It's probably a good thing that Microsoft did not follow the advice:
> They can't hire smart people anymore, but they could buy as many as they wanted.
> Buy all the good "Web 2.0" startups. They could get substantially all of them for less than they'd have to pay for Facebook.
> Put them all in a building in Silicon Valley, surrounded by lead shielding to protect them from any contact with Redmond.
Instead, Microsoft did the exact opposite. They picked Satya Nadella who had worked there since 1992. He has driven the innovation right out of Redmond, WA with no lead shielding required.
This highlights the magical thinking we can suffer from in Silicon Valley where we believe we’re inherently better at solving problems here. It sounds like what Paul was saying is good ideas can no longer come from Redmond and the best they can do is buy a chunk of Silicon Valley and hope to ride on whatever success that nets them.
Agree. Two of three largest companies in the world are now in the Seattle area with both Microsoft (#1) and Amazon (#3) thriving. Apple (#2), Google (#4) and Facebook (#5) are in the Bay Area.
I don't know about Apple but Facebook and Google have pretty big offices in Seattle as well, along with many other startups (Uber, Airbnb, Stripe, etc.)
It seems the play these days is to have the best of both worlds.
Companies at that level generally have offices in big cities all over the world. In Toronto we have Airbnb, Uber, Google, etc offices for example. I think the important thing in this context is where the HQ is.
And Microsoft and Amazon both have big offices in Silicon Valley. Bing is (was? it looks like the site is now under reconstruction) based out of their Mountain View complex that's almost literally across the street from the Googleplex, while A9 (Amazon's search subsidiary) has a big office in Palo Alto.
Under Satya, Microsoft went on a shopping spree. They bought a ton of startups: Minecraft, Github, Skype, Sunrise, LinkedIn (tnx comments!), Yammer ... and along with those it bought a mapping company, a multitude of ad/marketing companies, educational startups, AI startups...etc, etc. Those are just the ones we know about.
Did you know they almost bought Twitch.tv, but Amazon had more money?
And as far as locking them in the same building, that's not exactly operationally efficient in 2018. But they seemed to have done their best at either merging the products right away, or letting the team run on their own.
2) Most of the ad tech companies were also purchased by Ballmer.
3) Minecraft and Skype are not Bay Area companies
4) LinkedIn was a public company when Microsoft acquired them. So it seems like a stretch to call them a startup.
5) And finally it looks like Microsoft has done acquisitions throughout their existence. In fact, they acquired as many companies in 2007 as they did in 2018 and more in 2007 than 2017.
It doesn't seem like "buy lots of Bay Area Web 2.0 companies to infuse new talent" is the secret ingredient. Yahoo was the company that probably closest executed the strategy. That didn't go well as we all know.
Man, I worked at aQuantive when it got bought for $6B by Ballmer. They took a $6B write down within a few years. They should have kept their money in the services part of that business instead of the second-rate ad tech.
The other key part of that idea is the “lead shield” from the dinosaur parent company, and we know from many accounts that being integrated into the Yahoo culture is what killed their acquisitions.
Microsoft acquired Skype in 2011, before the 2014 Ballmer/Nadella change-over. Also, Minecraft isn't exactly a start-up, but more of a celebrity among gaming projects, ranking closer to an endorsement/sponsorship deal.
It seems like Big Tech has switched over to amassing data and behavior manipulation ("engagement", etc.) as their main business model. I think that Microsoft didn't feel threatening because they still mostly just sold a product.
I feel like we as a society haven't really entirely got a handle on this yet. Things emerge, like trusts, and it takes a while for people to understand what has happened.
But, with things like Windows 10 telemetry you can't turn off, and Windows turning into a service, they've gotten back on board with how Big Tech runs these days.
This pattern seems to hold really well. The giants mostly amass data and manipulate people. If they don't do this sort of thing, they don't really seem that threatening, regardless of their size. HP, IBM, Dell. Apple seems interesting, as a hold-out on just selling a product, but in the meta game, just direct manipulation of people really looks like a step up, and I don't predict they'll continue as anything approaching a "scary tech giant" in coming decades unless they change like Microsoft did.
But, with things like Windows 10 telemetry you can't turn off, and Windows turning into a service, they've gotten back on board with how Big Tech runs these days.
