I was on the DVLUP project where Nokia and Microsoft attempted to inject energy into windows phone app development. We could see the tension between the two companies as we were a 3rd party contracted by Nokia to build the platform. The Nokia exec we were in contact with was fantastic, and really tried to make Windows Phone a success. Unfortunately the Nokia IT teams we worked with were not happy and essentially tried to sabotage through inactions (we just needed OAuth / SSO to link accounts and track app installs, it took over 3 months of email chains within Nokia).
People blame Stephen Elop. But the person in control of the Nokia board was former CEO Jorma Ollila who headed Nokia through its glory days and had a lot of power. Nothing happened in Nokia without his approval during that period.
The board recruited and invited Stephen Elop. Part of his appointment was the board handing over the company on a silver platter to the new CEO. Negotiations for the acquisition started almost right away in secret. And most likely there were high level discussions ongoing with Microsoft and Steve Ballmer before Elop's appointment.
Jorma Ollila retired as a CEO around 2006; he wasn't really pushed out and Nokia was still performing well. He handed the keys to the company to a non technical CEO with a financial background (Olli-Pekka Kallasvuo, aka, OPK internally). That was the first mistake. OPK was a bit hands off throughout the Iphone and Android rollout. There was no vision, no leadership, just a lot of infighting between heads of various units.
The second mistake was selling out to Microsoft and all the share value implosion that came with that. Microsoft bought several companies over the years. Nokia was one of the smaller ones. That's how bad it had gotten. At the peak Nokia was worth 150 billion or so. MS bought the phone unit for 5 billion. The later Linkedin and Github acquisitions were worth more.
By the time Stephen Elop was brought in (by the board, headed by Ollila) to fix things, it was too late. There were a lot of internal battles as well between the big business units. A whole string of CTOs with no power whatsoever that were appointed and let go. Including Scott McNealy who never really made any impact and was there only briefly.
> Jorma Ollila retired as a CEO around 2006; he wasn't really pushed out and Nokia was still performing well. He handed the keys to the company to a non technical CEO with a financial background
I wouldn't say that Ollila had a technical background either. Ollila has 3 MSc degrees, 2 in economics and 1 in engineering. But after graduating for the 2nd time, he worked first in an investment bank (Citibank), then in the finance department in Nokia before rising to the executive level. I would say he has a financial background.
Was this around the time Microsoft kept switching what "windows phone" was? I remember that the one friend who got into it loved it, and then they released something completely incompatible and he abandoned the platform.
2009 and earlier: Windows Mobile was based on WinCE. The UI was garbage but the innards were pretty functional, and there was desktop-like multitasking. Unpopular opinion: they should have just done a UI refresh of that thing and moved it to an NT kernel. There were a lot of cool third party hacks on this platform.
2010: Windows Phone 7 was still WinCE, but they removed full access to WinCE APIs, and got rid of PC style multitasking. They had a new UI framework for first party apps. Then for third party apps they had a port of Silverlight that imitated the new UI style. The latter had really terrible performance.
They had to base this release on WinCE because the NT kernel port to ARM wasn't ready yet. Blocking access to "good" APIs could be seen as a way to ensure app compatibility for the next release.
2012: Windows Phone 8 had the NT kernel. Also, windows 8 and windows RT shipped. But the silverlight-inspired UI framework of Windows 8 was different from the Silverlight fork from Phone 7. So you had yet another UI framework rewrite to cope with.
At the time Steve Jobs was putting his foot down against allowing Flash on the iPhone because the performance was so pants, Microsoft was going all in on Silverlight which had exactly the same problem.
The first iPhone had a 400Mhz processor and 128MB of RAM. It was so resource constrained that you couldn’t put a picture on your home screen because it would have taken too much memory.
It could barely run Safari. If you scrolled too fast, you would see checker boxes while trying to render the screen.
When Flash did finally come to mobile on Android, it required a 1Ghz processor and 1GB RAM. The first iPhone with those specs didn’t come out until 2011.
Even then on Android, Flash ran horribly and killed your battery. I had a high end Android phone on Sprint back then.
Wait, it wasn't about performance, but it was about Performance?
It was that a pseudo-machine/VM approach put the VM IP owner (Adobe, Sun/Oracle) in driver's seat for control of the product's precious HW resources while letting their affiliates define the UI. What could go wrong, knowing that to invite in the vampire of their bloat & risk was to give those IP owners a competitive leg up to override all your design choices and serve their own markets, contrary to everything Jobs had done to rescue Apple from its clone wars.
Oh, and that Flash and Java were the world's most popular malware/APT delivery vehicles at that that continued to wreck PCs for many years after 2007.
But it wasn't about performance!
Or why Jobs choose to not drive a stake into his own heart to defend from vampires.
It would be interesting to see a companion presentation from the POV of Cingular/ATT. They likely also were very surprised and entertained by Jobs' device!
It's easy to forget how popular flash was in that decade. A lot of us found it annoying on desktops too. Not to mention Linux, where we'd deal with binary blobs that were pretty unstable, not because we liked it but because you needed it to interact with the world.
I have not so pleasant memories of having a few different versions of their plugin and I'd try to figure out which one worked for a given website, symlink the right one and restart the browser. And that was the way to watch videos online...
Flash as an animation tool and applet platform was already on the downswing when the iPhone happened, though.
The consumer demand for Flash on mobile seemed to be mostly about video streaming, because at the time Flash was experiencing sort of a second life as the least-bad way to do streaming video on the web. In that context Apple's point of view of "as an industry let's finally fix browser-native video streaming, rather than being stuck with Flash forever" seems pretty reasonable.
Yes, I also think around 2008 or so the most widespread use of Flash might already have been newgrounds et al. I don't remember really ever caring for Flash on Linux though.
I do remember writing CMS backends for Flash websites in 2001, but that was the early time I think, before AS3 and really cool stuff.
Oh, the flashbacks.. (pun intended). Same here. Every new flash release, download, extract, rename to have a version number, copy to "folder of last 10-15 released flash .so files", symlink, restart browser and hope it works.
I think it got to be so common that firefox supported reloading the library without restarting the browser if you changed the symlink and opened the "about:plugins" page.
And then they started releasing both 32-bit and 64-bit versions...
Your 90s area PC also had disk swapping and wasn’t running on a tiny battery. The Flash of 2007 was much more processor intensive and in the 90s, I doubt you were streaming quality video with Flash.
Mostly because a whole generation lost that part, and finds cool putting Web everywhere, saying this as someone that also does Web projects, I only don't see a value using it as a hammer for all kinds of nails.
Apparently the whole Windows UI mess is also related to Microsoft not able to hire new folks with Windows development experience, probably they only saw Win32 after joining Microsoft, funny how things come around.
> When Flash did finally come to mobile on Android, it required a 1Ghz processor and 1GB RAM.
It worked on Maemo years before that though, with 600MHz Cortex-A8 CPU and 256 MB RAM. Nokia N900 had out-of-box support for Flash in its Gecko-based browser.
I believe Symbian had some support before that too, but I don't remember and haven't checked the details.
I remember having a Flash app on Nokia E70. I never used it. The phone was lauched in 2006, but I don't know if the app was there from the beginning because I bought it second-hand in 2008.
> It was so resource constrained that you couldn’t put a picture on your home screen because it would have taken too much memory.
Clearly you drank the Apple koolaid that later artificially limited wallpapers to 3GS (or 3G?) and above when they introduced the feature in later iPhone OS updates.
We had wallpapers and great homescreen and dock themes on jailbroken iPhones without a significant perf impact.
P.S. Contemporary Windows Mobile phones had Texas Instruments OMAP ~200MHz processor IIRC with less RAM and iPhone (2G) was comparatively great.
> We had wallpapers and great homescreen and dock themes on jailbroken iPhones without a significant perf impact.
Untrue. There was a noticeable UI lag when scrolling between app pages. I've tried it in both the iPod touch and previous generations iPhones. It felt like how Android used to feel like back then.
Yes “I drank the Kool aid” when Adobe couldn’t get Flash to run decently on a 1Ghz/1Gb RAM Android. But it was going to run smoothly on a 400Mhz, 128Mb RAM first gen iPhone?
Was Safari with Flash going to run well when Safari without Flash could barely run?
I didn't read a word about flash in the comment you replied to. They commented on the mention of wallpapers in your comment about flash, but they didn't mention flash at all. What they said is that you believed things that Apple said, that weren't true, about why they wouldn't allow wallpapers. They characterized this as a nitpick.
But back to wallpapers - while the jail breaking community didn’t care, between performance (lot easier to redraw a black background), memory and battery life, background images would have adverse affects on the iPhone. it wasn’t that it couldn’t be done.
> It was so resource constrained that you couldn’t put a picture on your home screen because it would have taken too much memory.
This is the sentence in your original comment I had responded to (and I quoted it in my original comment, not sure where's the misunderstanding here). iPhone was resource-constrained, but not that resource-constrained.
I do agree with your characterization of Flash being slow and clunky at the time for the most part, hence prefacing my comment as "nit," although I do not for one second believe that's the primary reason Jobs killed it. If he wanted a fast Flash, he would have made Adobe dance to his standards.
And it made the UI slower - as confirmed by another comment and used battery.
> Untrue. There was a noticeable UI lag when scrolling between app pages. I've tried it in both the iPod touch and previous generations iPhones. It felt like how Android used to feel like back then.
How was Jobs going to force Adobe to get Flash to run on a first gen iPhone when they could barely get it to run 4 years later on phones with 8x the memory and 2.5x faster?
Apple struggled to get Safari to run.
As another counterpoint. Google and Motorola tried to release an “iPad Killer” with the Motorola Xoom promising it would have 4g and Flash. Adobe was late releasing Flash for Android tablets leaving the Xoom in the unenviable position that you couldn’t visit the Xoom product page from the Xoom itself because it required Flash.
Haha I remember the Xoom. Company bought one at the time to test our product on, since it might be a big deal. Several of us were intrigued and negotiated who would get to take it home after work. After about 3 days it became clear that it was terrible and it was relegated to a drawer.
Wait...you're telling me I can't charm a cute girl with the Motorola Xoom and its amazing capability of being able to play a Youtube video in full screen?
PS: In reality the stuttering masses were probably using a respectable device that actually provided long term value (and probably had longer OS support than the Xoom haha).
I had the original iPhone, then swapped it for a flagship android and hated it. I tried Windows Phone out a few years later and loved it and had two of them over the years. Some apps didn't exist for it and that sucked, but the OS was snappy and something different that I really liked. The Nokia windows phones were great. I knew it was doomed though as when I got them from the carrier, I was apparently like the only person despite it having its own wall at the brick and mortar store lol. I still miss it.
A relative of mine had a Windows Phone and kept it running at least until the mid 10s. It was fast but he could basically only do calls and SMSes with it because nobody wrote apps for that OS. Everybody in app development (devs and their customers) was keen to see Windows Phone die quickly so they could spare time and money and develop only for two OSes.
Thy were really great on supporting a bunch of less popular platforms (feature phones and the like) WhatsApp supported Blackberry for a long time. WhatsApp probably supported a potato for quite a long time.
My cousin says the same thing... 25 year IT veteran. Early adopter for almost all new tech. He says his 1000-whatever Lumia phone was one of the best phones he ever owned. I know it ran Windows Phone OS, and I remember playing with it a bit.
The Lumia was essentially a N9 ported from Linux to Windows. The N9 was the best phone I have ever owned. The UI was fantastic. In particular, the offline navigation application was incredible.
Nokia could have succeeded in the smartphone market. They had the 770 since late 2005. But they were a typical corporation, conservative and plagued by internal politics. Bringing Elop on board, with his Windows agenda, didn't help either.
I remember having a Windows PDA when I was in college, and developing a bit on top of Windows M. It was a reasonable platform.
But Microsoft was too greedy with their licensing schemes and demanding too much adaptation from the hardware and chip manufacturers. You’d think they would adapt their OS and drivers, but no, you had to tape out new silicon for them. So they’ve lost the mobile OS market.
It feels like something like this may happen with the AI OS now. They are pushing hardware manufacturers to conform to their standards while Linux is adapting to what is available and working already.
I don't think so. As I recall, the different UI (not just from older MS stuff, but also from iPhone) was really front and center of Microsoft's pitch at the time.
Besides, WinCE PDAs were very much a power user / enthusiast device, with relatively few around. People who used them and thus were familiar with the old UI would be well aware that WP7 was completely different, and people that didn't use them weren't exposed to the old UI in the first place.
I think the issue was some people never even got around to seeing the different UI. They heard Windows + Phone and immediately thought of the older tech and noped out.
I had forgotten the name, but yeah, I had a Lumia for my first one. Hardly anyone I knew had one, but the ~5 I knew were absolutely in love with theirs.
