Hacker News new | past | comments | ask | show | jobs | submit login
Microsoft will shut down the Cortana iOS and Android apps in 2021 (engadget.com)
79 points by behnamoh 49 days ago | hide | past | favorite | 61 comments

I was a big Cortana + Windows Phone fan (big enough even that I had one in use in 2014) and it did start out competitive. Over time though, Siri got better, Alexa got better; Cortana never did. Microsoft shows an inspirational video at Build every year, but what they show never seems to become reality.

In contrast, I started investing in Alexa mid/late-2018 and while it was great to start, it has become increasingly useful. It makes phone calls to businesses, and gets it right very consistently; it knows business hours; the automation has become more powerful over time (a category Cortana never even began to touch); and the integration ecosystem is superb.

Alexa keep improving, too. Yesterday I invoked a routine that included playing music, while the invoked service was busy playing on another device. While in the past the music just wouldn't have played, now she says, 'Amazon Music is already in use on another device. Did you want to play Thunderstorm Sounds on Bedroom Sonos?' That's just one example, but they are numerous and regular enough that I was able to pick one from just yesterday.

Microsoft missed a massive opportunity to implement deep integration between Windows on the desktop and Windows phone. What I mean by that is a smart phone that become such an intimate extension of your computer that it blends into it during use.

I remain flabbergasted that today, in 2020, phones don't blend into desktop and laptop computers in any significant manner. I can't browse files on my iPhone X from my Windows 10 desktop through Bluetooth. There are band-aids and work arounds through apps and services. My point is that this should be a natural integration by now, not a pile of band-aids. To be sure, companies like Apple are at fault as well.

Think about everything you can do when you plug in a second, third or fourth monitor into your computer. You should be able to do that and more with your multi-core, multi-gigabyte, hyper-connected, high display resolution smart device.

I should be able to start a Zoom meeting on my phone, swipe the call onto my computer screen and back. I should be able to drop PDF and other files onto my phone by dragging them into the aforementioned native interface.

I still think MS have an opportunity to come out (again) with a phone that is a such a good extension of Windows that it would feel like you ripped the phone right off the Windows desktop and file system. The problem with this is it requires leadership and commitment they might not have.

> I remain flabbergasted that today, in 2020, phones don't blend into desktop and laptop computers in any significant manner. I can't browse files on my iPhone X from my Windows 10 desktop through Bluetooth. There are band-aids and work arounds through apps and services. My point is that this should be a natural integration by now, not a pile of band-aids.

On Android+Linux, KDE Connect is a single solution to do all of this. Files, clipboards, bidirectional music/volume control, pausing computer music during phone calls, notifications, SMS (desktop app), type from computer onto phone, and more. I've heard there's some work going on to make it work on Windows+Android, and for Linux there's already a KDE Connect applet to make it work on desktop environments other than KDE.

Completely agree. I want a computing model I called PAO, where applications are singular and available across all of your devices. If you're interested in a read: https://tiamat.tsotech.com/pao

I wonder if some sort of method of sharing Redux actions over a websocket could possibly work to make an app work like this - although it would quite obviously add a lot of overhead to development complexity and production resources, so only certain apps would really benefit enough for the complexity. Worth trying, though.

> I remain flabbergasted that today, in 2020, phones don't blend into desktop and laptop computers in any significant manner.

Microsoft did something like that in 2015. The high end Lumia's could plug into a dock with a keyboard, mouse and display and show a windows desktop with a start menu and apps.

It was called Windows 10 Continuum and it never caught on.


Why does it need to be it's own fledgling-sounding thing?

e.g. Win10 has automatic tablet mode. Perfect.

> I should be able to drop PDF and other files onto my phone by dragging them into the aforementioned native interface.

Apple have gotten pretty far along that road with Airdrop, it works great.

As for the rest, MS tried some of that stuff and it landed with a thud. The mass market doesn’t actually care for the stuff (how often are you actually taking a Zoom call on your phone and wanting to transfer it? I like the idea but I can’t think of many times it would actually apply).

