Apple has announced that it will be delaying the controversial technologies that tried to match known illegal CSAM images uploaded to iCloud Photos and scanned for explicit pictures sent or received by minors in Messages. Developer complaints are at long last coming home to roost, as regulators and lawsuits target Apple’s App Store policies—we explain how Apple is being forced to adapt. Finally, Glenn Fleishman describes how and why he upgraded from his ancient version of Quicken 2007 to the current Quicken Deluxe. Notable Mac app releases this week include Acorn 7.1, BBEdit 14.0.1, Fission 2.7.1, and MarsEdit 4.5.1.
In a statement released to various media organizations, Apple said it would be delaying the launch of its CSAM detection features, previously slated for inclusion in iOS 15, iPadOS 15, and macOS 12 Monterey:
Last month we announced plans for features intended to help protect children from predators who use communication tools to recruit and exploit them, and limit the spread of Child Sexual Abuse Material. Based on feedback from customers, advocacy groups, researchers and others, we have decided to take additional time over the coming months to collect input and make improvements before releasing these critically important child safety features.
We covered the initial announcement in “FAQ about Apple’s Expanded Protections for Children” (7 August 2021) and Apple’s subsequent mea culpa and explanations in “New CSAM Detection Details Emerge Following Craig Federighi Interview” (13 August 2021).
As I noted in the second article, Apple thoroughly botched the initial announcement and follow-up, so it’s unsurprising to see the company delay in order to collect input and improve the technology. The question is why it took so long for Apple to do so—the company has plenty of experience with removing technologies from forthcoming upgrades, as with the recent delay for SharePlay.
My guess is that Apple plans to refine both the technology and its messaging over the next few months and reintroduce it sometime next year. I would hope that announcement comes with a much better story about how effective the technology will be in reducing the spread of CSAM, preventing revictimization, and resisting government-level subversion. That may help the media do a better job of analyzing and reporting the second time around—even if Apple bungled the initial announcement, much of the overheated media coverage was based on inaccuracies and misunderstandings about how Apple said the technology would work.
The other possibility is that the entire effort is now tainted, making this “delay” just a face-saving way for Apple to drop the technology like the hot potato it became. Would there be a massive public outcry if 2022’s Worldwide Developer Conference came and went with no mention of CSAM detection in iOS 16?
Courts and regulators have been slowly chipping away at the bricks protecting Apple’s walled garden. We recently covered a new South Korean law that lets developers in the App Store and Google’s Play Store offer alternative payment methods (see “New South Korean Law Mandates Alternative App Store Payment Options,” 31 August 2021).
But that law applies only to apps sold in South Korea, and we would be surprised to see Apple voluntarily expand that program to other countries. However, two recent settlements have global implications.
Settlement in US Developer Class-Action Suit
Apple announced changes to the App Store as part of settling the Cameron et al v. Apple Inc. lawsuit, but the impact seems limited. From what we could glean from Apple’s announcement, much of the settlement involves Apple continuing to do what it’s already doing, such as agreeing to maintain the App Store Small Business Program for at least three years, even though there was no previous indication it would ever end (see “Apple Drops App Store Commission to 15% for Small Developers,” 18 November 2020).
The only two changes that seem notable, if not earth-shaking, are developers being allowed to notify their users about alternatives to in-app purchases in email (but not in apps themselves), and Apple expanding the number of price points from fewer than 100 to more than 500. So you might see apps priced at $1.49 instead of $0.99.
We don’t know how best to interpret this settlement because we have no skin in the game. For a roundup of responses from well-known Apple developers, we recommend reading Michael Tsai’s blog post. In short, they aren’t impressed, which casts a shadow on Apple characterizing the settlement as with “US developers” rather than “the lawyers for a set of developers who don’t seem to represent the opinions of the larger Apple development community.”
Even if Apple’s concessions are minimal, they, along with the South Korean law, show both how much pressure Apple is under and that Apple will respond to developer concerns when necessary.
Settlement with Japan Fair Trade Commission
Apple’s settlement with the Japan Fair Trade Commission (JFTC) goes further. In a small but significant change, Apple says it will allow developers of “reader” apps to include in-app links to their websites so users can create and manage their accounts. For instance, the Netflix app could have a link to the Netflix website. What a concept! The change will go into effect in early 2022 after Apple updates App Store guidelines and its review process to accommodate the change.
