Over the past few days we've published at least four [1, 2, 3, 4] articles about Richard Stallman's (RMS) talk at Microsoft. While waiting for a reply from RMS. It took about half a week. As the saying goes, however, better late than never.
“Microsoft is only one of the actors in this context. I've been following ARM's attempts of acquisition of GPL projects in embedded for many years. For some reason they push to remove copyleft from the projects they want to acquire.”
--AnonymousAs expected, many folks have dubbed that a "mistake" (the talk), possibly due to the way it was presented by Microsoft PR people; they monopolised the narrative. RMS needs to tackle misunderstandings.
In IRC (#techrights) one person pointed out the CopyleftConf from earlier this year, showing sponsorship from Microsoft and the FSF and therefore insinuating, tacitly, there was a compromise if not ‘sellout’. He did not share my view that Stallman wasn't corruptible and was likely misunderstood instead.
“The old method of hiring the project manager to remove copyleft from the project -- see also FreeRTOS becoming Amazon FreeRTOS (this time GPL->MIT).”
--AnonymousThat person't isn't a troll. That person knows a thing or two about Microsoft's tactics. Last night he showed me these two pages [1, 2] from microsoft.com in response to what I had published earlier in the day. To put it very succinctly, these are technical documents about Azure IoT Device Management and Windows 10 IoT Core Services. "I'm following Microsoft strategy entering IoT/Embedded market," he explained, "since it's the area I'm more involved in right now, and they are doing a lot of promotion of these around, I've seen the same people from Microsoft Germany demoing at embedded trade shows around Europe."
I decided to joke: "Microsoft loves data... I mean, Linux..."
Well, "indeed," he responded, "and cloud lock-in. And stomping on standards. So I see the market fit there, where standards for cloud communication and firmware upgrades are thin and just taking off. I saw a nice panel at RIOT-OS summit in Helsinki this week. Video will be out soon. It contains a call for action to not trust third parties for key management in IoT devices. Microsoft is only one of the actors in this context. I've been following ARM's attempts of acquisition of GPL projects in embedded for many years. For some reason they push to remove copyleft from the projects they want to acquire. That's how PolarSSL became mbedTLS at the time and lost its GPL; looks like they re-instated a GPL for mbedTLS recently (probably due to compatibility problems with Apache2 in GPL2 projects)...
"Amazon's strategy has moreover pushed some Free software projects to become borderline proprietary.""The old method of hiring the project manager to remove copyleft from the project -- see also FreeRTOS becoming Amazon FreeRTOS (this time GPL->MIT)."
In recent months we also saw Apple (and to a lesser degree Google) joining the anti-copyleft/GPL 'cause'. This is problematic and it shows that G.A.F.A.M. are no good to Free software. Amazon's strategy has moreover pushed some Free software projects to become borderline proprietary. They recently hired the person who had advocated this (Mac Asay). It's the same person who opened the gates to Microsoft at the OSI over a decade ago -- certainly a 'village fool's move'. He did that after he had sought to be hired by Microsoft. Know your friends....
As for Microsoft? In the next post we'll show that it is still attacking the GPL, but usually not directly (so as to hide its role through the likes of Black Duck).
At the end of last month Microsoft pushed its proprietary and patented file system into Linux -- something that sites in Microsoft's pockets celebrate until now. Just published by one such site:
Among the benefits of having exFAT available as a Linux kernel driver are its availability out-of-the-box, and better performance. According to Greg Kroah-Hartman, who submitted a patch to the Linux kernel adding support for exFAT, a kernel exFAT driver is much faster than its FUSE equivalent module. Specifically, FUSE modules have higher latency and worse IOPs, which is usually not a big issue on desktop-systems that have plenty of computing power and energy available but it is on embedded and IoT systems.