I’m sceptical it would ever be a thing now. There’s a reason why so many companies make Electron apps: desktop apps just aren’t all that important, and they’re not likely to dedicate dev resources to one.

the "Your Phone" app on windows 10 does some of this. At least I can look at SMS and my pictures and stuff! Over wifi even!

Windows phone was actually legit, I was one of the few who loved it I guess. Those big tiles etc...

There are dozens of us!

I absolutely loved my Windows phones. I had a couple of different Lumia phones, including the flagship Lumia 920 which had a pretty decent camera for the time, but my favorite was the Lumia 520. The Lumia 520 was only $60, and didn't need a case or screen protector or anything fancy like that. When it was dropped, the back popped off and the battery flew out which reduced the impact on the screen. It had offline maps, a web browser, and the battery easily lasted for 3 days. And if it was lost, it was just $60 to replace! I went through 3 of them before my bank removed their Windows Phone app, which meant that I couldn't do mobile check deposit. I "upgraded" to an android phone, which cost much more and was so much more complicated to use.

Ahhh I do miss my Lumia 920. Solid hardware and a refreshing take on software.

I was a windows phone dev. It was friggin -AWESOME-. Really sad it never materialized. I had worked on android before and M$ really knows how to treat a dev right.

In many ways Windows Phone came out at the wrong time (i.e. way too late) but it felt like a real sweet spot for UI. Android had stalled out on 2.x for a year or more while they focused on tablet UI (woops) and iPhone was starting to feel very samey. When I saw the WP UI I was immediately sold. That plus the keyboard made for a phone that was a genuine joy to use.

But... no apps. And then they pissed everyone off by announcing that Windows Phone 7 devices wouldn’t get an upgrade to 8. That was the point at which I realised MS didn’t really understand what they were doing and switched to iPhone. I still think there was potential there, but alas...

The UI might have been nice for small pockets of technical folk but it was a mess for a lot of others.

For example WP7s emphasis on big long lists made it very hard for me as a dyslexic to quickly scan for the one app I wanted in that list. My eyes would constantly jump about and the lack of other visual clues made it hard to anchor to the item I wanted. I don’t even have this problem menu bars because those have more visual clues that WP7 did.

Another issue was how the flat design made it impossible to distinguish between a button and a highlighted block. Also between a hyperlink and normal emphasised text (frankly, hyperlinks do not belong on UIs outside of document markups. There was one time my mum changed her WP7 language to Polish (she has Polish heritage) to see if she could read it. She couldn’t and that mistake was totally user error on her part. But she could read enough to get us to the language settings and it literally took both her and I more than 10 minutes to realise that there was a button that needed pressing after the language selection because the button blended in with all the text on that dialogue. If your only distinguishing feature of interactive elements is their caption then that’s simply bad UI design.

To give balance, I’ve used nearly every Windows device on phones or PDAs. While Windows Mobile (and CE before it) had its own issues they were still enjoyable devices in their own ways. But I hated WP7 because the UI constantly left me guessing. My mum hated it for the same reason so switched to Android. One of my friends, who is a reasonably technical Windows fanboy but doesn’t work in the IT industry, he hated it too and then switch to Android. It was not a well design UI at all. And while I’m sure there are pockets of engineers who will miss WP7, I’ve never met a single end user who liked it.

I am also one of those people I enthusiastically bought one back in early 2015.

I was one over also by the UI, found it super easy to use.

I still have 3 Windows Phones, 2 Lumia 950s and 1 Wilyfox Pro. Still use my Lumia as my main phone and will keep doing so for as long as I can keep getting new batteries. Windows phone was a superior product. It is too bad Microsoft gave up on it. It saddens me to this day.

Some parts of it were very superior and some parts were abysmal.

The overall flat design was a big pain in the ass when you couldn’t tell what was clickable / editable.

The accessibility of putting the data first was awesome. Apps being tiles or a simple list of names was great. They started that with the Zune and it’s still a winning play on the Xbox. It worked great on the phone.

