Call me pessimistic, but how likely is this acquisition by Apple going to result in them integrating it directly into iOS versus completely shutting down in the future?
Perhaps Apple is using the acquisition strategy as a method to shut down an app getting around iOS's "preferred workflow" and UX model. Sort of a strategy taken by Microsoft, Google, and Apple whereby they buy competitors, claim to leave the service running, but ultimately shut them down so it doesn't compete with their main product (see: Sunrise, Acompli, Sparrow, Timely, etc.)
We already know how Apple gets rid of things that do things on iOS that they didn't want people doing: they pull them from the store.
Apple didn't do that, in fact they've left it in the store and made it free. They've given it a design award in the past (which seems to tacitly endorse that they like parts of it), and we know they've had things to help with automation in some degree on the Mac for years (AppleScript, Automator). This may be one of those cases where they just saw a really good product and decided to start from it instead of building from scratch.
Not unlike what they did with MusicMatch Jukebox and a number of other things throughout their history. (iTunes was amazing. I'm not going to argue about the last 10 years or so, but when they bought it it was really nice software for many years).
I think they're trying to fight for the iPad Pro in some respect. Though they won't endorse programming on it, they will benefit from more automation systems on the iPad making it into a more viable alternative to the general purpose computer.
Given that Apple axed the position in charge of Mac automation tech (AppleScript, Automator, System Services, Apple Events, etc) last November, I can't help being pessimistic. Updates to products in the iWork suite repeatedly dumbed down their scripting support, and it's pretty clearly not been a priority for Apple in a long while.
Some background and commentary on Sal Soghoian's website. Until recently he was Apple's Product Manager of Automation Technologies: http://macosxautomation.com/about.html
"I hear you no longer work for Apple; is that true?
A. Correct. I joined Apple in January of 1997, almost twenty years ago, because of my profound belief that “the power of the computer should reside in the hands of the one using it.” That credo remains my truth to this day. Recently, I was informed that my position as Product Manager of Automation Technologies was eliminated for business reasons. Consequently, I am no longer employed by Apple Inc. But, I still believe my credo to be as true today as ever."
The quote implies that Apple left him in the dust after 20 years, but if you look at careful phrasing it leaves open that Apple offered him an alternative position.
I agree. I doubt it is as black and white as people have interpreted. I suspect it was more along the lines of "we want to take a different approach to _automation_, are you with us or not?"
What's so "wow" about this ? This happens every second of every day across companies. People are always made redundant for business reasons e.g. restructures, strategic realignments etc.
It's definitely not unique for Apple which recently has seen them do a purge of areas they no longer want to be in e.g. Displays, Wifi routers.
He spent 20 years building technology to empower users, and Apple decided that user empowerment is an an area they no longer want to be in. That's wow a new era in computing, that is all in on changing users into mere consumers
They didn't decide that. They decided this particular guy was not necessarily the guy to do that. You don't just go and buy the best automation app on the marketplace because you don't like automation - you do it to bring in new blood and a new perspective.
This is all just speculation. We don't know anything. But it's possible that there was a difference in opinion about what sort of user to target.
Courting power users used to be much more popular than it is today. It seems that today all large tech companies target IT professionals and consumers with a completely separate set of offerings. There's very little in between any more. On the consumer side they go for ultimate scale because nothing else moves the needle for them.
It wouldn't surprise me if Apple had decided to deprecate everything that isn't accessible to all consumers, which would have frustrated someone who has worked for decades to let power users automate tasks in ways that the average consumer never would.
As I said, all pure speculation for our entertainment :-)
True, but automation on the Mac is still a lot easier just because of the system design. In about sixteen years of using OS X (er, "macOS"), I've only used AppleScript for automation when forced, usually by programs like BBEdit that came of age in AppleScript's glory days. In most cases, writing a shell (or Python or Ruby or...) script is just easier. To be fair, AppleScript can be really powerful, but it's just not a very good language.
Also, it's worth noting that Workflow is, in a lot of ways, inspired by Automator on the Mac. At this point it's better than Automator is, albeit with all the caveats forced on it by the aforementioned differences in system design, but it wouldn't require an awful lot of effort to bring Automator up to par.
