When Apple acquired DarkSky, they absolutely destroyed a service that I loved and relied on. Four years on, Apple Weather is less reliable than DarkSky, and not even close to feature complete.
But DarkSky was a cross-platform service, whereas Pixelmator is software that's already Apple-only. I'm wondering how much I should be worried, and if I should already be abandoning ship.
On the other hand, Testflight has a pretty good acquisition story. It got "merged into main" and is now a first class citizen of the iOS development ecosystem. Workflow being acquired and turning into Shortcuts is a pretty successful outcome IMHO. Beats still continues to make slightly cheaper headphones. FoundationDB is still there.
Apart from Dark Sky, what other products with users has Apple acquired and shut down? Being acquired by Apple doesn't seem to be the obvious death knell that it is for other companies.
BuddyBuild was acquired by Apple and used for their server CI platform, almost immediately cutting off Android support. That was a shame because it was a great platform for Android builds.
I wish every acquisition improved the end product as much as Workflow's did. All Apple's OSes got better for embracing it. I can write and have written AppleScript things, but Shortcuts is a vastly more convenient UX for the things it's good at.
I think the crucial different is Workflow was targeted at what Apple should have been doing from the start. I see it the same way Karabiner, BarTender or QuickSilver (DropBox?) improve the OS experience in ways the platform owner would ideally could have figured out on their own.
Those fundamentally tend to butt against the OS limitations and benefit from becoming a blessed first party utility or feature.
Let's see, I have 157 shortcuts defined at the moment. That wasn't due to some mass effort, just a bunch of little things that accumulated over time.
I have one shortcut that shares the song I'm currently listening to in Apple Music to Mastodon. I use iA Writer for my work notes, and another shortcut creates a new note with today's date with wiki links to yesterday's and tomorrow's notes. (I use that one with Keyboard Maestro: if I'm in iA Writer and press F2, it opens that note (or creates it if it didn't already exist)). One runs on a cron job and copies any new links I've added to GoodLinks to my Pocket account so that it'll sync to my Kobo. Here's one that runs a custom sorting script on my OmniFocus projects. This one dims my office lights; I use Keyboard Maestro (again) to link it to one of the buttons on my Stream Deck.
Basically, for me it's the equivalent of shell scripting for GUI apps. I wouldn't want to write a whole app with it, but for quick and dirty automation jobs it's terrific.
I have tons that solve small annoyances or paper over things I forget. As an example, I listen to an audiobook in audible many nights to sleep, but I often forget to set the sleep timer. Very annoying to have to scrub back hours to find the last thing you remember. I have a shortcut that activates when my iphone is in sleep focus that automatically sets the audible sleep timer for me. It's a little thing, but it's a great quality of life improvement and eliminates my need to think about sleep timers.
- toggle the white point setting on or off to warm and dim the display for nighttime,
- present a menu that makes and displays QR codes for my contact (from vCard text), Wifi info, and more,
- turn off Wifi and cellular at the same time (this one's on my homescreen),
- upload a .torrent file to qBittorrent's watch folder via SSH.
I use Shortcuts at work, too, like sharing a Wifi network with visitors - easier than fiddling with settings and they can take a picture of the QR code to share with others in their party.
My favorite and most handy Shortcut took a picture of an order form, OCRed it, applied a regex to find the order #, and finally showed a QR code I could scan with my scanner; This was at a job where customers would come to pick up, and would often have their order email on their phone or as a printout. The Shortcut meant I could snap the photo first thing and then chit-chat in the time it took the Shortcut to run, instead of them passing their phone to me or reading out the number.
Shortcuts is one of the things that keeps me on iOS.
According to an help article by Apple [1], if you activate wifi or bluetooth during airplane mode, it will remember that and not disable those when you use airplane mode again.
I was also surprised reading your comment as for my phone wifi and bluetooth stay on during airplane mode. I must've activated them once and then it just stayed like that ever since.
I use this in an automation that runs on my iPad. That automation is set to run when the battery charge goes above 80%.
My iPad charger is plugged into that smart plug, so this effectively gives me the 80% charge limit option that so far Apple has only added to the settings for iPhone 15 and 16, iPad Pro (M4), and iPad Air (M2). My iPad is a plain iPad.
• A series of shortcuts to control my Denon A/V receiver.
My receiver is from 2012, long before voice control was available. It does have network control (and serial port control). I've got a script on an RPi that can use the network control to change inputs, mute/unmute, and change the volume.