Yes, the Windows Nokia Phone was quite an interesting alternative. Though I never owned one I played with one and was pleasantly surprised, the 'workflow' was very good, the UI as was nice, it was snappy. If they were around today I'd probably think about owning one.
I miss so many things besides the UI. seamless integration of Cortana with in-car bluetooth to read incoming SMS, live tiles, fantastic cameras in Nokia devices.
Yes... If I remember we were aiming for the newly released "Windows 8"-based Phone OS, and the previous version was fully incompatible with it so all apps had to be redone. Tiles were the new thing to build for.
Microsoft tried to do the same thing on the Desktop side too, but on the desktop they were forced to keep the backwards compatibility in place so it didn't finish off the platform the way it did on the Phone side.
Amusingly Microsoft is still trying to make the walled garden happen. Lots of cheap Windows laptops and Desktops ship in what is called "Windows S" mode where only Microsoft Store apps are allowed to run. But again because PC owners don't abide that kind of bullcrap they also have to supply a way to tear down the walls (it's surprisingly easy, albeit permanent: just download and run a free app from the Windows store) if you want to use the machine in a normal way.
If memory serves, it was a custom kernel and OS, then a semi-custom kernel with a few OS components shared with Windows 8, and then the Windows 10 'core' kernel (same as on the Xbox One?) with many shared OS components.
At each step they left the majority of devices behind.
What was equally worse was the triple (quadruple?) switch of app frameworks. If I remember correctly it was a dotnet abomination, then ?? then WPF and finally Xamarin.
Good luck convincing your platform 3rd party developers to entirely relearn and rebuild their app four times over in the span of a few years.
Interestingly enough, Windows Phone itself was far ahead of it's time. Buttersmooth UI, flat UI, built-in global and app dark modes, all in the early 2010s.
WinCE (which was rebranded as Windows Mobile at one point) basically had a cut-down version of Win32 as its app framework. There was also .NET complete with a WinForms port.
Windows Phone 7 had Silverlight as the app framework, which, to remind, was itself basically a rewrite of a subset of WPF in native code for perf (although the public API remained .NET).
And then after that it was WinRT / UWP, which was effectively further evolution of Silverlight in terms of how it looked to app devs.
WP7 was a really low point for the series because not only the new app dev story was completely and utterly incompatible with anything done before, it also had a very limited feature set in terms of what you could actually do inside the app - much worse than the iOS sandbox.
WP7 -> WinRT transition was easier because WinRT was so similar to Silverlight in terms of APIs (in some simple cases you literally just had to change the using-namespace declaration to compile). It also added enough functionality for more interesting apps to be viable. But by then, the reputation hit from both devs (who were being told to again rewrite everything they already rewrote for WP7) and users (who were being told again that their devices won't get the new OS, and the new apps are incompatible with the old OS) was too much for the platform, IMO.
And then on top of all that Google actively sabotaged it by refusing to make apps for its popular services - such as YouTube - and actively pursuing third-party apps that tried to fill that gap.
After WinRT transition, Microsoft sabotaged themselves, due to the way WP 8 => WP 8.1 => WP 10 happened to be, with rewrites, promised upgrades that didn't happen, deprecation of C++/CX, and plenty of other missteps.
It was incredibly smooth. The Windows Phone 7 browser was also very smooth compared to the iPhone/Android browsers of the time. Some miracles worked somewhere.
I had a few Windows phones, and butter isn't a word I'd ever use to describe the performance of the UI. Heck, I wouldn't even use the word margarine to describe my experience with it.
I had two Lumia flagship phones - Lumia 800 with Windows Phone 7 and Lumia 930 with Windows Phone 8 (which I later upgraded to 10).
Both look and feel awesome, not cheap at all. At the time, Microsoft were paying developers to port apps to Windows Phone. There were developers who took the effort to make their app look native, and I'd say Windows Phone 7 had the best UX to this day.
Didn't it end up as UWP? At one point they were trying to pitch running the same app on mobile and on desktop, and it .. kind of worked, although obviously very sandboxed and restricted in APIs.
I couldn't believe my eyes when I read that they had hired Elop and was concentrating on Windows phones. I immediately knew that was the end. Unbelievably incompetent by the board.
I think it was actually a good idea. I think they correctly predicted the Android market and saw Windows Phone was a good way to differentiate their phones from everybody else. If you look at the history of Android manufacturers, it was a long slog of brands trading off popularity and hardly making any profit until Samsung eventually became the only mainstream player.