Yes, I can confidently say that I miss the old Microsoft. They were strongly closed-source and proprietary (not that it didn't stop people from understanding how things worked anyway...), but they did not do the sort of data collection, behavioural analysis and advert-driven manipulation that the others were doing, and still highly valued backward-compatibility and stability.
Now they're going the same route as the other tech giants, silently collecting information while also open-sourcing a lot of other things almost as a form of distraction/diversion. You could consider the fact that they're open-sourcing something to mean they're not interested in selling that as a product anymore, and that's because their profit is coming from somewhere else.
It was flagged dead on HN because the perception was "do no evil' Google good, M$ bad.
They tried harder later on with the somewhat cringy Scroogled campaign and found out the public didnt care while Google laughed all the way to the bank while increasing data tracking, collection and usage. And here we are.
The interesting thing is, how battle-proven is the current ad/monitoring obsession as an actual business? It's trendy, but not every successful firm has to-- or should-- get on board.
Microsoft's classic businesses are pretty comprehensible. Software (boxes or as a service) and cloud services are pretty solid business models. They don't require unrealistic expectations or gimmickry to work. There might be shifts within that market (the MSRP of a browser is 0, and that of an OS is plummeting towards 0 quickly), but you can realistically say "I want to sell a software package" or "I want to run a cloud service provider" and make a business case without fantastic assumptions.
In contrast, much of the value of Google and Facebook and similar companies is based on anticipation of the future value of those big data stores, a largely unproven asset.
Will consumers reach an "uncanny valley" moment for personalized marketing messages, resulting in a downturn in click-throughs?
Will advertisers always be willing to pay a premium for improved targeting? I could see a competing product offering spray-and-pray banner ads with .01% click-through rates at a few cents per click outperforming hyper-targeted ads that draw a 5% click-through rate at several dollars a click.
Will they be able to successfully retain, grow, and monetize their data hoard after the inevitable regulatory backlash? (I could imagine a sufficient data breach resulting in regulations that make the GDPR look small potatoes)
I find windows firewall is very good at blocking programs/processes from accessing the network.
By configure it to block svchost.exe from access network except 192.168.0.0/16, the windows 10 system service can have very light load on my laptop.
Combine with ProcessExplorer to monitor various processes's network traffic, I can make my laptop where the only two programs can access the internet are Firefox and Chrome.
Once every 6-9 months, when I feel like it, I do release the svchost.exe from the jail for a day to free win10 to get the latest security update.
With this setup, I can make the Win10 home edition laptop to have system uptime of 6 months or more.
10 years ago I complained about Microsoft, but basically was happy with windows and recommended it to people. Now I'm just creeped out by windows. I really wish they had left the ads and data collection out, the trust is just gone.
How does your model for the fact that a lot of MS growth is coming from things like Azure? [0] This is pure IaaS (compute/storage/networking + managed services) and it's increasingly popular. To the extent there's telemetry involved it's mostly to manage services effectively.
But, with things like Windows 10 telemetry you can't turn off, and Windows turning into a service, they've gotten back on board with how Big Tech runs these days.
They sort of have no choice but to play in this space given that Google does, they can’t just surrender that territory. But if the “attention economy” went away tomorrow MS core business would be fine, how much actual revenue do we think they even make from Windows 10 telemetry? Compared to Azure and O365?
I think this paragraph really sums up the whole article for me:
> The last nail in the coffin came, of all places, from Apple. Thanks to OS X, Apple has come back from the dead in a way that is extremely rare in technology. [2] Their victory is so complete that I'm now surprised when I come across a computer running Windows. Nearly all the people we fund at Y Combinator use Apple laptops. It was the same in the audience at startup school. All the computer people use Macs or Linux now. Windows is for grandmas, like Macs used to be in the 90s. So not only does the desktop no longer matter, no one who cares about computers uses Microsoft's anyway.
If you live in a forest you'll think the world is made of trees.
Paul Graham is a very clever man, and I have a lot of respect for him, but he lives in a high tech bubble. People in high-tech start-ups use apple; people who work in creative industries (mostly) use apple; most of the rest of the world use Windows PCs. So to declare the desktop (does this include laptop?) "dead", especially back in 2005, is most definitely premature.
Question: has anyone out there come across a non-high-tech multinational corporation that uses apples instead of PCs?
Even within FAANG as a developer You the developer gets a choice of MBP vs PC. But PG was alluding to the specific fact that Macs outnumber PCs within the YC bubble which still holds true.