I've got shortcuts for mute, unmute, various volume levels, and getting the current settings for input, mute, and volume. They work by using the Shortcut action "Run script over SSH" to invoke the script on the RPi.
I've got an Amazon Echo Dot next to the Denon, with the line out of the Dot connection to one of the Denon analog inputs. When line out is connected to a Dot the internal speaker is disabled, which makes that Dot useless for things like checking the weather or timers unless the corresponding Denon input is selected, so I've got the mic on the Dot disabled and use an Echo Show in the same room for all those things. The Dot's job is strictly to supply input to the Denon.
So let's say I want to listen to some music. I can say "<Apple>, Denon Dot" to switch the input to the Dot. (I'm writing "<Apple>" instead of Apple's phrase to trigger their voice assistant so if anyone ever happens to run this through text to speech it will be less annoying. Similar for Amazon's voice assistant).
Then I can say "<Amazon>, on Dot listen to Classical KING FM on Dot" and the echo Show will start the Dot streaming KING FM to the Denon. (It is only actually necessary to say "on Dot" once, but which of the two is necessary occasionally changes, and it is easier to just say it twice).
• Open the "OTP Auth" app.
That's the app I use for TOTP codes on my iPhone. It supports Apple Watch, but asking Apple's voice assistant on the watch to "open OTP Auth" isn't reliable (maybe...[1]). My first choice alternative to voice would be a widget in the Smart Stack but OTP Auth does not have one.
Shortcuts can be put in the Smart Stack. Hence, a Shortcut on the watch that simply launches OTP Auth.
[1] I say maybe because that is based on my old Series 4 watch. I recently upgraded to a Series 10. That has on-device Siri, and I've noticed that is way better. It did not occur to me to check of "open OTP Auth" is reliable.
As a fellow 2012 Denon owner (AVR 3313) you can probably remove the RPi and just use HTTP. Just use devtools on your browser to do an action in the web UI and see what request it sends, then replicate it in shortcuts.
And there's a internet radio capability built in - you have to pay $20 or so a year for it but it should be easy enough to set up a competing service since it uses unencrypted HTTP - probably just a matter of reimplementing the protocol and then DNS redirect.
What I originally wanted was just a better web interface. Mine is an AVR 1913, and the web page served by the receiver is kind of slow. The shortcuts came later.
I did indeed look at the requests the browser made when on the receiver's web pages. I also used packet sniffing to see what the Denon app on my phone did.
I then wrote a simple web page that just showed current status, and had big buttons for the three sources I use, and for mute/unmute, and for several volume levels, and used JavaScript on that page to send requests to the receiver.
That's when I learned about CORS. The receiver does not send CORS headers, and browsers take that to mean that scripts running on pages that do not come from the receiver should be blocked from receiving any data back.
For the commands to change source, mute, and volume that was OK. They are simple GET requests with the change as query parameters. They could be done without triggering a CORS preflight check. Whether or not the browser blocked the response didn't matter.
Not so for getting status.
That's when I switched the approach from the web page using JavaScript to talk to the receiver to having it be a form and having the web server talk to the receiver. The web page source looked like this:
<?php
...a bunch of functions to control the receiver
...code to process the form and invoke those functions
?>
...the HTML for the web page
Later when I realized being able to control the receiver from shortcuts would be nice, it was a simple matter to copy the web page source, delete the HTML, replace the form processing with command line processing, and have a command line script for controlling the receiver, and then use the run script via ssh shortcut action to invoke it. It never even occurred to me to consider issuing the commands directly to the receiver from the shortcut.
I just gave it a try, using the "Get Contents of URL" action to GET the URL http://ip_of_receiver/goform/formMainZone_MainZoneXml.xml and it worked. It gives back a blob of XML that includes the data I care about (source, mute, volume).
On my Mac I'd probably handle that by passing the XML off to a script to extract that data. I'm not sure how I'd do it on an iPhone or iPad.
There are commands to get individual data items such as the volume whose results might be easier to deal with, but I'm not sure they can be used from shortcuts. They are POST requests to /appCommand.xml, with the command (or commands if you want to batch commands) in XML in the body. For example to get the volume you post
The "Get Contents of URL" action does support POST, but the only options it gives for the post data are JSON, Form, and File. For JSON and Form you give it name/type/value triplets and have no direct control of how the post data is formatted. Maybe file would work to get the XML the receiver wants sent but it probably won't have the right content type. I have no idea if the receiver would be OK with that.
> Beats still continues to make slightly cheaper headphones. FoundationDB is still there.
