As good as Wolf's points are, it doesn't change the fact that more and more Mac developers are starting to shift to a MAS-preferred market.
At this stage, even though I would rather buy most of my apps direct from developers, I'm often half-afraid to do that because of the inevitable "what happens when they shift to MAS-only" question arises.
It's already happened with 1Password, Coda 2, Pixelmator and a number of my other favorite Mac apps -- it's hard, as an end-user, to support the old regime when maintaining two separate software branches continues to be a PITA for developers.
And that's the real problem. Developers retain two branches because they want to offer customers options (as well as legacy support for regular pre-MAS releases), but most of the Mac developers who are both selling direct and in the MAS are seeing much more income from MAS. If forced to choose, I can't imagine a developer picking the option that gets in front of fewer customers.
Still, there are some very, very real concerns for Mac software due to the new Sandboxing rules. Fortunately, now that the deadline has passed, we'll actually get to see just how bad it is.
1Password and Coda are both sold direct — what do you mean by "it's already happened"? A developer big enough to set up a direct payment channel is likely to see it pay for itself through bulk sales and site licenses even if that's the only reason they keep it around.
I'd also dispute your last point that we'll be able to clearly see the effects — despite the continued ability to sell direct (Developer ID looks like a great thing), sandboxing as Apple's chosen to implement it is going to result in fewer apps, not merely worse ones.
Hey all! Friendly neighborhood AgileBits Support personnel here.
This was previously true. However, we've not determined that for sure at this point. Users spoke, we listened. 1Password 4 is still a ways away and we'll determine whether it goes MAS exclusive sometime closer to release.
That said, nothing is ever set in stone with AgileBits. We do after all have Agile in our company name. :)
Our goal is to try to make our customers happy. We did make the MAS version cheaper upon introduction for users to make the transition a little easier. The pricing was roughly equivalent to what an upgrade would cost from our direct store.
We would also love to see a way to transition users to the App Store from our Direct Sales, but as you all know, Apple doesn't provide a mechanism for this.
We do always welcome users to comment and offer their comments and suggestions. This isn't the place to do that, but feel free to voice your opinions and suggestions (politely please, we are human after all) at our support site: http://support.agilebits.com
I'll do my best to follow up here, but due to a busy day here I may be a bit slow. Best to contact our support site directly where on duty support can assist you more quickly.
Apple should provide the ability to import "legacy users" (Cydia offers this feature, for the record, and it has come in very handy and made the decision to support it much easier for both developers and customers, so I'm not just talking out of my ass without knowing the ramifications of having such).
I don't disagree, but how do you define and identify "legacy users"? Most importantly from Apple's perspective, how do they properly identify "legacy users" who purchased apps from a vendor's site before the App Store launched, vs. users who purchased apps from a vendor's site after the App Store launched.
If Apple allowed vendors to import any user as a "legacy user", then that's a clear and supported way for vendors to get around Apple's 30% cut.
As has already been discussed ad nauseum, Apple is not bein irritating to get a "30% cut" (which is a bothersome way of describing it as they don't den get 30%: they get what is left over after paying you your 70% cut and then handling their numerous fixed costs with relation to selling your products): they want control over distribution. For more information, read this entire thread.
As for how you do it, in practice 99% of licensing schemes are based either on "serial number", "mac address", "license key given to user and typed into app", or a combination of such. While some people then can't import still, almost everyone can, and that is enough to be game-changing.
Note: you don't even need to allow this list to e changed after you import the product if you are still concerned about a "30% cut". (Though, Cydia does allow this, and even fully automates it by way of a backend licensing API vendors can implement to take control of the purchasing process, and in practice vendors that start using this mostly use it as a temporary measure and eventually just switch over entirely).
Apple doesn't care about the distribution of apps. Gatekeeper makes that abundantly clear. Please present a cogent argument that takes that under consideration.
Apple, however, doesn't want to be on the hook for maintaining and supporting and spending bandwidth on users who haven't given them a cent. I can't blame them. Again, I'd love to be granted the ability to bless "legacy users" into the Mac App Store, but I also don't understand why Apple has any incentive to ever allow me to do so. It costs them money and resources, with absolutely no benefit in return.