This change affects one of Apple’s most-loathed policies, the so-called “anti-steering” rule that prevents developers from providing links to external payment systems in an attempt to force them to use Apple’s in-app purchase system. Allowing reader apps to let users link out to account management systems seems like a small and obvious option, but Apple has thus far been steadfast in rejecting apps that present such external links.
What exactly is a “reader” app? In theory, that category includes content-consumption apps like Kindle, Spotify, and Netflix. However, as I pointed out in “Developers v. Apple: Outlining Complaints about the App Store” (13 August 2020), Apple’s definition of a “reader app” is arbitrary. Here’s how Apple defines them in its App Store guidelines:
Apps may allow a user to access previously purchased content or content subscriptions (specifically: magazines, newspapers, books, audio, music, and video). Reader apps may offer account creation for free tiers, and account management functionality for existing customers.
But Apple also apparently once counted Dropbox as a reader app since the only option when you opened the app was a login. It now offers various sign-up options. In any case, we expect Apple’s definition to change or at least become more concrete. The question is if Apple will have to get the JFTC to sign off on the new definition or if it’s allowed to make such decisions arbitrarily. JFTC’s statement doesn’t clarify that, but Apple agreed to a review once per year for three years.
On Daring Fireball, John Gruber shared a couple of anonymous anecdotal accounts of developers who don’t trust Apple’s new linking policy and suspect there are strings attached, such as being required to also offer subscriptions as in-app purchases or fearing Apple will invent new reasons to reject their apps for links. It shows how much trust Apple has squandered with developers.
There is also the question of how this intersects with Apple’s program for “premium subscription video entertainment providers,” which allows users of the Amazon Prime Video app to make content purchases within the app without using Apple’s in-app purchase system (see “You Can Now Make Purchases in the Amazon Prime Video iOS and Apple TV Apps,” 3 April 2020). A handful of other video services are included, but not the likes of Disney+, Hulu, and Netflix.
The new linking rule won’t apply to games or in-app purchases, thus doing nothing to resolve Apple’s legal battle with Epic Games, which has exposed to the public a treasure trove of internal emails that reveal Apple’s thinking on many of these matters.
More Bricks at Risk
We suspect Apple will be forced to make additional concessions in the future. India has opened an antitrust case against Apple, targeting the company for forcing developers to use its in-app purchase system. Apple is also fighting a similar battle in the European Union.
Meanwhile, the bipartisan Open App Markets Act proposed in the United States Senate would go even further, allowing third-party app stores and sideloading, letting developers use alternative payment methods, and allowing developers to communicate openly with customers. We have no idea of its prospects for passage, but it’s unusual when Tennessee Republican Marsha Blackburn and Minnesota Democrat Amy Klobuchar agree on something.
Apple sees where things are going and is taking independent actions aimed at staving off regulation, such as its just-announced News Partner Program, which halves the subscription commission rate to 15% for publishers who offer their content in the Apple News Format. And last year, Apple offered a big concession with its App Store Small Business Program, which dropped the sales commission to 15% for most developers (see “Apple Drops App Store Commission to 15% for Small Developers,” 18 November 2020). But such moves might be too little, too late.
At this point, Apple might be best served by making a dramatic overhaul to its most controversial App Store policies and behaviors, such as allowing alternative payment methods within apps, dropping App Store ads (are they really worth it?), eliminating special deals with big developers, and overhauling the review process to better catch counterfeit apps and not unfairly punish legitimate developers. Apple could also allow streaming game services and emulators into the App Store. iDOS was one of the neatest iPad apps in years, but after initially approving it, Apple changed its mind and pulled it.
Then Apple might have a stronger position from which to resist clunky and likely misinformed regulation that may have unintended consequences. And perhaps then developers would lay down their pitchforks and torches.
I couldn’t remember when I first installed Quicken to manage my personal and small-business finances. It seems like I’ve used it forever. But after a recent long-delayed update to the current version, Quicken Deluxe, I can now pinpoint the time: mid-1998.
After 23 years of using older versions of Quicken, including Quicken 2007 since its paradoxical release in 2006, I’ve finally severed my cord to the past: to PowerPC and Intel processors, to outdated record formats, and to clunky manual entry of transactions. I waited so long, however, that I wasn’t forced to give up anything in the process.
This is my journey, and if you’re still on Quicken 2007, it’s one you can take too.