Final Cut Pro was bought from Macromedia. And Logic from Emagic. And off the top of my head Astarte (iDVD), FileMaker (FileMaker Pro and Bento, though that was originally spun out or Apple in the first place), SoundJam (iTunes), Siri (Siri).
All of these were mildly- to hugely-successful products.
Note how most of them are from the era when Apple cared about software, especially pro software. And how Siri is a barely functioning stagnant service.
You'd be hard pressed to find any more recent success stories.
Shake was a big one. Apple loved to put Lord of the Rings and King Kong on its homepage, but Jobs always seemed pissed that they couldn't dumb Shake down. Artists at Weta, ILM, Etc and others were not about to tolerate a gimped product.
Unfortunate side note: Apple was going to open-source Shake, but abandoned the idea after realizing it would face an endless parade of patent trolls if people were able to scour the entire codebase line by line.
Another controversial one was Lala.com, if I recall right — they shut it down right away, but it had an avid user base.
I get why Apple wouldn't want to maintain two music services, so that engineering talent likely got absorbed into iTunes. It's yet another story where the competition was offering something really good / unique, drawing in customers interested in those differentiators, and it ended up disappointing a lot of people getting bought out.
It's built in to Siri though. So Apple benefitted from Shazam. I'd say Apple's Weather app has also significantly benefitted (albeit a bit late) from Dark Sky's acquisition
Filemaker might as well have been bought when dinosaurs roamed the Earth. It's such an acquisition from another era that it's a fully owned-subsidiary of Apple inc. that is completely separate from the rest of the company. Do you know how many other subsidiaries Apple has?
It (optionally) integrates with the Apple Music Android app now, and offers to add to your library there whenever you scan a song, so I assume it's a good funnel for them to get people into their service ecosystem.
I would bet a million bucks that Jobs put that price in because he basically said well if they buy the Linux version we're down one Mac sale from them so charge them our profit margin on a Mac Pro.
That iGesture pad was a life saver when I was using a PC at work. I eventually switched to Mac and got used to their trackpad gestures, although I still change the settings to enable 3-finger click-and-drag like the iGesture.
The talent and ideas that were Pixelmator will be substantially diffused as it's absorbed by Apple... most of what you liked about Pixelmator is likely no more over the next year or two. Depending on Apple's reasoning for the acquisition (i.e. how much of it was just for the talent vs the product) you'll may see some small glimpses of Pixelmator's influence a couple years from now in Apple's stuff. Most of the time Apple doesn't keep the acquired product around.
On acquisitions like DarkSky (RIP), sure. This looks a lot more like a Logic-style acquisition.
Pixelmator would slot nicely into the same consumer set of productivity apps that ship with all Macs (Pages, Numbers, Keynote). Photomator will get them back into the market they abandoned when Aperture was shuttered.
Speaking of Aperture… am I the only person who remembers that Apple owns Claris? Why didn't Apple just hand off Aperture to Claris and say "just keep this thing working on new MacOS releases"?
And the management team that brought us ClarisWorks is still leading Apps to this day. Apple doesn't wipe out management during acquisitions, it permanently entrenches them in their structure.
> I still haven't found an app that's as good as Aperture used to be with my workflow in term of UX/convenience etc...
Me neither… I wanted to like Lightroom, which was the solution most of the community seemed to migrate to, but between the infuriating inconsistent UI and the predatory subscription model I did not use it for long. And now I have a Rube Goldberg thing that is janky and feels brittle.
Yes! And plugins are great, but the experience is not smooth, and quickly annoying when working with many photos. Also, switching libraries is not good. I wish it were more integrated because on paper, a photo management app combining the features of Affinity, DxO, and others sounds fantastic.
Logic and and Final Cut were bought and developed since. Pixelmator fills the open Photoshop space in an Apple way, and will plausibly go the same way — no vague pessimism required.
The (almost) direct counter example is Aperture. That was the “Pro” photo application and it was killed for seemingly no reason with no notice. It’s fairly reasonable to be pessimistic about this acquisition given that history.
The main worry is that it will be an acquihire into the Photos app and Apple doesn’t actually want to have a separate image editor (let alone two).
They used to have Aperture competing with Lightroom and then decided pro photography wasn’t a space they needed to be in, has something changed where now they want their own Photoshop competitor?
Dark Sky would've added more value if they'd just renamed it Weather and made it the built-in app, and yet...
I do hope they'll offer Pixelmator as an included app on Macs and Pixelmator Pro alongside Logic, Finalcut, and other "Pro" software. The lack of a built-in image editor can be annoying.