Things are a lot better now that you can do OSA scripting in JavaScript. I have a few OSA-JS scripts that are in my crontab. If I had to write AS rather than JS, I wouldn't bother.
Simple stuff in Workflow basically looks like Automator, and advanced stuff in Workflow becomes inscrutable wizardy[1] even faster than AppleScript.
Workflow's advantages are that it A) covers the whole gamut of simple to insane with a single product, and B) has a sharing gallery so you can easily use someone else's magic without understanding it.
Maybe Apple killed the automation position because they were planning to replace it with Workflow, but do you expect them to backport the whole thing onto the Mac? In which case, are they going to actually maintain the user-submitted workflow gallery that makes it useful? Will it be a generally available API, or do they lock it down and use this as a stick to make people stop leaving the Mac App Store? I just don't have an easy time being optimistic about this, given where their priorities lie lately.
Wikipedia is simply missing information for the other 50%. Many are already in existing products or services e.g. Broadmap, Cue, Acunu.
The others are scheduled for upcoming products. LuxVue for example is widely predicted to be the basis for the new iPhone 8 screen. PrimeSense, Perceptio, Faceshift are all part of Apple's upcoming AR push.
In Apple's case, no, because their growth over the last decade is insane. But to be clear that table you posted shows that $1 in 1977 is worth $4.40 today. That is a big difference, mostly cause by the insane inflation of 1979-1981 (around 12% a year for three years). That makes the average inflation rate over the last 40 years 4%, which is double our target of 2%.
Apple's acquisition of Fingerworks provided them with the multi-touch technology (and just as important, IP) that they used to build the iPhone. Definitely one of the most important acquisitions in their history.
I have an old Fingerworks trackpad and it works amazingly well even today.
I think there's some survivor bias in your list. I remember Apple acquired a public transit app (Embark, I think) and I'm not sure the fruits of that were ever known. Apple Maps has public transit features but it's a shadow of what that app provided.
I am one of the Embark founders. I'm no longer at Apple.
Our acquisition actually ended up with the spirit of our product living on in a way that I am proud of.
At Embark, we innovated by taking a regionally nuanced and tailored transit App and giving it scale. When Embark operated from 2008 to 2013, there were small bespoke apps and there were larger more generic experiences (like Google) and we filled a void in between.
Apple's approach was quite similar. Like Embark, Apple Maps Transit has a more regionally tailored experience than many bespoke transit Apps out there, but they're also able to bring it to scale. It's now at a scale we never got close to reaching at Embark.
I seem to remember an incredibly nerdy and thoroughly enjoyable article about the arcane details of drawing subway lines. It was written, IIRC, by a startup that had automated the process of optimising these maps, only to have Apple launch their maps – which had brute-forced the problem by doing it manually.
You're cherry-picking the biggest, most visible of Apple's acquisitions. You should be comparing Workflow to the dozens upon dozens of Workflow-sized acquisitions they have made.
I can't recall many of them off the top of my head anymore, because they're gone and lost to memory.
While I think you are being pessimistic, I don't think the pessimism is unwarranted. That being said, I have a feeling that this is, more than likely, an acquisition to bridge the gap between Siri, HomeKit, HealthKit, and all the other built-in iOS functions that are just ripe for "friendly" automation.
This was my first thought as well. Most home automation solutions are very limited and require integration with something like IFTTT to perform anything but the simplest on/off scheduling.
Workflow can only be better if its integrated in the OS. They've also made the app free and said it's staying in the store. I don't think it will be pulled (if ever) until it's integrated.
I share the pessimism, but there are a couple of glimmers of hope. Firstly, Acompli merely got rebranded, and Sunrise is (purportedly) getting merged into it. More cynically, though, if Apple wanted to shut it down, there are easier ways they aren't above.
Edit: Furthermore, they gave it an Apple Design Award in 2015, so they're clearly alright with its existence.
Ah, sorry, the comments weren't contiguous and so I missed that. I'm not sure what Microsoft acquisitions prove, but I'll continue thinking on it until enlightenment happens.
> More cynically, though, if Apple wanted to shut it down, there are easier ways they aren't above.