I want to try light phone and hope it doesn’t suck.

I have to agree with not knowing sometimes what is clickable. I have that problem now in Windows 10 UI, such as in the settings app. Often the text is clickable but that is not immediately apparent. This is a flaw in their design imo.

I'll always have a soft spot for my Lumia. I loved it too

Obviously Cortana was a laggard, but I wonder how much ROI any of these voice assistants are really delivering for anyone. Besides maybe Alexa. I have not observed anyone talking to their phones since these were a novelty.

> I have not observed anyone talking to their phones since these were a novelty.

Note that people are much more likely to talk to their phone in private than in public. It’s rude/weird to talk to one with people around. I use mine at home or when away from people.

I use mine for the following:

* turn on/off the lights * play some music * what’s the weather * call people * add X to my grocery list * start a timer * occasionally, open an app. Usually on apple watch where this is faster * Remind me tomorrow to do X * Get me directions home * Show me pictures of X * Start/stop time tracking timers

The common thing is they’re all simple functions, quick to say, and faster by voice than going 3-4 steps into a UI and typing stuff out.

This is on iPhone. I imagine all of the above use cases work well on android, and I think others would be effective too, as google assistant is reputed to be more powerful.

Note that siri used to be garbage at understanding me. After using it a while it seems much better. Either she learned or I learned to speak differently.

I forgot to note that I have the airpods with always on hey siri. That has increased my usage as I don’t have to touch the phone to start the above.

I talk to my Google Home pretty regularly. I use it to set timers when cooking, and ask it about the weather about once a week. My car has Android Auto or whatever it's called. I can push a button on my steering wheel which activates the voice assistant. I tell it where to navigate me to, and sometimes ask it things like when certain stores will close. While I'm driving, it reads any texts I get to me.

Using it is one thing. I'm wondering about ROI. Alexa has a whole ecosystem that built around buying stuff from Amazon. Google Home is a solid #2 in the home assistant space but what does Google get out of it?

I think like many Google products, Google Home can best be understood as a moat protecting their search business.

Maybe voice search will be a big thing, maybe not. But if it proves to be one, Google needs to have an established product in that space before it takes off and they're left playing catch-up.

The other day I said “Hay Siri, take me to Buena Vista park.” She responded “the entrance buena vista Avenue?” and I responded: “oops I actually meant corona height park” and she instantly started navigation there. I actually held a productive conversation with a computer. Magical.

The issue with voice assistants is frequently after they get the thing wrong the first time I might not bother interacting with them again. Especially if there are other people around.

I use and see others use smart watch triggered from time to time. Occasionally I see people use Bluetooth headphone triggers.

Believe it or not a lot of use is out of the public eye, though. Lots of usage is in cars (where there has been a sharp shift to phone powered interfaces), and as an assistive tech. They are great for breastfeeding moms as an example.

I don't think the voice assistants on phones are as useful generally as a smart speaker. The phone is typically in your hands already, so just touching it is often faster. The few times I use Siri are to set alarms, and that has more to do with how bad (imo) the time entry control is on iOS.

I do have smart speakers and I use them every day to run my lights, music, set alarms, etc. It is wonderful to be able to say 'Alexa, next song' while in the shower and have it just work (with a Sonos speaker, btw).

I have exactly two uses for voice assistants: while driving and while cooking. Saying "Hey Google, set a timer for 15 minutes" when cooking or "Hey Google, remind me in in twenty minutes to add cilantro to the grocery list" when I'm driving and I remember something is cool.

Mostly voice assistants are a joke to anyone outside those trying to prove they are an early adopter.

My own experience, having given Siri and Googles alternatives are fair chance on my phone, went “Hey Siri/Google. Do this thing... OK why did I even bother? It would have been faster to just do it myself”

I use them a lot to send people messages while I’m driving.

ROI? You are literally listening to the market. And it's speaking to you. Yes, there are privacy issue. But,sadly, there are work arounds, if not public apathy.