Photos works for some stuff, Preview includes basic adjustments too, but sometimes you just want something like a hue/saturation adjustment instead of color temperature and pink/green tint, or multiple layers so you can experiment with different edits non-destructively.
Eh, I don’t think it’s the same thing. The gulf between “photos user” and “pixelmator” user is quite high, much more so than “weather app” and “weather app but better”.
In particular, if you have the average user Pixelmator, they’d be worse off. The same isn’t really true with weather or darksky - they really just do the same thing.
We still have iMovie and FinalCut, GarageBand and Logic. Apple has kept two different product lines before.
Also remember that some of those have been crippled in the past. iMovie used to be way more advanced. Older versions of Pages had (pretty basic but still) layout options that were completely removed.
It's also not impossible that Apple moves a few of Pixelmator's tools into Photos but kills the rest of it, either actively or just by stagnant development.
> This would be very short sighted as Pixelmator adds way more value to the Mac platform than a better Photos app.
This is comparing apples to oranges. A better photos app isn't even remotely comparable to shipping a raster image editor. One is concerned with overall rendering of the products of a camera, the other is concerned with precise editing of a raster image.
Apple sells Macs. The Mac platform is enhanced by the existence of Pixelmator as an exclusive app.
If Pixelmator were to disappear then the value of the Mac platform would decrease. There is nothing that the Pixelmator team could do to the Photos app to make up for that.
> Pixelmator as a product is literally what Apple would have built anyway if they made an attempt.
Even accepting this premise there's little reason to think Apple would have cared about this particular market before they bought Pixelmator. Why would you think Apple would target a given market segment?
There’s no way Apple can build this. Their human interface people all seem to be gone on the desktop. So many things work so bad these days when they migrated to the new ui framework.
At the time Apple bought eMagic, Logic's UI sucked. It actually had dialogs that told you to "reboot the dialog for changes to take effect."
Given how well-regarded Logic is today, it must be drastically improved. I haven't looked at it lately, but am considering the bundle with Motion and FCP.
One piece of software Apple built in-house is Motion. While it suffers from a few UI gaffes, it was an innovative product that still has no competitor in the motion-graphics space.
I want to point out that the same management team that brought us ClarisWorks is still leading Apps. Apple drag and drops teams into their org chart and gives them tons of autonomy.
> The talent and ideas that were Pixelmator will be substantially diffused as it's absorbed by Apple...
The "ideas" in pixelmator are mostly updating traditional image-mutation patterns to match the native environment language. Let's not pretend that this was some kind of revolutionary application for image development.
Is it implemented well? Absolutely. But this is hardly an example of developing new software practices or processes.
I don’t understand why companies buy other companies for the talent and not product. Why not just make everyone working there an incredible offer at the same time? It would cost so so so much less than these massive buy outs. Maybe not all of them would take you up on it, but if you buy the company a lot of them may not stick around post-buyout anyway. I feel like this would be a lot more effective also because in a buyout, employees just make the same old salary at the new company. In my method, they make a ton more and are more likely to stay
> Why not just make everyone working there an incredible offer at the same time?
Under civil law this is regarded as tortious interference. Businesses have a contract with their employees and if you interfere with it to harm the employer then you are liable for damages.
If you tried to make a mass offer like this, the employer could likely get a judge to place an injunction against it immediately.
If they don’t notice until further down the line, watch out: damages are unlimited. They can extend to a judge breaking up your new business unit and handing it back to the original employer or rewarding damages of the entire lifetime value of the business unit.
it's ridiculous how america is all about free markets except for the instances where rich people could lose money, then suddenly free markets are bad and evil
Which rich people are you talking about, the buyer or the seller? Presumably the buyer of a startup is richer than the startup founders. If poaching all the employees of a company was legal, then we’d end up with only monopolies by the largest and richest, and it would be legal for big companies to crush smaller competition. The playing field in the U.S. and everywhere globally is definitely biased toward the rich, but you’re inadvertently arguing for even greater concentration of wealth, it doesn’t seem like this argument is well thought out.
Step 1: make everyone an incredible offer
Step 2: get them all hired away from your competitor who is now out of business
Step 3: in a year or two, restructure all these people out (or just fire them if your jurisdiction allows)
Step 4: your competitor is gone, and all it cost was a year or two of salaries.