It sort of depends on the definition of "shut it down"(see Beats, Swell, FoundationDB, etc.). At minimum, Workflow goes away and rises from the ashes (with a new name) as an automation feature for a future version of iOS. Or, maybe it's just an acqui-hire[1] and they'll start over with a new codebase on something more integral to the OS.
You're pessimistic. But perhaps you are not wrong to be so here.
On the other hand, this functionality would be a brilliant match for Siri, and they've been looking to move in exactly that sort of direction for some while now, as Alexa and Google Assistant have increasingly made Siri's rough edges and limitations show by comparison.
Also worthy of note is that Apple doesn't have a first-party tool in the iOS automation space. So I'm not sure what purpose you suggest it would serve to buy Workflow for the purpose of killing it.
Well it's most a matter of Apple sherlocking features vs. shutting down because they want complete control over the user interaction model. As with most acquisitions, unless the team will continue to work on that app (and based on Sunrise, etc. even if they did), the existing version of Workflow will probably languish and the team's output probably won't surface to actual users for another couple of years.
It's mainly the idea that Apple doesn't want anybody else dictating how iOS should be used.
> Call me pessimistic, but how likely is this acquisition by Apple going to result in them integrating it directly into iOS versus completely shutting down in the future?
I think it's pretty likely. I also think it's likely to return as something more integrated, but of course, the gap between when they shut this down and when the new thing arrives will suck for current users.
If they just wanted to shut it down, they would've done that. There's nothing in the history of the App Store that indicates Apple needs to buy an app to shut it down – all they have to do is change a rule or interpret it a little bit differently and boom - no more workflow. Apple didn't do this, they bought it instead which makes me very optimistic.
A couple of years ago there was a patent by Apple that made the rounds on tech blogs detailing automation in iOS. It actually patented things that, say, Tasker on Android was already able to do. So, I think integrating it into iOS is a sure bet.
> Workflow won an Apple Design Award in 2015, with Apple Accessibility engineer Dean Hudson waxed ecstatic about the way that it handled accessibility. “When I first saw the app, I was just like man, this accessibility is cray-cray. This is off the charts!”
I think that's he strongest indication is that they made the app free. Why would they want to massively boost the user base of a product they intend to axe?
Apparently over the last few weeks they've been sending out contracts to the developers of apps they're integrating with to obtain written permission to use their URL schemes. Google and Uber could've just said no or not responded.
This is according to Marco Arment a few minutes ago on the live recording of the Accidental Tech Podcast, who said he was asked to sign this document for his podcast app.
Google offers contracts and standard business relationships for some commercial uses of their API. Workflow could have reached out to the appropriate Google product person/lawyer, instead of sending a generic email as Marco speculates, if they were keen to use Google Translate instead of switching to Microsoft Translate.
Switching to Apple Maps exclusively over Google Maps is almost certainly a condition of the acquisition as I can't imagine Workflow would willing cripple their app that way, lack of Google email reply be dammed, without getting paid handsomely to do so
Just to be charitable - I can imagine Workflow enter discussions to be acquired and get told that they need to make sure that they have permission to link to those other apps by Apple Legal before anything can happen.
Being a small dev team, and not a corporation with a giant legal department, they panic at the amount of permissions they need to acquire (Workflow deals with a lot of apps) and rush through it as quickly as possible. Coupled with the fact that they, unlike Apple, are a tiny company that doesn't actually make much money off the app, so don't want to end up entering long-standing commercial agreements when the deal may still fall through.
Apple then insists that anything that they haven't had a formal agreement to on a particular date is removed from the app if they want the acquisition to proceed.
Whether that means that they now contact Google again is another question entirely (personally, given that the point of Workflow is to link as many apps together as possible I would hope so)
>Google offers contracts and standard business relationships for some commercial uses of their API. Workflow could have reached out to the appropriate Google product person/lawyer, instead of sending a generic email as Marco speculates, if they were keen to use Google Translate instead of switching to Microsoft Translate.
Sounds like you're speculating about details based on a second hand retelling of a podcast host's individual experience. A host who presumably doesn't have an "appropriate person/lawyer."
I was actually listening to the ATP podcast live as they were recording and I heard Marco's words in real time.