> Windows is for grandmas, like Macs used to be in the 90s. So not only does the desktop no longer matter, no one who cares about computers uses Microsoft's anyway.
This sounds like a very ignorant thing to state. Wherever I worked, the best programmers used Windows and Linux desktops. Why? Because they too busy coding to get caught up in the Apple hype is my best guess. You know just like he talks about being ignorant to what was going on in Microsoft world, many coders I worked with did not know or care about what was going on in Apple world. There was one group of people who I did notice flock to Apple OS's, wannabe developers. The type that talk a lot about new tech but never wrote code that got deployed to production.
As a Windows/Linux developer my reason is that both Windows (at least historically) and Linux offer far more opportunities for customisation than macOS, as well as being far less about hiding things and dumbing-down for the computer-illiterates, or locking the user out in the name of security.
I agree with your observation about those who prefer Apple products: it seems to be more of a fashion statement.
MacOS has been a buggy mess for years. From 2008 to 2014, I used to have three screens in front of me: Fedora (local dev server), Windows (Office and business apps), and OS X (iOS dev).
I had by far the most problems with OS X. Windows has been the “just works” OS for me since Windows 7. I’ve had zero bugs and zero crashes. I know that’s not universal, but I’ve spent thousands of hours with various Linux distros, MacOS, and Windows, so it’s also not exactly anecdotal.
The opinion part of that matters a lot less than the silly elitism of “good developers choose ____” OS. I like to be able to customize every aspect of my OS, so I preferred Linux for a long time and now typically use Windows (because it’s customizable enough and easier to use).
Well, I've just left a company where I have been using Mac OS for 6 years. IT recommended that I don't use Linux because the drivers for the Mac hardware are supposedly buggy and slow... so I had to go on with Mac OS.
I've never had so many problems with a computer since Windows 95, and it was consistent across two different model Macbooks. Being unable to hold on to the correct resolution on external display, problems with peripherals (e.g. not detecting a mouse until after a reboot or taking out/putting in Apple mouse batteries), every upgrade of the OS breaking something or requiring re-installs (XCode, etc.), grey screens of death, Terminal crashing(!), mic stopping and needing one to kill the audio driver to work again (that hit many people, hilarious in meeting), insane memory usage (on a 4Gb laptop it's simply impossible to do work because it's swapping pretty much half the time... compared to Windows 7 that is slow but serviceable on 2Gb with similar workload; 16Gb laptop was better), VPN mysteriously dying (works after reboot), GPU driver hanging and causing a reboot ("relaunching WindowServer"), slow boot time compared to a Windows laptop from like 2014, and many more issues. Some issues with memory and scheduler that I looked for in-depth explanations of, I was thinking "is it really possible to f-up Unix so badly? You need some sort of a special talent."
And almost every time I'd go online and there are tons of people having the same problem for years with no resolution, just like Windows 95.
When I started using it, I'd say "lol just like Windows" on every glitch, nowadays when something wrong/slow happens on Windows I say "lol like a Mac"
Oh, also - needing 3rd party software for simple things like setting high-enough mouse sensitivity, window maximization that is not idiotic on multiple monitors, window snapping, or system-wide equalizer.
Really I found the only good thing about Mac OS is a terminal (that is until Terminal app started crashing after some update, taking all my tabs with it - a known bug months ago, still not fixed). But then in Windows I set up ConEMU + MinGW64 for only slightly inferior experience, in like an hour including research.
The fact that you're talking about Terminal and not iTerm makes this whole story very suspect, or at least puts you into a category of "people who never really tried to get the most out of their MacOS experience".
I don't really care much for bells and whistles around the command line, as long as the command line is good (which it is, because it's Unix - I just wish they'd stop making changes so it won't crash).
Yeah, I never really tried to get much out of my Mac OS "experience" because I just want the OS to allow me to find stuff and launch stuff. I found most UI (well, not UI itself but various UI features) outside of actual applications (e.g. the IDE) plain annoying, just like the same crap in Windows 10. Spotlight is on par with Windows 10 search. Finder is worse but more or less the same.
However, I do want my monitor to stop blinking from 2560 to 1920 for no reason, and my terminal from crashing, and to stop saying "hey I can't hear you, kill coreaudiod" in meetings. And actually come think of it, it would be nice to open files from non-open dialogs (I don't know why the context menu in dialogs cannot just be a normal context menu ), and also can I not have the thing where the files created on the date that is after the date the app was opened, show up in "No Date" section at the bottom of its open/save/etc dialogs, when grouped by date? That's really annoying when you want to attach a latest file.
IMHO iTerm isn't all that great. I have problems with common keyboard shortcuts that work in every other *nix terminal emulator (eg forward over word). The response seems to be "just redefine these keyboard shortcuts yourself!"
What exactly does iTerm do that makes you regard it better than Terminal?
Input broadcast, for one. Insane customizability for another -- I have scripts that, in 1 click, can SSH me to all hosts for a service (dozens). That's super valuable.
The fact that you're talking about SSHing into dozen hosts and not e.g. pssh makes this whole story very suspect, or at least puts you into a category of "people who never really tried to get the most out of their terminal experience" :P
Definitely true, but I'm not on here defending the merits of the terminal!
I'll check out pssh. The cool thing about iTerm input broadcasting is you can turn it on and off rapidly, so you can start out broadcasting, stop for a sec to adjust one host, then resume. pssh looks like you'd have to establish a separate connection.
Not being alone and being correct are entirely unrelated.
Also, an OS having bugs and an OS being "terrible" are also entirely unrelated.
Should I just link the bug trackers for each of the desktop environments available for Linux? Should I link the Linux kernel bug tracker (not that either such thing is actually fair)?
You're flamebaiting more OS wars, and honestly I'm pretty sure that's explicitly disallowed on HN, per the rules.
Like, really - do you not realize how biased you sound? Or rather - how _few_ people will agree with you, making the "statement of fact" pure flamebait?
BTW, please note that on Windows I can use various types of Linux variants right from the Windows App store. So the excuse of having a Nix shell on Mac OS is no longer a point for Mac OS. In fact, because we use Ubuntu in prod, and I run Ubuntu on Windows, that point might now be for Windows for many users.
I still prefer using a real nix instead of the windows subsystem for Linux. Mostly because the file system integration with WSL is still bad and Windows still doesn’t come with a decent terminal emulator.
Linux, the kernel, is one of mankind's greatest achievements, I do believe this. The Linux desktop experience, however, is a living nightmare. Your computer itself becomes your second hobby, in addition to whatever you do on your computer. For a professional who's at work to produce novel software, that creates an additional time-sink they don't have the luxury of indulging.
This isn't even really a debate for developers, and pretending like it is just baits the idiotic "Great OS Wars" debates that have zero rational actors. "What's the best OS (for x)" has been, for probably 25 years, a flaimbait level question.
Besides, I use "all" 3 OSes (MacOS, Linux, Windows) for various things, why does this have to be a, "YOU CAN ONLY PICK ONE!" conversation? I could happily develop on Linux or Windows, I am (along with most in the industry, based on my conversations with people over the years) just happiest on MacOS.
"just happiest on MacOS" is far from what you wrote earlier, "MacOS is, by far the best generic developer OS". This points to a problem I notice, hyperbole and exaggeration when discussing software development tools and technology.
Because it is the best generic developer OS (it's *nix + good desktop environment, no Linux distro can even come close to the latter), I am also just happiest with it.
> The last nail in the coffin came, of all places, from Apple. Thanks to OS X, Apple has come back from the dead in a way that is extremely rare in technology.
The last sentence is ironically funny now that MS is back on top and Apple is slipping, and the article is about the "death" of MS.
> Microsoft closed Friday with a larger market cap than Apple’s, making the Redmond software giant the most valuable U.S. stock.
The funny thing is Google is the new Microsoft for precisely most of the reasons he mentions. The worst part of Microsoft's culture (empire building by middle managers) has now infected Google as well. Amazon is the only one that has been able to stave it off which seems partly due to their insanely customer obsessed culture.
The analysis fails because it looks only at technical and cultural determinants for what is primarily an economic process.
To counter this with my own dubious and oversimplified explanation, Microsoft did not fail because it has created a monopoly, a walled-off ecosystem they control, and they managed to protect it in most desktop segments. Google has it's own monopolies, as do all of the successful companies that are still around. Microsoft's failure to extend their monopolies into mobile and web costed them the crown of the tech industry, but they're far from dead.
Apple rebooted solley on it's iPhone walled garden - their desktop was moribund and the iPod, while an inovative product and short term cash cow, was not conducive to a monopoly, so it was quickly duplicated and made obsolete. Yahoo was not able to create and maintain its monopolies, so it's dead.