I use Siri once or twice a day, mostly for setting alarms.

Do Android users who are part of the Google Home or Alexa ecosystems not use their phones for `OK Google` or `Alexa`? Or do such people only use dedicated Google/Amazon hardware?

I'll use Google Assistant on my phone pretty much only when driving. Other than that, I've customized my home screen to the point that it's faster just to navigate to what I need.

I use a couple of commands regularly. "Navigate <destination>", "Play <song>". But the thing is, 4 years ago when Google Assistant launched the hotword was unreliable, the query understanding was very limited, and the set of available actions was also very limited. Today the hotword is unreliable, the query understanding is very limited, and the set of available actions is also very limited.

I couldn't name a single thing that has improved my usage of Google Assistant since launch. In fact, Assistant isn't any better for my usage than the previous thing, Google Voice Search, available years before. Whatever all the people working on these assistants are doing, it isn't the right thing.

I dunno. I was pissed off when Google removed ability to listen to wakeword on Nexus 6 with the screen off. I used quite a lot of stuff through it.

I use Siri regularly to add reminders and control Apple Music while I’m driving and that’s about it. Hands free is the only utility IMHO.

Somewhat related, there is "keyboard free" usage for e.g. AirPods and HomePods. Music is a major usage domain for both, the former can also be useful for maps.

> Microsoft is tweaking the software to better compliment apps like Outlook and Teams.

I wish MS didnt think shiny when thinking about these two apps. Outlook and Teams (at my workplace) log me out from laptop and mobile without warning and send me an SMS OTP I have to use within 5 minutes. They don't even have a resend OTP so I have to restart the apps and hope the SMS arrives on time. Outlook web is slow as molasses on Firefox and of course years behind Gmail.

Yeah, that happens to me. I leave my desktop turned on a occasionally get authenticator prompts at random times when not using the computer. And then see the login dialogue when I log back in with no idea which app needs to authenticate.

I've had issues with Outlook on Web in Firefox, mostly when I have the tab open for a long time. If you haven't tried, I suggest reloading the tab when it gets slow. It has been really fast for me otherwise.

Contrary to popular HN opinion, it's not uncommon nor a user-trust catastrophe for large corporations to turn down products, services, and features over time.


Remember: You can't please HN crowd.

Ideal corporation according to HN commenters:

1) Moderately successful, but not too successful (my Foo can't compete with Bar)

2) Run by at most 20 people (why do they need so many people, me and my buddy could do it in a week!)

3) Maintains all products/features indefinitely even if less then 1% of the users use it (Google killed RSS Reader)

4) Directs all earnings back into the product (they have so much money, surely they can afford to all maintain all products/features indefinitely)

5) Pays below market rates (engineers make too much money)

6) Grows organically with no VC/investor money

7) Makes no mistakes. All experiments end up being total hits / cash cows forever.


>Maintains all products/features indefinitely even if less then 1% of the users use it (Google killed RSS Reader)

1% of all Google users is what, a gazillion people?

Nice try.

1) Moderately successful, but not too successful (my Foo can't compete with Bar)

Actual issue: They use their success to bludgeon competitors to death. E.g. what Microsoft used to do to Linux, which is why we still don't have a viable Linux on desktop even though nearly everyone agrees that they only use Windows 10 because they have to and not because they want to. Plus, now that Microsoft has apparently "done a 180 on open source" (read: we used the bludgeoning time to figure out how to profit from OSS), their success is the equivalent of breaking a sleeping man's knee and handing him a stick to walk, and then patting yourself for your generosity.

2) Run by at most 20 people (why do they need so many people, me and my buddy could do it in a week!)

Actual issue: They don't provide support commensurate to the number of people building it once the project's hype factor goes down. A good example is nearly every not-very-cool-anymore OSS library which was announced with great hype 4 years ago, is not used a lot today, and the employees of the BigCo treat everyone who raises GitHub issues with general disdain and disinterest. Although, to be fair, I think this is the fault of the suckers who go and use this OSS library. As they say "Fool me once..."