Seems like a great way to help out budding monopolies.
it seems like you can just prevent this by providing incentives for your employees to not get poached, and also companies that mass-hire-mass-fire would get reputations for doing so, and people wouldn't fall for it. making it illegal instead of requiring businesses to actually pay for retention and loyalty in a free market way is so silly
When a mass employment offer is made to steal or destroy another business, it's usually something ridiculous. For developers it might be a million a year each, for example. It's not an amount intended to be paid perpetually so it can be larger than the defending business can be paying to retain.
It is not illegal to do general hiring at good rates and shop for employees at a particular company. That wouldn't have the same results as buying a company. Plus, you wouldn't own their creations; you'd have to rebuild or clean room steal it.
The 'people wouldn't fall for it' is in error.
People aren't rational actors and don't have complete information.
That's a bold statement, I know, but it's at least as correct as 'people wouldn't fall for it'. I'm pretty sure it's easy to make a case for 'too many people will fall for it'.
Multiple reasons. That it does happen should be reason to question your assumptions, rather than assume some obvious imagined alternative has been overlooked by everyone, right?
While poaching one employee at a time might be usually legal, attempting to poach all employees of a company might not be legal, and either way is considered unethical.
Paying off the investors may be the goal.
Eliminating the product or competition ethically may be the goal.
Buying the competition’s customers, and/or distribution channels may be the goal.
Acquiring the top talent, while giving them the expected reward for having bootstrapped a company, might be the goal. Founders are often uninterested in a salaried position for themselves, but may be interested in a return for the company and payoff for everyone in it - as backpay for their investment, completely separate from their salary going forward.
Also, your hypothesis is not accurate. Buyouts are not always, or even usually, massive. It’s common for them to be small and medium sized. It is definitely not a given that making persuasive individual offers would be any cheaper than an acquisition, let alone “so much” cheaper. Depends entirely on the situation.
I'll admit, as an attorney, this isn't my specialty, and every jurisdiction varies, but the ye olde common law of tortuous interference requires something more than mere competition, this is America, not the EU.
2 DAN B. DOBBS, THE LAW OF TORTS §§ 448-52 (2001)("you are thus free to induce my customers, employees, or suppliers to deal with you instead of me, as long [as] they are not bound to me by contract").
Restatement (Second) of Torts § 768 (1979) (stating that interference with a competitor’s contractual relations is permissible if it does not employ wrongful means and is intended to advance the competing interest).
Wal-Mart Stores, Inc. v. Sturges, 52 S.W.3d 711, 726 (Tex. 2001) (" we conclude that to establish liability for interference with a prospective contractual or business relation the plaintiff must prove that it was harmed by the defendant's conduct that was either independently tortious or unlawful. By "independently tortious" we mean conduct that would violate some other recognized tort duty.").
Correct, tortious interference has criteria, and making competing employment offers doesn’t necessarily or automatically meet those criteria, but it could if there are other factors involved. Again, it depends on the situation.
You’ve asked two different questions. One about legality and the other about public perception. There are lots of things that are legal and still considered unethical. And there are lots of things that might or might not be legal, that businesses avoid simply because there’s legal risk, and/or avoid because there’s risk of negative perception.
If everyone involved in a startup agreed to be individually hired, and divest interest in the startup, and there was mutual understanding on all sides, then there may be reasonable chances of success and no lawsuits. I think that probably has happened before. If not everyone agreed to it, and a company tried to acquihire all the individuals of a company forcefully without agreement by the investors and founders, there’s a high likelihood (risk) of legal conflict, and the likelihood will increase under US law if the acquiring company would start to look anything like a monopoly on the market in question after the unofficial “merger”, right?
Agree with the other person - there's nothing unethical about hiring people in right-to-work laws and systems however you like. employers can fire at any time with no reason, the reverse also has to be true that they can hire at any time with no reason
buyouts are often massive considering the alternative, which is the cost of recruiting and possibly inflated salaries for the people you recruit, which frankly happens often in buyouts anyway
Like the other person, you’re arguing about individual hires, and not considering the implications of whole-company mass poaching.
Sure some buyouts are big. But plenty are small. Most aren’t “massive”. The histogram, I speculate, is probably something like the Zipf distribution: the frequency of buyouts of a given size is probably inversely proportional to the size, to a first approximation. https://en.m.wikipedia.org/wiki/Zipf%27s_law
Ya that’s a solid point. Though many startups give their employees equity options… so you have to factor that in too. Also buying a start up for talent seems risky since many people that join startups are looking for a totally different energy than a large corporation, so it seems reasonable that there’d be a big drop off of that talent as soon as it gets acquired… especially if the vision is not aligned
Apple historically tends to look for shipping results, and the underlying software and services (such as using DarkSky's algorithms and server code as starting points) are often worth it over just putting offers out to key people.