Quicken 2007 Was Old When It Was New
When Intuit, Quicken’s original owner, released Quicken 2007 for what was then called Mac OS X, it was a major step forward for Mac users in synchronizing online accounts and producing financial reports. Quicken 2007 included and improved on several bookkeeping needs I had and still have:
- Transaction-based entry
- Hierarchical categories (like “Business:Hardware:In-state” to mark purchases on which I paid sales tax, or “Income:Consulting:Conferences”)
- Financial institution transaction syncing and automatic (or at least heavily assisted) reconciliation
- Highly customizable reporting to extract lists of transactions or summaries of income and expenses corresponding to city, state, and federal tax guidelines
- Separating personal and business transactions by using categories to assign any transaction to any purpose
But Quicken 2007 was already behind the times in 2006! Apple had already begun its transition from PowerPC to Intel chips, and Quicken 2007 had only PowerPC code in it. Intuit never released a true Intel-compatible version of its flagship software but somehow pushed out a hacked-together version that sidestepped the need for Rosetta (see “Intuit Releases Quicken Mac 2007 OS X Lion Compatible,” 8 March 2012). After years of excuses, the company released an entirely new app, Quicken Essentials, that was so stripped down as to be essentially… useless.
Intuit’s attempts to reimplement the full capabilities of Quicken 2007 never quite hit the mark; see “Quicken 2015: Close, But Not Yet Acceptable” (2 October 2014). I regularly updated that “2015” app version and even bought some upgrades to see if Intuit had added what I needed.
At various times from 2014 to the present, I also tried Banktivity, Mint, Moneydance, Xero, and others. None met my requirements, though many came close. Close wasn’t good enough: if I was going to shift away from a two-decade-old system, I wasn’t willing to give up functionality.
Through macOS 10.14 Mojave and the end of 32-bit apps, Quicken 2007 limped along with a surprising series of minor updates that allowed it to continue to function mostly unimpaired.
And that’s how I found myself in early 2021 with the digital equivalent of a bulging shoebox of old receipts, unsure of how to move forward in a way that would save me time and frustration. Fortunately, Quicken was a step ahead of me—and a step ahead of where I thought it was.
Lost in the Desert, but an Oasis Looms
I continued to run Mojave on my 2017 27-inch iMac even through the release of macOS 11 Big Sur for a few reasons. Two of those were Quicken 2007 and Mailsmith, a text-only email program I’ve used for 20 years. A third was the poor performance of the iMac’s Fusion drive, which I eventually fixed with an external SSD (see “An External SSD Gave My iMac a New Lease on Life,” 9 April 2021).
Once I added the SSD and upgraded to Big Sur, I had the performance necessary to run Mojave within a Parallels Desktop virtual machine. That let me keep Quicken 2007 and Mailsmith running while I sought replacements.
Before I settled on any, my iMac bit the dust in the most expensive way possible. A local shop traced the problem to a failed motherboard. At a cost of several hundred dollars, it made no sense to replace, so I opted to purchase an M1-based Mac mini to act as my anchored office computer. (The M1-based 24-inch iMac didn’t impress me enough, and I wanted two 27-inch monitors.)
The move to M1 left me in a crisis. How could I keep running Quicken 2007? I was in the middle of an overwhelmingly busy month, so I punted and set up a virtual Mac. I chose MacStadium, a co-location service for people who need remote Macs, which was offering a first-month discount. (There are other virtual Mac options, including some at Amazon Web Services, but none are priced for “using a single app for a few hours a month.”)
In an effort to find a new-to-me Mac for Quicken 2007, I searched eBay for a 2012-era Intel-based Mac mini that could run Mojave, was inexpensive enough to make sense to purchase, and wasn’t a total slug. I found and purchased a 2012 model with 4 GB of RAM and a 500 GB hard drive. In its stock configuration, it was terribly slow, but once I installed Mojave on an unused 256 GB USB 3.0 SSD I had on hand, it turned into a pretty zippy single-purpose Quicken 2007 appliance. After a quick configuration via a monitor, I set it to wake on network access and used exclusively via screen sharing.
To keep this Quicken 2007 appliance out of the way, I used a Humancentric Mac Mini Mount ($17.99) to attach it to the underside of my desk. (The mount has screw holes spaced for VESA mounting, so you can attach a Mac mini to the back of a monitor, too!)
After posting a photo of my hidden Mac mini, an online acquaintance said I should check out the latest Quicken Deluxe. A few years ago, Intuit sold Quicken to a private equity group, H.I.G. Capital. Instead of destroying the product, the firm breathed new life into it, seemingly by shifting to a software subscription model.