3) Maintains all products/features indefinitely even if less then 1% of the users use it (Google killed RSS Reader)

Actual issue: Unlike BigCo, which can kill whatever product they want and not face the smallest problem, the user base of these products in terms of actual usage is so big, it can support a company or two or twenty. People who are investing the time to build on your bloody platform are regularly treated in the same way Mark Z thinks of his user base ("dumb fucks").

4) Directs all earnings back into the product (they have so much money, surely they can afford to all maintain all products/features indefinitely)

Actual issue: Not actually committing resources to product support.

Here is something which should make you laugh, if it didn't also make you cry in case you are a Google Cloud developer (you cannot remove yourself from a Google Cloud project that someone else added you into):


The issue is marked as priority: P2, severity: S2. Really? And Mr Thomas Kurian wants developers to evangelize Google Cloud when they wouldn't even take care of these kinds of things for over 5 years?

5) Pays below market rates (engineers make too much money)

Actual issue: Hard to say, because on the one hand, they are definitely overpaid with respect to the difference between their actual performance and their own perception of it (for e.g. even the biggest supporters of the tech that comes from BigCo, when pressed, will point out a laundry list of bugs). Not to mention the perennially lousy support on the projects deemed by the engineers as not-sexy-anymore. On the other hand, if some of these folks end up going to jail for things such as trying to manipulate elections (which I firmly believe is a question of when and not if), then you want them to be compensated for that risk.

6) Grows organically with no VC/investor money

Actual issue: Taking investor money is just a neon warning sign today: "expect user hostile behavior in our product very soon". If the vast majority of VC backed companies didn't become user hostile, this wouldn't have been a big issue.

7) Makes no mistakes. All experiments end up being total hits / cash cows forever.

Actual issue: these experiments are often accompanied by extreme hype and overpromising. To be fair, once again, I say "Fool me once..."

Did you notice I finally put most of the blame on the people who trust BigCo, because they repeatedly act like suckers? There is indeed a sucker born every minute, and the surest sign of that is the fact that people are actually signing up for something like Facebook Shops. So the HN crowd is the voice of the skeptic. I actually hope it is never pleased about anything.

Not a catastrophe. Just not a sign of goodwill, when done with such wanton disregard for the user.

If only they will shut down the windows app. The darn thing eats up 5-20% of my cpu and my VM doesn’t even have a mic. I’ve tried deleting, disabling, cortana’s process comes back and does nothing useful.

Give me windows subsystem on Linux any day (for mainly this reason). I need a windows machine but it's so jarring going from my Linux laptop running off a HDD running six months worth of updates and a kernel while I watch a film to windows needing a restart to update a .NET minor version(!?) when running off an SSD.

I really like WSL, but I can’t turn off cortana from running at the same time.

I can only imagine that both users of the curtana app must be devastated.

> To be eligible, you'll have to have used your Invoke speaker to ask something of Cortana between July 31st and before today's announcement.

I'm assuming that "July 31st" means 2019? Otherwise, that's a few-hours window.

So Microsoft Cortana will have had about a 7 year lifespan, just like the Cortana in Halo it was named after.

So let's see if it goes Rampant, found an ancient galaxy guardians, and return to Earth for that "mantle of responsibilities"

Impressive speed of obsolescence.

Should manufacturers be forced to support devices for a set amount of time (5 years?) or should the customer start demanding it?

Yeah they just abandoned everyone who invested in their Lumia phones. I had difficulty finding batteries after 2 years for my 950s.

I’ve never used Cortana and tried in vain to remove it, what exactly is it supposed to be used for?

It is supposed to be a virtual assistant like Siri or Google Assistant.

The only place I ever found Cortana to be useful was on my XBox One. At some point it stopped responding though, so I stopped using it. I suspect if I did some work I could get it working again, but all I ever really used it for was to pause/resume Netflix.

Applications are open for YC Winter 2021

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact