7
$\begingroup$

I'm talking about this one research project:

but MS closed it in 2008. I've read many articles about it, about high perfomance (30+% faster against tradinional monolithic kernels), much more secure etc. So it should be really a revolution in OS designing. Why they did it? I would be really happy if they didn't... So only one question i have: why? Was "effective managers" who destroyed it or some else? I'm thinging about it for many strenuous months...

$\endgroup$

closed as off-topic by Karolis Juodelė, David Richerby, J.-E. Pin, Subhayan, frafl Nov 10 '13 at 17:27

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "This question does not appear to be about computer science, within the scope defined in the help center." – Karolis Juodelė, J.-E. Pin, Subhayan
If this question can be reworded to fit the rules in the help center, please edit the question.

7
$\begingroup$

First you need to understand the goals of Microsoft Research.

Our labs are unique among corporate research facilities in that they balance an open academic model with an effective process for building research results into product development. This approach, unique among corporate research facilities, pays off for Microsoft as enhancements to virtually every product and service it offers and as entirely new lines of business for the company. At the same time, it produces significant technological and intellectual advances across the spectrum of computer science.

from: Inventing the Future of Computing

Second you need to see how these ideas are translated into current projects.

Windows 7. The latest version of Microsoft’s flagship operating system incorporates a number of “under the hood” technologies from Microsoft Research that contribute to the product’s improved performance, reliability and usability. These include enhancements to handwriting recognition, graphics performance, networking and security. Additionally, the product development team made use of a number of Microsoft Research technologies to write and test code efficiently and more thoroughly identify potential security issues.

See: Bringing Innovations to Life and Making Computers More Capable and Useful

Lastly you need to understand that Singularity has been used as a springboard for other projects:

Microsoft describes Singularity as a research project "which demonstrated that modern programming languages and software engineering techniques could fundamentally improve software architectures." Singularity ended up spawning and/or influencing a number of other operating system research projects at Microsoft, including Barrelfish, Helios, Midori (Wikipedia) and Drawbridge.

from The father of Microsoft's Singularity OS moves on

$\endgroup$
  • $\begingroup$ So i hope it will be used. Because the further they go, the worse it will be, compatibility with legacy code is a very big anchor, that clog evolution of OS. Win7 is a little service pack of Vista. Windows 8 - GUI little changes. Windows 8.1 - "Start menu" recovered. 2006-2013 - they did nothing. Because of compatibility with all systems until win95... This is not funny anymore... $\endgroup$ – Alex Zhukovskiy Nov 10 '13 at 8:36
  • 1
    $\begingroup$ Of interest: Under the hood of Microsoft's Windows Subsystem for Linux $\endgroup$ – Guy Coder Apr 23 '16 at 20:47
  • $\begingroup$ Wonderful, ty for a link. $\endgroup$ – Alex Zhukovskiy Apr 25 '16 at 9:33
  • $\begingroup$ My fault, sorry :) $\endgroup$ – Alex Zhukovskiy Apr 25 '16 at 13:03
  • 1
    $\begingroup$ @AlexZhukovskiy I put these links here for everyone, and as a reference for me as I find things of interest related to the topic. With the latest update to Windows 10, I can now run TensorFlow on Windows because of the work from Singularity, Drawbridge and WSL. See: How to install TensorFlow on Windows? $\endgroup$ – Guy Coder Aug 5 '16 at 15:42
6
$\begingroup$

it is probably not so accurate to say the project is "closed". wikipedia says it is "completed" (2007) as basically a "proof of concept" type project. the code is nearly open source and still available for researchers to build on. the site is still available. one "deliverable" of the project can be seen as academic papers of which there are many available.

the goal of the project was re-architected design principles/design patterns (massively refactored/streamlined) for a future operating system, not nec. built by MS. such a system may still be possible. it seems likely that some of the ideas may already be circulating in new OSes eg Android and/or new versions of Apples OS & also some variants of Linux.

not all R&D projects in microsoft or other corporations are slated to be converted directly to products. especially from more pure research, the results may not be "productizable". from GCs article

Microsoft officials often publicly downplayed (at least publicly) the significance of Singularity -- not wanting company watchers to consider a research project a threat to Windows, one of Microsoft's biggest commercial cash cows. But at least one member of the Singularity family, Midori, may still end up a commercial project at some point.

in other words the new design was so different than the old that it could not merely be incorporated as an enhancement. its a very ambitious near-revolutionary rework/ground-up redesign of the entire operating system. one can see eg with Vista that there were massive initial feature ideas (such a as a rework of the filesystem architecture) that turned out to be too ambitious to carry out and was scrapped.

in other words, somewhat counterintuitively Windows may now be considered to be a legacy system running on a PC. in earlier history there were no legacy systems running on PCs because they were so new, it was more a mainframe concept. but after nearly 2 decades of Windows on the PC, its in unequivocally in many ways approaching legacy system symptoms/status.

it is a very large amount of code, increasingly unwieldy to maintain/change, that cannot be changed in major ways without causing major breakage. another relatively new concept to describe this is technical debt.

a classic book describing the powerful/overwhelming inertia effect of attempting change in large codebases is Brooks Mythical Man Month.

other aspects to consider are Microsofts competitive position in the industry which has faded somewhat in recent years, making it less able to pull off/carry out very bold and game-changing new designs and making it necessary to prioritize (and yes cut) some projects. Microsoft routinely goes though massive internal reorganizations which it is involved in currently with the departure of the longtime ceo Ballmer.

a reality of applied CS principles in industry: complex internal politics not entirely grounded in technical merit/finesse or gamechanging potential plays a role in what projects live and die in Microsoft or [quite similarly] any large corporation. it does take on Darwinian aspects at times esp in the fast-moving/evolving IT industry where one leader, Grove of Intel, also espousing/emphasizing its Darwinian aspects, famously said "only the paranoid survive".

$\endgroup$
  • 1
    $\begingroup$ example of a singularity design pattern making its way into other software. singularity considers "software isolated processes", SIPs. google chrome isolates tabs into separate processes. $\endgroup$ – vzn Nov 11 '13 at 16:20

Not the answer you're looking for? Browse other questions tagged or ask your own question.