"Apple should..." Ha! Apple, as a whole, doesn't care one whit about the way people used to buy Mac software. The App Store, both iOS and Mac, doesn't allow for any sort of nuanced models like timed trials, paid upgrades, package discounts, student discounts, etc. etc.
I think this is a overreaction to unregulated markets where merchants are constantly playing pricing games. As a consumer, it feels unfair when different people pay different prices depending on what hoops they've jumped through. However, Apple's approach feels unfair to me too, particularly the lack of trials and paid upgrades. I feel that this pushes everything towards "disposable" software rather than products that could grow over many revisions supported by upgrade from their users.
Just a small note here, but Coda 2 is for sale on Panic's website just like previous versions. After some consideration I went with the MAS-option because of iCloud-integration. Being able to easily sync your sites from your computer to your iPad sounded like a nice thing.
We'll see if I'll come to regret it in the future, but it would be nice if Apple allows app-developers who get crippled by future restrictions to offer their users a licensecode for the non-MAS-version as a replacement.
Right you are -- I misspoke there. Like you, I opted MAS for Coda 2 for iCloud -- especially since I imagine that'll come in handy for Diet Coda in the future.
I am curious if there are any numbers comparing revenue from direct sales vs MAS sales.
I sell a Mac App exclusively via the MAS and have occasionally considered adding a direct sales option to our website, but the extra overhead of adding a secondary licensing scheme to the app, a web store front, processing payments, and dealing with sales support doesn't seem like it would be worth it.
Revenue numbers would be very interesting indeed. I can imagine (small) developers receive more purchases from the Mac App Store compared to their own website, since apps are very easy to find in the MAS.
I'm not willing to give out concrete numbers but, from my experience, that's absolutely the case. Our Mac App Store reveunes far eclipse our own store's revenues. I certainly wouldn't say maintaining and supporting our own store hasn't been worth it – we're able to offer deals and discounts to bulk and educational purchasers that we couldn't on the MAS, thus making us more appealing to those markets – but I'd give up selling on our own store before I ever considered giving up selling on the MAS.
For us, it's more in the 4:1 range, but I suspect you still might be right that many apps see a MAS:private sales ratio of 5+:1.
Because we're able to offer bulk and/or educational purchase discounts through our own store, we see a fair number of licenses move through that vector. Other apps might not appeal to bulk and educational markets as much.
"Fortunately Apple now only accepts sandboxed Mac apps, clarifying the situation: customers should buy Mac apps directly unless there’s a good reason not to."
This article seems heavily biased towards the app developer's interests. Is it really better/safer for everyday customers to use non-sandboxed apps?
Unless I'm mistaken, the two are not the same. There is a middle ground possible: a developer can publish their own sandboxed apps.
This would get you about half of his "buy non-sandboxed apps directly" advantages (More/Faster Updates, Less Risk, More Money), while still offering the user the advantages of the sandbox (e.g., I can see that a graphics program I downloaded only uses the entitlement "com.apple.security.files.user-selected.read-write" and thus can't open network connections behind my back).
Why? Because of exposure app store gives you app, and far more people would buy it from app store, as it doesn't require you to enter your credit card number in some random, no-name developer' website.
Many of the arguments provided by the article are straw men and ignore a provision that Apple offers (and encourages the use of!) – custom sandbox entitlements. Apple has repeatedly stated that custom entitlements can and will be granted for apps that need access not granted by existing sandbox entitlements. Certainly not all exception requests will be granted, but for some of the ones he listed – Growl, file read access for Omnifocus – those are perfectly legitimate exception requests.
Apple is working hard to provide a secure and still fully functional system. Some things will break, undoubtedly. Some apps will be abandoned, either because of legitimate technical limitations Apple refuses to address, or because of political stances taken by the apps' developer. Things aren't perfect right now, and Apple still needs to continue refining and enhancing the OS X sandbox.
Is it as clear-cut as Rentzsch makes it out to be, though? Absolutely not. In fact, I think he takes a very biased stance in this article. For the vast majority of users, sandboxing is an enormous step forward for the safety and security of their data and system. For that reason alone, I encourage everyone to purchase apps via the App Store. It's a safe and trusted distribution vector that is now helping improve OS X's platform security. That's an enormous benefit to almost all consumers.
