Microsoft taking a sip of Midori
Yes, Microsoft is pursuing a different type of operating system, which goes by the name of Midori. And, no, it's not the next version of Windows.
The Midori subject has gotten a great deal of attention in recent days, with the fires only fanned by the fact that Microsoft has refused to say anything about Midori beyond confirming that it is an "incubation project" within the company. ZDNet's Mary Jo Foley noted its existence in her Microsoft 2.0 book, while more recently SDTimes posted a bunch of details on Midori based on internal documents.
From there, there has been plenty of speculation about what Midori is and isn't.
Eric Rudder, Microsoft's executive vice president for technical strategy, is heading up the Midori effort.
(Credit: Microsoft)Here's what I've been able to confirm:
Midori is related to Singularity, a research project that dates back to 2003 and is basically a look at how one might architect an operating system from the ground up, given what we know now about computing and where things are headed in terms of parallelism and cloud computing. Longtime Microsoft engineer Eric Rudder is the one leading up the Midori effort.
Whereas Singularity was a research effort firmly confined to a small team of researchers inside Microsoft's in-house labs, Midori is an effort to see if there is something commercially viable that could come out of it, though it could be years off and come in pieces if it comes at all.
The one public mention I found to Midori was within a research paper on a bug-finding program called Chess. On one PowerPoint slide, it mentions a list of "current Chess applications" of which one bullet point is "Singularity/Midori (OS in managed code)."
That syncs with the SDTimes report, which talks about Midori as an OS for the age in which computing resources can be either local or in the Internet cloud and in which processing tasks can be split among multiple processors and multiple machines.
It's worth noting that Microsoft often has incubation projects that seek to explore whether an all-new approach to a product might be justified. That said, up to this point, every update to Office and Windows has been some type of incremental improvement, not a ground-up rewrite.
Back in 2000, the company had an effort called NetDocs that many thought might replace Office with an online productivity suite. Eight years--and at least three Office versions later--people are still wondering when we will see such a product from Redmond.
That suggests to me that the arrival of Midori or some similar approach as a Windows successor is something that is a long way off, if it ever happens.
Microsoft has struggled to change even single subsystems of the Windows operating system, such as the file system. Microsoft has had both Cairo and WinFS projects, ultimately opting instead to stick with trying to build on top of what is already there.
The fact, though, that Microsoft is thinking about new ways it might do an operating system should not come as a surprise. I'd be surprised if they weren't exploring that idea.
What will be more interesting is if Microsoft actually does release something all-new. With its much smaller and mostly consumer base, Apple has shown several times a willingness to sacrifice compatibility in order to take the Mac in a new direction--most notably the shifts from 68000 processors to PowerPC, from OS 9 to OS X and from PowerPC to Intel chips. Microsoft, though, with its huge base of business and consumer users, has long favored compatibility over new capabilities and approaches.
The question is how long this approach can continue. I (and others) have been asking for some time whether Windows hasn't gotten too difficult to update. Take Longhorn/Vista. After a few years of work, Microsoft decided the major architectural changes it was planning were too drastic. It went back to the drawing board, but even the more modest changes it made with Vista have come under attack.
I wonder if, in this day and age, it wouldn't be possible for Microsoft to emulate all of Windows, while moving forward with a more modern software approach, sort of like Apple did with the "classic" mode in OS X. I have no idea if that's anywhere in the cards or not.
But if anyone wants to share some more Midori--I'm up for another round.
During her years at CNET News, Ina Fried has changed beats several times, changed genders once, and covered both of the Pirates of Silicon Valley. These days, most of her attention is focused on Microsoft. E-mail Ina.
However, I have to wonder if you actually read the article of if you simply saw that Microsoft was trying to develop a new OS and had these thoughts. Building on the BSD core (in my opinion) would bring Windows (or whatever it would then become) too close to Mac OS. People use Windows because they like the functionality of being able to use (pretty much) the same thing they've used since the early 90s. If Windows became a BSD based OS... it would be viewed as EVIL MICROSOFT Operating System XX as a direct competitor to AWESOME, GREAT, BEST IN THE WORLD Apple Mac OSX. You can make the absolute BEST product in the world, but if no one buys it or believes you... it is worthless.
The point of the article is that Microsoft might try to build on the semi-sucessful Singularity project to make a consumer version (apparently codenamed Midori). The conclusion that the author makes is that Windows survives and THRIVES on the fact that the people who really USE it don't want it to change that much. They want it to do more things (obviously), but not take away things they're used to. They've maintained 90+% marketshare against the self (and fanboy) proclaimed BETTER Mac OSX and Linux by not really giving it's customers a really strong reason NOT to use it. They're struggling to gain additional market share for the same reason. While they're not giving their customers a reason NOT to use it... they're not really makeing a very strong case for new users to START using it.
These are just opinions... I'm not an expert... I'm not a professional analyst... I'm just a person who can see that what is is what is simply becuase it is... and not get twisted by people trying to tell me the sky is a big curtain and the grass is really just a very very very large shag carpet. Don't pi$$ on my head and tell me it is raining.
FYI... I'm a linux user. I've been using Knoppix, Mandriva, and Ubuntu for about 5 years.
Have you tried http://midori.sourceforge.net/
Yeah, I'm going to switch my whole business to Vista because this guy over here that doesn't even realize he's using Vista when he's using Vista says Vista is cool. Yeah, like Vista or hate Vista Mojave didn't prove anything.
It's not entirely obvious to this ex-MSFT person exactly what drive MSFT to "start over" with yet another OS, unless it were for small-scale devices. That might be a valid reason -- a thin client for things like cell phones, low-cost laptops and even XBox.
On the PC side: The issues with Vista are NOT with the kernel, right? It seems to have withstood any security challenges, etc.. Yes, it's a large OS, but at this point, the PC vendors have adapted in general with 2 gigabyte RAM systems. The issues with Vista were with third-party compatibility and the UI. I think I"d argue too that the issue is with IE -- in a day of web apps, isn't it IE's mission to get software vendors excited, not the OS anymore? It's also not obvious to me that fragmenting Windows right now (where Ballmer is pointing a dearth of applications being written for it) is the way to go.
As to cloud computing, it is an excellent way for the writer of the software to control what you do. If parts of the application run on some server outside your system, they can decide when you upgrade. They can make you rent your software as a service VS purchasing it for a one time fee, and they can turn you off anytime they want (They can even take your data).
"Microsoft will no longer support the managed portion of Windows 2011 after July 31 2015. Users must upgrade to Windows 2014 by that date or Windows will no longer be usable."
> has been some type of incremental improvement, not a ground-up rewrite.
Not true: Windows NT has been completely new code.
The problem there was that instead of the modularity of UNIX Dave Cutler choose the monolithic architecture of VMS... but you can't blame a dinosaur for being a dinosaur.:-)
Hopefully Midory will bring some freshness and will also spur development into the Open Source community to go beyond Linux.
I think part of Apple's current success is their approach to that. It has been reported that their next OSX upgrade will be a nuts-and-bolts upgrade that reduces the size and increases the speed of OSX. It will be the applications running above the OS that benefit from that reduced size and increased speed. If the OS is optimized then it gets a speed boost as well.
HA, Ha: if a human creates a secure system; then another human, by defination, can hack it. Such is true, even from before the early days of code-breaking ; such will be true until the second coming!
-
by mwebb64
August 5, 2008 6:23 PM PDT
-
Reply to this comment
-
(19 Comments)