My remarks of the possibility of standardizing their Translation API use aren't speculation; Google's terms/offers are pretty public: https://cloud.google.com/translate/faq
While Marco tried to put a nice spin on it, it should be obvious why Workflow dropped Google services
Oddly enough, book Uber and open clipboard link in Chrome are two of the featured workflows in the tutorial. I added both and neither of them actually work :)
I think open sourcing a language you built is rather different from acquiring companies and locking out competition.
Apple has more incentive to open source Swift than they do to let anyone use content from a company they acquire (remember FoundationDB's fate?). I think Authentec went the same route but I am uncertain.
Apple's attitude to third party devs aside the system is broken for professional work from the get go. It's entirely locked down and siloed.
I recently had to fix some issues with my parents iPad and it was a hell of a eye opener into why it's a terrible idea to rely on these devices at all.
So my mother does painting as a hobby and recently has got into using Brushes (2010 Apple Award Winner), at some point her iPad decided to update itself to iOS 10 (note you cannot disable updates anymore, only postpone them) and was a bit surprised she could no longer open any of the paintings she'd made in the app over the last 3 years. I tried to update the app, it's the latest version
I tried to open them > crash, tried to email them to myself > crash, set up a dropbox account to sync them off the device > crash.
I look online and find this app despite having won awards and being used by famous artist David Hockney is no longer updated, but good news it's open source so someone else has taken up the job of updating it to run on """modern""" iOS, so I install Brushes Redux, it works but none of her paintings are in it. So 3 years of work is on the device, we have a version of the app that should be able to open that work, but we can't because all files are siloed within app.
Long story short after 5 hours of trying to sort this out I finally used a very dodgy looking app to pull the hidden files out of the old brushes app and into the new one. Let's just hope Hockney manages to figure this out too or millions of pounds worth of artwork might be lost for good.
I use software in my day to day work way older than this brushes app but this scenario will never be a problem in my work because my OS isn't locked down, siloed and can disable updates.
There's a shoe here that hasn't dropped. Clearly the ability to automate into features of third party apps is a key benefit of Workflow, which does still hook into plenty of non-Apple services. The question is, what is the blocking issue in these cases?
Is Apple asking for concessions from companies in order to put hooks into them in Workflow, or are these companies asking for concessions from Apple in return for hooking into their services?
In Workflow you can create custom actions that hook into third party apps using the iOS URI scheme. Can that be used to re-create actions on third party apps that are no longer provided in the app by default?
They have had similar tools in the past (automator, quartz composer) and they habitually have abandoned them despite of their success. Why are they making this move now?
(For those unaware, Viticci's is the name that sprang into every "power" iOS user's head when they saw this news. He writes often and well about using iOS devices as "computer replacements" and uses Workflow extensively. c.f. https://www.macstories.net/stories/one-year-of-ipad-pro/)
It wouldn't surprise me if he's one of the biggest users in the entire world, or even uses it more than some of the people who made it. Following him is the only reason I knew what this app is.
I've had Workflow on my phone for years but have never found any reason to use it. Things that I wanted to do were just out of reach. Someone else mentioned in here changing wifi networks... but also things like: enabling/disabling bluetooth, scheduling mute and dnd, scheduling a text message, scripting auto replies to texts based on received texts etc...
There were tools like this for android back when the first android phones were being released. I had a few cools things set up, like making the ring volume silent when I got to work, but then when I upgraded phones the app (there were a few) never got reinstalled because they were never really missed.
I rarely use it on my phone but it makes my iPad incredibly useful. I have a whole series of workflows for sending standard emails and automatically scheduling follow-ups, preparing for meetings, publishing blog posts and so on.
I wonder if it'll result in just Workflow alone being able to access all the good stuff, like how Swift Playgrounds today is the only iOS coding app with the ability to fork().
Oddly enough, I had a premonition about this happening a few weeks ago. I was thinking about how powerful iOS can be with simple workflow-like scripting, and how much better it would be for iPad productivity if workflow were a 'native' solution.
Could this be a response to the whole "Apple is ignoring the pros"-narrative (albeit planned, I'm not saying Apple is acquiring companies based on just som bad press)?
And yes, yes, we want a MBP with 32 GB ram and a new Mac Pro and all other kinds of stuff, but I still feel there is somewhat of an overlap between the Workflow crowd and the "ignoring the pros"-crowd.