Quicken Deluxe costs $51.99 per year, but search for discounts—I paid $34.39 for my first year by clicking through a search-engine ad. There’s no free trial, but the company offers a “30-day money-back satisfaction guarantee.” I made the purchase, installed the software, tested it out quickly, and found to my chagrin that Quicken Deluxe finally had everything I needed. Sadly for all my extra effort, it may have had it as far back as 2019! I wish I had been paying more attention.
My next task was to convert my old data and bring myself up to speed on the changes.
The (Quicken) Deluxe Route to Big Sur and Beyond
First off, I wanted to import my Quicken 2007 data. I was resigned to the possibility that I wouldn’t be able to convert a 14-year-old data format with 20-plus years of data, but I held out hope that it would work. Quicken Deluxe requires a round trip through a cloud server for Quicken 2007 format conversions, which it warns you about since your data will leave your computer.
It took several failures before I sorted out what was going wrong because Quicken Deluxe’s errors didn’t help. Quicken 2007 can run on a macOS system that has an HFS+ or APFS startup volume, and its data file can be stored on either kind of filesystem. But you cannot correctly copy the Quicken Data file from an APFS volume. (It seems to use some outdated package format that relies on HFS+.)
After finding this answer online in a forum, I temporarily installed Quicken Deluxe on the 2012 Mac mini volume on which I had Quicken 2007. That let me avoid copying the data files from APFS and start the conversion.
It took about half an hour to convert and import tens of thousands of transactions, but it succeeded. In Quicken Deluxe running on my M1-based Mac mini, I can pull reports dating back to 1998, and all my accounts and categories remain intact. I lost only Quicken 2007 custom reports, which I can’t complain about when everything else came through.
The next task was setting up online synchronization of accounts. Banks, credit card companies, and other institutions had maintained support for Quicken 2007 for many years, but synchronization and export features began to fall away around 2015. I had shifted to entering transactions manually—a horrible pain, but I had no alternative.
I spent nearly two hours working through all my banking, debit, credit, investment, and retirement accounts until everything synced correctly with Quicken Deluxe. The app offers a few ways to set up synchronization; I started with Quicken Connect (also known as Express Web Connect), which uses a secured login. This worked in all cases. Another option, Direct Connect, may require paying a financial institution a monthly fee, but then you can use Quicken Deluxe for direct bill paying, too.
Of all these accounts, only a Citi credit card failed to match existing transactions and duplicated the last year’s worth of downloads. This was an easy problem to fix because Quicken Deluxe tags the source of transactions in a column that can be sorted—downloaded ones use a little green ball. I was thus able to delete those duplicate downloads quickly. I had a few other tweaks to make in other accounts, where I used Quicken 2007’s split function to break up one transaction into multiple categories or transfers that the downloaded financial data didn’t match, but it was quick to resolve.
One bump I hit applies to any transaction-tracking app that performs these syncs. Some institutions require a two-factor authentication code every time you sync rather than occasionally or after you validate a particular computer or network. I had three accounts in that category: PayPal (separate personal and business accounts) and a Bank of America credit card. Quicken Deluxe handles this modally, so it’s a tedious process. The app tries to log in, recognizes it needs a code, asks you for the code, you enter the code, it validates it, repeat, repeat. Because it’s modal, it winds up stealing a minute or more. To avoid this wasted time, I disabled automatic sync on these three accounts because they have relatively little activity and it’s easy enough to trigger them to sync manually.
Finally, I began to re-create my needed reports. Quicken Deluxe has a lot of reporting power—much more than Quicken 2007, unsurprisingly—but the developers could improve a few aspects. Tweaking certain settings renames the report, which is generally unwelcome. And performing an operation across the most recent year’s worth of transactions—several hundred in my case—takes far too long for something that’s clearly database-driven. I should have to wait seconds, not a minute, particularly when I’m configuring a report and every change incurs the same computational churn time.
But it was all worth it. Just two weeks into using Quicken Deluxe, I’ve already saved a couple of hours keeping up to date. Over the course of a year, I’m somewhat embarrassed to admit, I might save a cumulative 40 to 80 hours. The Quicken 2007 time hit was particularly severe for quarterly and annual local tax filings and assembling material for our accountant for federal taxes. Quicken Deluxe’s better reporting will let me pull a precise report instead of multiple reports I had to consolidate through a spreadsheet.
Will the roughly $50 I pay every year for the rest of my Quicken-using life—plus regular bumps up in that price—be worth it? Absolutely—saving even an hour or two a year would. Saving a week or two of time I could spend earning money? Sign me up! Oh wait, I already am.
Now if only someone wanted a 2012 Mac mini suitable for under-desk mounting. I’ll even throw in the external SSD!