(And I say all of this as a developer who's having to wrangle with the sandbox right now.)
They are temporary because they are granted for situations where Apple has not yet implemented an API or formalized an entitlement. Apple bends over backwards to be developer friendly, they are not going to pull the rug out just for the hell of it.
The new 'iOS' which is still surprisingly developer friendly, despite the attitude it receives on forums such as HN.
iOS, for all of it's failings and deficiencies, still offers one of the finest – if not the finest – development and distribution platforms in the world.
Yes, it's not open. Yes, it's restrictive. Yes, it's censored.
But, perhaps, those qualities are some qualities that make it the most successful software sales vector we've ever seen.
I agree it's not perfect. But to say it's not developer friendly is quite disingenuous. The iOS development toolchain is actually quite powerful and flexible (though still faulty), the distribution mechanism is broad and simple (though restrictive and censoring), and I can't think of a single other platform, including the web, that has been as developer friendly. At least for developers who wish to make an income. (Something about which I admit, without any bit of judgement, some developers aren't concerned.)
I'm sorry, but making people pay $100 to develop on systems they own is not developer-friendly. I would much rather they had NO apis and we were allowed to access the hardware we bought.
Oh yes, I get all warm and fuzzy every time I think of how friendly Apple is to developers... ummm, are you crazy?
How friendly is a company that uses secret APIs to compete with you? How friendly is a company that rips off your software and then kicks you out of their app store? Gimme a break...
I'm not convinced the Mac sandbox adds much security at all yet, for one big reason: do you know what entitlements a sandboxed app has?
There is a command-line tool to list an app's entitlements, but it's not at all easy to use, or even find. I haven't found a way to determine the entitlements of an App Store app without installing it (and hence, paying for it, if it's not free). I expect that basically no users will ever look at this. It's a bit like the device in Dr. Strangelove: the whole point is that you're supposed to tell people!
In a perfect world, Apple wouldn't approve an app for the App Store which did something nasty (like upload the user's address book to a spammer), but it's impossible for them to catch everything even if they never made a mistake. Besides, every entitlement has legitimate uses.
Security is helped by transparency. It would be great if users could see what sandboxed apps are allowed to do. They can't, and at worst this makes users feel safer without actually being safer.
And notably absent is the idea that there could ever be an advantage to sandboxing for the user. A glaring oversight that renders the one sided conclusion invalid.
Having recently suffered a hard drive failure, I have a new appreciation for the idea of being able to download and re-install my applications from a single location.
You have backups but apparently no feasible restores.
I have always an up-to-date image of my Macs created and updated each night with Carbon Copy Cloner (before that, I use SuperDuper!). In case of a problem with any of my Macs, I can simply boot from the image on another Mac and continue to work. If I cannot fix the problem, I copy the image to the respective Mac or a new Mac …
Buying direct has another important advantage if the buyer doesn't live in the US: Pricing!
In the Mac App Store, prices in non-USD currencies don't compare favorably with the original price in USD due to the exchange rates Apple uses. If you can still buy an app directly from the developer in USD, the price is usually less than in another currency. In many cases, the difference in pricing is substantial and the developer gets more from the paid prices anyway.
I get updates for my apps through the App Store, and more than just minor bug fixes...feature enhancements as well as new features and big version changes. I think it really depends on the app.
For what it's worth I have also bought Divvy directly so I don't know what you mean when you say you "won't receive updates" for it.
At this stage, even though I would rather buy most of my apps direct from developers, I'm often half-afraid to do that because of the inevitable "what happens when they shift to MAS-only" question arises.
It's already happened with 1Password, Coda 2, Pixelmator and a number of my other favorite Mac apps -- it's hard, as an end-user, to support the old regime when maintaining two separate software branches continues to be a PITA for developers.
And that's the real problem. Developers retain two branches because they want to offer customers options (as well as legacy support for regular pre-MAS releases), but most of the Mac developers who are both selling direct and in the MAS are seeing much more income from MAS. If forced to choose, I can't imagine a developer picking the option that gets in front of fewer customers.
Still, there are some very, very real concerns for Mac software due to the new Sandboxing rules. Fortunately, now that the deadline has passed, we'll actually get to see just how bad it is.