Crisis is the time when people are most willing to accept changes, and Microsoft was indeed in crisis. Still is in some ways, but their turnaround has been impressive.
When PG wrote this down, perhaps the vast majority of Microsoft board agreed with him. It was a giant company without a roadmap. A company that always behind.
Microsoft has changed a lot since, and still putting much of an effort and money to gain trust amongst the community. Just remind yourself the reaction when GitHub acquisition announced. For three weeks I read blog posts filled broadcasting fear and loathing as if the government of China has bought Facebook, Twitter and Slack altogether.
I recall seeing a book in the cutout bin of a bookstore that made the case that IBM was an unstoppable juggernaut that would inevitably take over the world. It was written in the mid 80's. The irony was it was in the cutout bin because it wasn't long after that IBM was well on its way in its slide to irrelevancy.
I wish I'd bought the book, or at least remembered its title.
Well few decades later IBM is still generating billions in profits. So I'm not so sure what to conclude from that, other than that juggernauts, while maybe not unstoppable, do not die easily.
I remember a newspaper article cut out and posted on the notice board in the Computer Science Dept at my University, probably around 1986, entitled "Looking forward to the day when IBM rules the World".
It's easy to laugh now, but in 2007 this was spot on.
It was 10 years before Microsoft managed to drag itself up and do something useful with itself.
It's really remarkable that Microsoft managed such a big turn around after such epic failures (like mobile); others (like IBM...) are still struggling to do it.
Just goes to show, competitive pressure is a good thing. :)
In addition to other comments about Microsoft's current market cap and new CEO, this article was posted also posted before Windows 7 was released; arguably the best version of Windows ever made. In fact, that era of software including Office and their server products was very solid.
You're comparing an OS that has one major release (windows 7) to one that has been essentially a rolling-ish release for ~10 years. (Mac)OSX (which you linked to) has been going/supported far longer than windows 7.
Yes but I'm also looking at the individual years. OS X has many releases and Windows has many releases; they're both effectively one product. Windows bugs are counted twice if they exist in different versions but are still effectively the same bug.
This is not entirely true. Yes, Ballmer made many mistakes, but he also architected one of the most sustainably profitable enterprise software businesses (on the heels of all MS software products we love to hate). These enterprise customers are more than knees deep with Microsoft, and deep account penetration bought at least 5 years for Satya to steer the ship in the right direction (giving up on B2C mobile, Windows obsession, etc.)
One of the most common biases I see in how we assess leaders is how we try to attribute success & failure to a single person at a discrete point in time (ex: Here in the US, the extreme Left claims Trump is taking credit for Obama's foundational effort for the economy while the extreme Right says it's all Trump). In reality, it is never that simple nor discrete but rather complex and continuous.
I really enjoyed re-reading that. It captures the very essence of "instant in time" incorrectness, it is why I stopped believing anything I thought was "definitely true" in the tech business actually was what I thought it was.
I felt the same way that Paul did, Microsoft had calcified into this Office/Enterprise crust, living off their contracts with big businesses, while the real world was cruising to a new, agile, and open source beat. I had also been at Sun when it was still small, and cringed when bad news from IBM put a big negative impact on Sun's nascent stock price, even though, at the time, IBM's pain was coming from Sun not that we'd share in it.
But the trap that I had fallen into, and this essay expresses, is that "Today is every day" or more precisely, now that the world is to my liking, it will cease changing in ways that I disapprove. :-) I felt like the puzzle is figured out, I "get it" now, I can see the strings going from the puppets to the puppeteers and now I can see how the world is working.
But the cruel trick is that in systems, the puppets are recursive, or in mathematical terms, the forces influencing the future direction of technology are nonlinear at best, and likely chaotic. Everyone is changing in response to the changes they perceive around them, and in response some of their changes change those around them. Like the traffic analyst who figured out you could stomp your brakes going West on Interstate 10, and have the 'wave' of your braking effect roll south on the Harbor, get picked up going north on the 405, only to have it come back and hit you from the front as its echo headed east back along the 10. Sun made Workstations and Microsoft only cared about PCs except the echo of what a workstation was suddenly came back and hit Sun in the face as a big PC. Or Sun's prophetic, if ill fated slogan that "The Network is the Computer", which turned out to be literally true as clusters of Linux machines took out big iron symmetric multiprocessing systems. And then again, as Paul points out, when the Web browser ran its "programs" on the "network." Where did the computer start and the network end? It was no longer possible to tell.