This obviously isn't always true; they do have some longer-term research projects and strategic initiatives we've seen leak out (cars and non-invasive blood glucose monitoring are common mentioned ones), but I think Apple generally would prefer to let others succeed or fail in the research.
There's nothing _to_ Pixelmator IMHO other than the product. Apple knows how to do sepia tone filters already.
Ok so to be fair.. I own an iPhone for about 3 years now and only discovered it comes shipped with Shazam about 6 months ago and only used it twice since. When I told my wife (also a somewhat long-time iPhone user), she didn’t know it came build-in either.
I’m not a power user, neither is my wife.. I don’t think it is all that well advertised.
Shazam was bought to boost Siri’s ability to recognize music but Siri isn’t really good at much, so it hasn’t been fully absorbed. Now with AI eating the world I assume that functionality will get reproduced by a foundation model and actually integrated into the OS
That's interesting. Is Shazam a default control center button for new phones? I don't remember how mine got there. (There's still probably a discovery issue with those buttons as they're just icons.)
Really? Even though the company is in Lithuania? It seems like they’d probably keep on working on Pixelmator or something closely related since any other teams would be a long way away.
Doesn't seem that way to me. The predicted rain over the next hour looks the same as it did in DarkSky, and you can view the scrub the predicted clouds map timeline and see that it's predicting the same stuff. And the real-life quality where I live has shown no change, nor is there any obvious reason why there would have been. I presume Apple bought DarkSky for their tech rather than their userbase, so it wouldn't make any sense for them to reduce its computational quality.
> and not even close to feature complete.
To be honest, I don't really remember what else was in DarkSky, I just used it for its main feature -- rain over the next hour. But the Apple Weather app has a ton of features. Is there one or more specific features you're missing?
I think it sucks for Android users that Apple bought it. But for iOS I've been totally happy to have it integrated, rather than dealing with 2 separate apps.
I am a regular runner. The accurate micro-forecasts on Dark Sky were a huge help for me to plan ahead so I wouldn't get caught in the rain. Apple Weather mostly fails at this.
Additionally, I really dislike the Apple Weather dataviz for the day's trends. This time of year, the my local weather can wildly change from early morning to late afternoon, and I want to plan what to wear. I could glance quickly at Dark Sky and see the trend almost instantly. Apple Weather requires this awkward tap and drag gesture to see actual temperature values through the day.
Apple weather puts all sorts of weather data at the same level, despite the utility being wildly different. I need to know the temperature trend for the day, or rain chance. Wind speed isn't very useful to me day to day, yet they are at the same "level" of UI access. It doesn't feel very driven by user needs, but perhaps there are a lot more sailors using the app than I realize.
> Apple Weather requires this awkward tap and drag gesture to see actual temperature values through the day.
You mean scrolling horizontally to see the values?
It's not an awkward tap and drag, it's just scrolling.
But if you don't like scrolling (which I understand), then just tap without dragging, and it'll show you a full-screen graph with a curve representing the temperature throughout the whole day. It's fantastic.
The interface doesn't make it clear that it's tappable, I'll certainly admit. But I hope that helps you. The graph view only got added maybe a couple of years ago, and I think a lot of people maybe still don't know about it.
I miss the visualization, but IMO the biggest feature loss is the history feature. You could select any day in the past, even going back decades, and get historical weather information.
The hyper-local rain forecasts were always accurate for me with DarkSky. The "rain starting in 3 minutes, stopping in 10" was accurate. But right now, Apple Weather is telling me it's cloudy and raining where I'm at, and I'm looking outside my window to clear skies and dry ground.
> I miss the visualization, but IMO the biggest feature loss is the history feature.
AFAIK this is still in the API (although it wasn't at launch). Apple is fine with third party weather apps that provide all the information within WeatherKit.
> The hyper-local rain forecasts were always accurate for me with DarkSky.
DarkSky didn't magically rectify the difference between the macro predicted weather and hyperlocal forecasting either. One is a legitimate weather model, one is vectoring based on the last few radar maps.
Apple just still puts the macro predictions up front, and treats hyperlocal as short term badging/alerts.
> But right now, Apple Weather is telling me it's cloudy and raining where I'm at
Does it say "rain will continue for the next hour", e.g. a hyperlocal forecast?
> But right now, Apple Weather is telling me it's cloudy and raining where I'm at, and I'm looking outside my window to clear skies and dry ground.
Huh, I definitely haven't experienced that with the chart that shows rain over the next hour, the part that comes from DarkSky.
What happens when you look at the rainfall map timeline from the past couple hours and the prediction over the couple next?
Are you just on the very edge of rain/sun? Or is it all super spotty? Or is it totally and completely wrong regionwide? And is the historical data from the previous couple hours accurate at least?
Just curious where the problem is coming from. Because it's visually pretty obvious how it works when you look at it.
> And sometimes it’s bizarrely off, like saying the UV index is 1 on a cloudless June afternoon. There’s no sanity checking to speak off.
That sounds like weather data that hasn't updated for hours because you have a bad connection or something.
It does drive me nuts that all weather apps I've ever used always show you the previously loaded data, even if it's 5 days stale. I absolutely despise this "optimistic" UX model where it assumes that the most recent data is "good enough" until new data is fetched. Especially since it never even tells you how stale the data is.
Like, if weather data is more than two hours old, I'd rather you show me nothing, because then at least I know to go outside and check, rather than be deceived by the app lying to me.
It happens after multiple refreshes, and it’s just a specific example I chose out of many cases… though it may be possible that the backend server just ignores all that and sends me old data anyways…
Apple Weather is still missing some of the data that Dark Sky exposed, like cloud coverage percentage and other niche info. I also find the UX a little worse, as I like more data at a glance. But you can tell they're using the Dark Sky backend, as it has the same bugs that Dark Sky had, like slowing loading map tiles which sometimes fail altogether. And there was the time they accidentally reenabled the Dark Sky API after an Apple backend deployment. :D
I was (selfishly) happy with the acquisition because DarkSky didn't support where I live. Now I have hyperlocal rain notifications I didn't get before.
Coming from the perspective of a non-DarkSky user, the DarkSky acquisition added tons of value for me. Apple's iOS Weather app is better than it's ever been in terms of accuracy and feature-breadth.
It seems that Apple made things worse for the (small number of) DarkSky users while improving things for (a huge number of) default-app users.
I've heard that Apple Weather is much less reliable nowadays outside of the U.S., but I agree that it's super accurate for me on the East coast of the states.
It’s worthless in Thailand. I was checking it last week with a Thai friend here in Bangkok. The forecast was clear skies while in fact we had an epic monsoon storm.
A fun fact about ITCZ is that you will simply not find a reliably correct weather forecast. In places like Bangkok or (depending on the season) Hong Kong locals normally know to use the weather radar.
Once going on a hike with a friend we got stuck amid torrential rain which for 40 min pretty much affected a less than 1x1 km area centred on the bench (with a roof) where we sat down. We knew it from the radar, since all apps showed mostly sunny weather. I didn’t bring the umbrella since it was supposed to be sunny and estimated cumulative precipitation was insignificant—who knew it would all fall directly on our heads!
The radar won’t give you a forecast, but (if you are lucky to not get hit by weather developing on top of you) show you an animated map of where in town all hell is breaking loose now vs. where it was 15 min ago and you make your own conclusions. Newer versions of Weather app include a mini map of precipitation in some areas but I assume not all local radars agree to feed it their data, and even if some do the extra moving parts involved in getting and processing the data introduce too much of a lag for real-time weather developments. I doubt optimising that is Apple’s priority.
I enjoy a good poking fun at weather apps (back then Dark Sky, now Weather) as much as the next guy, which is exceedingly easy while you are in ITCZ, but the reality of fluid dynamics on this big rotating ball is such that some places worry about a cold front they can see coming days in advance while others live in weather that may develop within minutes right there and then. Guess in which of the two do most paying customers live!
I learned that there’s no umbrella weather. Either rain’s too light, or the damn thing snaps/flies away but the street’s basically a river anyway.
The galaxy brain is to wear flip-flops and care less (or, if you are a local in Bangkok, move by car/wear one of those thin plastic raincoats, depending on your class).
If you mention sidewalks, remember cars that splash you with dirty water. You’d think umbrellas are of limited use, until one snaps in the wind and you cut yourself trying to fix it—then you’d hate them!
Late this summer, Apple Weather finally lost me (I'm in Indiana).
We had a storm roll through, and the temperature dropped 15º. Guess whose weather app continued to report the higher temperature?
But the real problem: rain forecasts were painfully unreliable. I spend the summer driving topless in my Jeep, and it's helpful to know these things in advance.
