This _may_ have been the default, knee-jerk guidance for the Windows team, or part of it, but it was certainly not true for the whole of Microsoft, even back in 2010. In 2010, Microsoft was shipping a handful of cross-platform tools that we built and tested on Unix platforms including Mac and Linux, where it's literally impossible to avoid using GPL software.

Today, Microsoft ships even more cross-platform tools (Visual Studio Code) where again, it's required to use GPL tools, and we contribute to others. (I hack on GPL code as part of my job at Microsoft.)

Now, of course, you're right that this was much harder within Microsoft five years ago. And one's risk exposure is different depending on the product portfolio you own (with something like Windows being less risk averse than something like Microsoft Bob). But some very key lawyers with their heads screwed on straight have overhauled our system and it's much, much easier to deal with OSS than it was five years ago. But even then it was certainly _possible_ and actively happening in parts of the company.

This statement is false as a blanket. I worked at MS in 2010, and I had vim and git installed on my machine. What they didn't want me to do was to install (or look at!) the source code.

The amount of FUD from rank-and-file employees was shocking though. Based on ignorance, but they just weren't part of the GPL world, and didn't care to understand it deeply. So I don't doubt that you may have asked someone and been told "Oh god, don't even look at that stuff". But it wasn't any kind of company wide policy.

I did end up talking to folks that were trying to make such a company wide policy, but even then what they were pushing for was some kind of GPL registry, where if you installed any open source software, you had to register it with the powers that be. I don't think that went anywhere though.

There were some people at Microsoft using GPL software. For example, Microsoft's Virtual Server 2005 R2 supported Red Hat Enterprise Linux, according to this book from 2007 : https://books.google.com/books?id=BAtUWStlEUwC&pg=PA41&lpg=P... .

And http://www.cnet.com/news/microsoft-embraces-gpl-opens-hyper-... from 2009 says "Microsoft embraces GPL, opens Hyper-V to Linux with LinuxIC: Redmond is proving that it can compete with open source on open source's terms by licensing LinuxIC under the GPLv2 license."

That said, the concern is indeed real, as http://www.zdnet.com/article/microsoft-admits-its-gpl-violat... from 2009 shows ("Microsoft admits its GPL violation; will reissue Windows 7 tool under open-source license").

This is such nonsense, using GPL software can't 'infect' software you are writing. I could understand having a rule about not putting GPL source code on your machine in the case you might accidently reuse some of it -- but even that seems unlikely.

I guess it's just easier to say 'GPL bad' than actually explain copyright to every developer.

Um, the concern [for source aspect] was more real than you think: friends and I used to send pieces of GPL code to Microsoft, IBM, and so on that potentially solved kernel problems. The goal was to wait until it got distributed to legacy customers, identify the cut/pasted code, prove it in court, and get these companies' source. Just some devious, hacker stuff we pulled out of boredom for a few weeks rather than a full-on operation with any likelihood of success.

However, I have to imagine someone else tried or will try that same trick. It's why I strongly account for licenses of code added to any proprietary, code base. It's also why license management software is a real market rather than a concept. Better to avoid GPL source than try to explain to FOSS types in court later why it was accidentally distributed and you didn't really mean to exploit their work without following the license. Who would believe that given how most companies are about giving back?

And look at the stunning success Microsoft has had over the past decade! Much of that is a result of stupid policies like this one.

> The assumption is that companies want to extend the software and keep their extensions "locked down." That does happen, but I have seen companies willing to extend software, and make the extension public, as long as they didn't have to make the extension GPL.

...so? The point is that the GPL makes it illegal to lock down, while MIT and the like enable it. Yes, some companies will play nice, but they are doing so voluntarily.

>The argument in the article is wrong. The assumption is that companies want to extend the software and keep their extensions "locked down."

I think that's by far the most common use case for avoiding GPL though.

People being less religious and moralistic about use at any cost would also be nice.