Take away from this the certain knowledge that Google, Facebook, and Apple will all "die" (Apple for the second time :-)) at some point in the next 5 to 20 years. What we care about today will seem silly in the future, and the there will be technologies that enable completely different things to consume our time and resources than the things that do so today.
People create startups and dream new dreams for the same reason a surfer starts paddling their board toward shore, waves come up, and you have to be in position to get a good ride.
I don't think Google has ever really "died". You could potentially make an argument that when social came along Google was somewhat upstaged, but not really that much.
On the other hand, I don't think it's realistically possible for Yahoo to come back from the dead at this point (and same for the other Oath company, AOL).
Iirc (it's been a while since I read the article) I thought 'dead' was used to mean "they are not the dominant power in the world of software (in terms of tools, languages, leadership, driving change) etc, like they used to be in the nineties" not in the sense of "they are a declining company in terms of revenue or market value"
The present news just says Microsoft's market valuation is higher than apple's (today). Not sure if it really affects the central thesis of the article.
EDIT: literally the first paragraph of the article
"I was talking to a young startup founder about how Google was different from Yahoo. I said that Yahoo had been warped from the start by their fear of Microsoft. That was why they'd positioned themselves as a "media company" instead of a technology company. Then I looked at his face and realized he didn't understand. It was as if I'd told him how much girls liked Barry Manilow in the mid 80s. Barry who?
Microsoft? He didn't say anything, but I could tell he didn't quite believe anyone would be frightened of them."
People fear Facebook and Google and Amazon, if only for the immense amounts of damage they do to societies and polities. Who is afraid of MS (or IBM)? Do startup founders even think of MS as a factor in their plans (which seems to be what the article is about)
Microsoft is to computing as Ford is to automobiles. Both are still there if you want them to be your choice, but by the same token, if you don't need them in your life, they aren't forced upon you.
Me? I haven't owned a Ford since 1985. I switched to Unix/Linux in 1991.
> Nearly all the people we fund at Y Combinator use Apple laptops. It was the same in the audience at startup school. All the computer people use Macs or Linux now. Windows is for grandmas.
> They still think they can write software in house. Maybe they can, by the standards of the desktop world. But that world ended a few years ago.
And now, I bet a vast majority of people funded by YC use VS Code. Maybe even Azure.
I actually miss the time when it was this cut and dry. The MBP was hands down the best laptop on the market and it wasn't even close. OSX was a great mix of Linux and usability. Everything (mostly) just worked.
I don't recall it being like that before or since.
MacBooks are great laptops that last 5+ years. With a windows laptop I never got more than 2 years out of them. Maybe that was because I used Dell mostly but it was a stark difference. The Dell was cheaper though by at least 50% so maybe that would explain it.
This is also a testament to how the web won and now mobile/smart devices are eating the web. Before it was about where the eyeballs where. The web democratized access to these eyeballs.
Okay this comment might finally push me to stop subscribing to IntelliJ and try out VSCode for all sorts of code. Sounds comfortable. Does that thing do any Windows 10 like non-configurable telemetry?
I also switched to Win 10 from Mac (and Linux before that).
My Surface Book has a faulty battery and I am using an MBP when I'm not at the office. I not sure I will stick with Win 10 on my next computer, but I am sure I'm not going back to OSX.
WSL isn't perfect, but it feel more open and more linux-like to me than OSX. Maybe a Pixel Book,maybe straight Linux.
I found myself really missing (1) homebrew (2) a decent terminal (I tried a bunch...) and (3) good trackpad drivers.
Not huge things, but enough to make switching feel unappealing.
(There are a few things similar to homebrew, but none seemed to have gained enough attraction to cover basically anything you could want, the way homebrew does.)
> They can't hire smart people anymore, but they could buy as many as they wanted.
> Buy all the good "Web 2.0" startups. They could get substantially all of them for less than they'd have to pay for Facebook.
> Put them all in a building in Silicon Valley, surrounded by lead shielding to protect them from any contact with Redmond.
Instead, Microsoft did the exact opposite. They picked Satya Nadella who had worked there since 1992. He has driven the innovation right out of Redmond, WA with no lead shielding required.