming the mainstream applications.

I doubt that Marzipan will have much impact on more comprehensive applications. They require an almost total redesign on the UI side, and the stuff 'under the bonnet' is usually more portable anyway. Bottom line is that it's essential to observe certain constraints in the mobile space (e.g. low power consumption, limited connectivity, limited expansion, appliance-like behaviour), which clearly separates it from the freedom and power macOS brings to professionals. No one believes a 'compatibility layer' will bridge that divide.
 


I believe that "Marzipan" will be the way that Apple gets rid of the Mac. It will make it simpler/easier for iOS-centric developers to write for/port to macOS, and since the development tools will need to cater to the "least common denominator" (iOS), it means eventually anything macOS-specific will no longer be supported/eventually disappear. So Federighi was right when he said "No" to the macOS-iOS merge question, since the answer is likely that macOS will slowly get deprecated/disappear altogether.
I don't think I'd agree with this prediction at all. I believe that the collection of things you can do that work usefully on a machine plugged into the wall with a few terabytes of connected storage and a big screen or three are different from the universe of things you can usefully do on a small handheld.

Thus, I would much prefer to say that, for things which make sense on both classes of platform, Marzipan will make it easier to write once, deploy twice. But a Mac can do more things than a handheld iOS device can, so there will be Mac-specific capabilities and API's.
 


Bluebeam announced on Thursday 28 February that they will cease development on Revu for Mac as of March 2020. Not sure about others, but this is a great product (in my opinion) that I suspect went under the radar for too many Mac users, though fairly well known in the AEC community. Revu was reasonably priced with many features.
We’re writing to let you know that we’ve made the difficult decision to stop developing future versions of Bluebeam Revu for Mac. We understand that this decision will impact many users and organizations, and we’re committed to providing the support and resources required to keep you and your company on track during this transition. We’ll also release the Revu for Mac 2.1 update in Q2. This update includes improvements to our measurement tools and addresses some critical issues.

When we originally developed Revu for Mac, our intention was to extend the incredible efficiency and collaboration power of Bluebeam Revu to the Mac platform. While we pursued this multi-platform approach, advancements in cloud-computing redefined what was possible and provided an opportunity for us to reevaluate our strategy.

At the end of 2018, we committed to building a new cloud-based ecosystem of solutions that delivers powerful features, tools and workflows accessible from any device and from any location. You will first see this vision coming to life in new products like Bluebeam Atlas and features like Drawings. In order to invest in this long-term vision, we have made the decision to begin the process of winding down the Revu for Mac product.
 


I believe that "Marzipan" will be the way that Apple gets rid of the Mac.

It will make it simpler/easier for iOS-centric developers to write for/port to macOS, and since the development tools will need to cater to the "least common denominator" (iOS), it means eventually anything macOS-specific will no longer be supported/eventually disappear.
I believe [that] ignores a major problem the Mac has right now, which is people not writing software for the Mac. Of the 4 major platforms, iOS, Android, Windows, and macOS, macOS is the weakest. It might be worse when you consider the web browser as a platform - too many systems are Chrome-only. Thankfully, iOS is keeping Safari's presence on the web relevant.

Hopefully, Marzipan will result in apps that would never appear on macOS arrive. Truth is, many companies don't feel that macOS is worth a simple recompile, never mind any minor adjusting required to get their app working on the Mac. Marco Arment, who never considered porting his Overcast to the Mac, and he is a Mac fan, is now excited about the prospect. And perhaps we will see a real NetFlix app for the Mac instead of having to resort to a web browser.

If all goes well, we will see a bunch of new shovelware on the Mac, but also good products that would never have made it to the Mac. Looking back at Mac history going back to 1984, there were always many programs that were crappy DOS and Windows ports. The market has usually rejected these apps. The biggest question is what will Mac users accept. Unfortunately, Mac users are less demanding these days, partly because so much of the stuff out there is web apps, Chromium apps and Electron apps. Marzipan might actually raise the bar [vs. those].
 


For me, the thing that will signal that macOS is going away will be when Apple releases a version of XCode that runs on some other OS. You've got to develop your iOS apps somewhere, and right now Apple would really prefer that you buy a Mac in order to do so. If XCode is released on Linux or Windows, that will be the signal that macOS's days are numbered.
 


Ric Ford

MacInTouch
For me, the thing that will signal that macOS is going away will be when Apple releases a version of XCode that runs on some other OS....
Ditto for Final Cut Pro X and Logic Pro X. Of course, FileMaker Pro and FileMaker Server already run on Windows and Windows Server, and FileMaker Cloud runs on Amazon AWS, and iTunes runs on Windows, and Windows runs on Macs, and Apple Music runs on Android, and iCloud supports Windows, and there's QuickTime for Windows, and AirPort Utility runs on Windows, but at least Apple's not supplying Safari for Windows any more....
 


Ric Ford

MacInTouch
Of course, FileMaker Pro and FileMaker Server already run on Windows and Windows Server, and FileMaker Cloud runs on Amazon AWS...
And Apple's FileMaker subsidiary just changed executives:
Apple/FileMaker said:
A welcome from FileMaker’s new CEO

... Dominique Goupil has led this company for twenty years. As he steps down into retirement, he leaves a legacy of high product quality and personal integrity standards. I’m grateful to everything Dominique has done for the company, and I’m looking forward to building on this foundation.

It’s an exciting time for FileMaker. We’re seeing top and bottom-line growth. Last year, we introduced the Workplace Innovation Platform category. We are accelerating our strategy and will continue to invest in our brand to drive this forward, including considering potential acquisitions.

... As I step into the role of CEO, I look forward to hearing more from our community and working closely with our partners and customers. We have a great journey ahead of us.

Sincerely,
Brad Freitag
CEO
FileMaker, Inc. - an Apple Subsidiary
 


For me, the thing that will signal that macOS is going away will be when Apple releases a version of XCode that runs on some other OS. You've got to develop your iOS apps somewhere, and right now Apple would really prefer that you buy a Mac in order to do so. If XCode is released on Linux or Windows, that will be the signal that macOS's days are numbered.
The way I see the future of 'Xcode' development is this (note - I do not have any insider insights as to what might happen, this is just my own speculation based on my observations as a 10+ year Mac/iOS developer):

Someday, there will be an 'Xcode' for iOS that will allow developers to work entirely on iOS (Swift Playgrounds is already starting to get us there).

But I don't see it stopping at just 'Xcode for iPad' - Apple could very likely introduce its own 'build-code-as-a-service' thing, which would allow developers to easily develop on an iPad locally, but the actual compiling, packaging and distribution of binaries would be all handled by Apple, along with anything Xcode Server currently provides for CI (Continuous Integration), etc. (for which there are already many services out there). This, to boot, would give Apple even more control over app distribution, as well.

And once you don't need a Mac to develop software, well, "there's no reason anyone would need a Mac anymore, right?". And if no one needs a Mac, why bother supporting any Mac-specific API/features?

Just call me cautiously pessimistic, I guess.
 


Thus, I would much prefer to say that, for things which make sense on both classes of platform, Marzipan will make it easier to write once, deploy twice. But a Mac can do more things than a handheld iOS device can, so there will be Mac-specific capabilities and API's.
That's assuming Apple still wants to be building Macs two, three, maybe five years down the line, which to me seems unlikely, given its lackluster releases (downgraded Mac Mini, trashcan Mac Pro, "What's a computer", etc.) It seems Apple is only releasing Macs to bide their time and keep current Mac users from being too loud.

(See my other response for more context.)
 


Amazon disclaimer:
As an Amazon Associate I earn from qualifying purchases.

Latest posts