Nokia was dead company walking before Stephen Elop. Elop saw the writing on the wall and made one of the choices available. A different CEO would have made a different choice but ultimately at that point it would still have been too late to save Nokia.
Nokia was a great hardware company that missed the boat when the market changed to be based around software. When the market changed again to be based around ecosystems, Nokia was beyond saving.
Was there with the company as intern and junior during Nokia and Microsoft days for Nokia Maps.
In my opinion Microsoft fumbled the app store by bloating it with questionable KPIs on number of added apps by anyone able to submit templates apps, while not getting key apps like WhatsApp on board. S\so it was a hard sell to have people's ecosystems. Same syndrome as with Zune, Tablet PC, and Microsoft Store on Windows.
Build quality and hardware of the Lumias was second only to iPhones and definitly better experience than Android.
The old Nokias had no chance compared to those, and I agree with the assessment that Nokia as Android-Vendor would have made little sense either.
> In my opinion Microsoft fumbled the app store by bloating it with questionable KPIs on number of added apps by anyone able to submit templates apps
Worse than that. IIRC, Microsoft ran contests which specifically incentivized developers to create as many apps as possible, and most of the apps they got as a result were garbage (like copies of developer examples with some of the text changed).
Nokia with android vendor would mean Nokia would survive until today - just due to the brand (it was big) and build quality.
They released an android phone that sold... many years too late.
If they released it much earlier (no microsoft) probablh Nokia would still be here - competing with Samsung, or in worst case the tier3 brand cheaper smartphones.
> made a different choice but ultimately at that point it would still have been too late to save Nokia.
You think if they made just a single decision different and bet on Android instead of Windows, they would be in the same spot today? I wouldn't be so sure. Samsung hedged their bets across both and succeeded. Both weren't great at in-house software and Nokia made better hardware.
I don't think Nokia at that point would have gone with Android with Google services which what the market wanted. They would have gone with Android with their own services (Maps etc) and app store.
I don't think that would have succeeded against Samsung and the Nexus phones.
But TBH I think going with Android would have a better move than what Elop did.
Nokia is still around, because NSN survived this mess.
As someone on the Networks side, with occasional visits to Finland headquarters, Nokia Mobiles would have done alright, if they kept down the Symbian/Linux path.
The Burning Memo killed the remaining trusth from app developers, in a company and ecosystem that was pretty much anti-Microsoft, just made the transition to have Qt properly integrated in Symbian, with PIPS and nicer Eclipse based IDE than the previous experience.
Only to be told to throw away all that developer experience, adopt Windows and .NET.
I think the critical failure of the windows phone was that app development was not open. You can't compete with established walled gardens by building your own, you can only compete if you make a huge amazing park free to use just outside the walls of the competitors.
Translating this to windows phones, it would have only succeeded if it either:
1) Made browser applications first-class and pushed phone-specific APIs (gyro, bluetooth, etc) to be open. Then pick a fight with google and apple about supporting PWAs better. This would probably keep windows phones as a "low cost, crappy feeling" systems forever.
2) Made the windows phone native-apps trivial to port to run on browsers with a convenient and easy way to deploy those apps on ios/android (hopefully without feeling too much not-native on those platforms). Would require a lot more engineering resources and time, so much harder to pull off.
Well they did do something like that; Windows Phone apps were written in the same .Net UWP SDK as desktop apps, so the idea was that you could target both platforms at once (and Xbox as well). I think MS overestimated how much people cared about native PC apps by that point (basically not at all). Additionally, snapchat was the hot new app at the time, and there was no first party Snapchat app (and if you used the 3rd party one, you risked being banned from snapchat).
I would argue that windows desktop development using .net is a walled garden in itself as well.
They finally realized what I was saying when they acquired Xamarin in 2016. I never used Xamarin myself, but I hear it is not that great and kinda dying. So like I said the native open platform approach was a lot harder to pull off.
I was on the DVLUP project where Nokia and Microsoft attempted to inject energy into windows phone app development. We could see the tension between the two companies as we were a 3rd party contracted by Nokia to build the platform. The Nokia exec we were in contact with was fantastic, and really tried to make Windows Phone a success. Unfortunately the Nokia IT teams we worked with were not happy and essentially tried to sabotage through inactions (we just needed OAuth / SSO to link accounts and track app installs, it took over 3 months of email chains within Nokia).