I find it kind of weird that this is so similar to things like Automator and AppleScript. Which are things that Apple has never really put a lot of work into. They introduce it with sparse features and then it just stays that way forever. Apparently they already released a new version of the software that removes integrations which work with Google etc. So it's typical Apple, they're pulling back any features that aren't helping with Lock in and that's about it.
"Providing hundreds of actions that interact with the apps and content on your device, Workflow opens up infinite possibilities of what you can do with your iPhone, iPad, and Apple Watch."
Interesting that Apple is looking at meta-tools ordinary users can program to organise their workflow over more applications. [0]
Looks like the Workflow Gallery is completely unavailable, not just closed for new submissions. However, existing Workflow URLs appear to be working if you wait long enough for them to load: https://workflow.is/workflows/ca14840de5e542c88ceaad0f733f8d...
> a tool that lets you hook together apps and functions within apps in strings of commands to automate tasks
I'm curious, how do they access functionality in other apps? Does iOS provide an API for that? Do the apps somehow cooperate? Or did they hack into the iOS kernel?
Workflow uses urlschemes mostly, but at least one co-founder (Conrad Kramer) has a background in reverse engineering and jailbreak tools, which probably helped him discover undocumented urlschemes in apps. Here's a great presentation he gave on reverse engineering lyft: https://realm.io/news/conrad-kramer-reverse-engineering-ios-...
Mostly by URL scheme. Each app can provide any number of url schemes like workflow: or app-settings: or etc. Then they can use data from that URL or the clipboard.
Reading about Workflow makes me wish HyperCard were still around. If all of iOS was essentially a HyperCard stack, this the kind of automation provided by Workflow would possibly be more natural and powerful. I hope this is a sign that Apple is moving away from the hard line between users and programmers, but maybe that's too optimistic.
hah I emailed one of the guys on the team awhile back to ask him for help about how to dissect an iOS app after watching his video about reverse engineering apps. He was super helpful! Awesome to hear this!
Cool. Apple does need such a tool in their ecosystem. The problem is that Apple isn't really well regarded in terms of their cloud tools. Will be interesting to see if they nurture it or kill it.
Local automation on a mobile device has a limited set of use cases (and arguably most of them are also included in the use cases of cloud-based automation), but it's not useless or completely subsumes by cloud automation.
Why did this get downvoted? Haha... I'm excited about this acquisition, but Apple has been seemingly against the URL handler usage by third-party apps since having been abused by advertisers.
Also, usage of these has been limited in recent iOS releases.
It got downvoted because not only didn't Apple shut down the app, they chose to massively boost its user base by making it free. Not a policy very consistent with discouraging its approach.
I've got no idea. Conceivably they could come out with a new approach and transition Workflow to it by having it support both mechanisms. That might be easier to do and be less disruptive to users if the app is in-house. But even if they do introduce a new mechanism, existing apps would still have to adopt it which would take years.
So absolutely Apple could introduce a new mechanism, but their buying and encouraging adoption of Workflow is a strong indication to me that they have no intention of deprecating the URL mechanism any time soon.
I agree here. If they do have an idea for improved inter-app automation or "dialogue", this would be a great example case and implementation of it. Plus this team's knowledge in this field could help guide what's needed, and also help out with Siri stuff to do on-the-fly "workflows".
Luck plays a huge role in everything, but by all accounts this team created one of the best apps to grace the App Store. This wasn't a generic automation app.
Agreed. It may have been luck that they were successful at first (I honestly have no idea), but it seems very clear that they took whatever advantages they had and how to execute on them to continue to make a great product and keep improving it. By the time you get to today I think it's fair to say they earned their place. They're pretty far past where luck alone would take you.
I was at the same MHacks I think! My team won a smaller prize, but we didn't really do anything with the app afterward. It's pretty cool to see a hackathon project with a successful exit. This is very good marketing for hackathons in general.
Perhaps Apple is using the acquisition strategy as a method to shut down an app getting around iOS's "preferred workflow" and UX model. Sort of a strategy taken by Microsoft, Google, and Apple whereby they buy competitors, claim to leave the service running, but ultimately shut them down so it doesn't compete with their main product (see: Sunrise, Acompli, Sparrow, Timely, etc.)