Well, that and the new UI was so much more cluttered than Dark Sky's, but I stomached that for years before throwing in the towel.
Why worry about something you have no control over? Keep using it, but be exploring alternatives now in case it does. Don't waste energy fretting over this.
I feel like those ideas are contradictory. Exploring alternatives just in case is wasting energy.
For example, for months I’ve been thinking of trying Inkscape to replace Affinity Designer, yet I keep putting it off because I’m not exactly enthused about the idea of having to learn yet another vector app again and deal with all its bugs and quirks.
It is wasted, because in this situation you’re forced to do it. If you end up not switching, all the time you spent trying something else comes to nothing. If you do switch, you were still forced to spend a bunch of time looking for something.
For me, "nothing" is rarely true. When I've had to learn a tool that operated in a different way, I've often come away realizing that I could think about a common task differently, or that there are capabilities I didn't realize I wanted.
I this case I'm thinking of domain-specific tools meant for creation or curation, like an IDE, image editor, word processor, etc. That wouldn't apply to bureaucratic paperwork-type tools, where learning the site is typically a one-off and is pure waste.
Yes, I agree, “nothing” won’t always be true. But I felt like the idea came across and that having to overly explain and nitpick my own clarification was unnecessary.
I think it's more relevant to ask "Is there any way this can end well?". Unless a company is basically down and out an acquisition, especially by a mega-corp, is basically never going to make things better.
When Microsoft bought GitHub it actually seemed like GitHub started working more on developing their product, but this quickly turned into essentially starting to do the same busywork every other big tech company does with lack of quality control, pointless reshuffling of UI components in places, embarrassing deficiencies in what should be obvious and exposed places.
So, on a long enough time frame even (observed positive) promising acquisitions seem to turn into bad deals.
And now seemingly working only on Copilot. The release notes of Visual Studio Code used to begin with accessibility, an indication of commitment to an important pillar of their software. But the KPIs have moved, and now Copiloy takes the spotlight
While I am also very sad about DarkSky, it doesn't always go that way. Shazam was purchased by Apple many years ago, and many people have no idea. It still a stand alone app, but got control center options and Siri integration (even for those who didn't have the app installed). While the app does push Apple Music a bit, it's largely clean and without other ads, which would probably not be the case if they were still on their own.
How Pixelmator goes will largely depend on their plan. Do they want an app in this space, the spiritual successor to MacPaint, or did they just want the underlying tech (and maybe the team) to add a couple features to Photos? If it's a new value-added app, I think it's great. If they are just going to add some minor tweaks to Photos and throw the rest away, that would be pretty horrible.
I was a Pixelmator user from its launch, but switched to Affinity a few years ago. If Apple does something good, I probably won't be tempted to buy the next version of Affinity whenever it comes out. I'm a very occasional user.
In view of how AI is changing graphics tools, my guess would be that Pixelmator's pay-once business model didn't bode well for long term competitive viability.
We've already seen Affinity switch into releasing a "v2" after years of updates, and on the other hand hobby apps like LightTricks get far more out of their subscription services, despite being a consumer grade template-based app.
As for how it will be run. I'm guessing that Apple will just be their backer like they do for Claris. I don't imagine a MacPaint style app returning to macOS, but sharing code between the two to enhance mark-ups, the photos app and the like seems likely.
For random conjecture: this may also mean Pixelmator grows into the AR space.
I dunno, I thought "sherlocking" wouldn't be a thing if they acquired instead of duplicating their solution in-house, but it's the same effect, just more equitable to the original creators.
I can’t think of any Apple software that doesn’t get the job done, isn’t straightforward about its capabilities, and isn’t easy to use. Their whole shtick seems to be just that.
A minority of people will always prefer to use competing products, if for nothing but sentimental reasons. That’s fine, as such market will always exist.
Pixelmator was a successful team with a polish product and happy customers. What Apple brings to the table is money I guess, but was that a critical issue the company was facing ?
The talent could bring a lot of good to other Apple products, but I guess Pixelmator as a product has reached its peak at this point.
It started off only using the USA's National Weather Service as a source[0] but gradually added international support[1]. But even then, outside of the US/UK, you would have been better off finding an app that you know uses your region's weather stations.
When Apple acquired DarkSky, they absolutely destroyed a service that I loved and relied on. Four years on, Apple Weather is less reliable than DarkSky, and not even close to feature complete.
But DarkSky was a cross-platform service, whereas Pixelmator is software that's already Apple-only. I'm wondering how much I should be worried, and if I should already be abandoning ship.