Microsoft Continues Anti-OSS Strategy 857
MacDaffy writes "Microsoft's General Manager of Platform Strategy, Michael Taylor, continues Microsoft's press blitz against Open Source in general and Linux in particular in a CNET Interview. He says of Linux: 'You can build it, design it, and it will work great. The trouble begins when you want to add things to it...(due to) the brittle nature of the platform, when you do that, other things break.'"
And Windows never breaks, right? (Score:3, Interesting)
Which is why many places have test machines to test windows updates.
"Linux" is a Total Generality. (Score:5, Interesting)
Microsoft, however, in their positioning, are exploiting the human incapacity for understanding a generality when confronted with logo/brand positions. "Linux" is a huge field. You can't just say "Linux" and mean "All services that depend on a Linux-based solution". Its pathetic.
Microsoft know this; they frame the fight so that when they say "Linux" they mean all Linux-based distributions. But to a user of Linux who actually wants to use Linux, and knows how to use Linux, "there ain't no such thing as a Single Linux target"
I say this having used Linux now for 10 years, quite productively. I haven't used Microsoft-based products in that time. I hardly consider that a "GM for Platform Strategy" at Microsoft will have had that experience
Re:And Windows never breaks, right? (Score:2, Interesting)
Which is why many places have test machines to test windows updates.
So you're suggesting that with OSS it's not necessary to test and you can slap patches and updates onto production servers without trying them out first?
SSDD (Score:1, Interesting)
These tactics do work but they also prove microsoft can't compete on the merits.
Re:Microsoft don't need to spread FUD about OSS (Score:1, Interesting)
lack of cedibility causes fud failure (Score:1, Interesting)
fud can't work if the fudster has zero credibility and ms has exactly that. just about any criticism they could ever level at anyone (valid or not) is something they can rightly be criticized for as well.
Irony (Score:0, Interesting)
Re:*yawn* (Score:5, Interesting)
It's not made of glass... (Score:4, Interesting)
It's using the myriad of custom distributions against it. There are Linux distros for forensics, for security, for graphics, for portability, for a myriad of specialties. These distros are usually booted from CDROM, etc. They have nothing to do with an average workstation distrubution installation of Linux, which has perfectly capable package management using apt-get or rpm. Dependency checking is part and parcel of every decent installation shell. Across a boggling array of packages for every conceiveable app.
Microsoft is just working the edges, trying to make the somewhat busy rate of new distros into a negative. It's true, I just got the LAST Fedora Core in when the next one comes out. But it's hardly orphaned, is it? apt-get works just fine for something I may want to add.
Microsoft's war strategy is to drive major Linux distrubutions to being more static, to stop re-releasing new distro updates at such a frenetic rate. They can't compete in this area, it's too costly for them to do major Service Packs all the time.
Re:OSS IS stupid (Score:2, Interesting)
While someone may want to tinker with code at home enterprises want source to ensure their investments are protected.
OSS may be stupid but that puts it light years ahead of you.
Re:I kind of agree (Score:3, Interesting)
Transitioning Software (Score:5, Interesting)
I agree with Martin Taylor that transitioning software on a Linux platform can be difficult. I also believe transitioning software on ANY platform is difficult. If it wasn't, none of us would have jobs.
I also agree with Martin Taylor that going to a Linux platform may prove more costly than first expected. I also know from experience that Microsoft roll-outs have additional cost.
For Example: MS Exchange server compared to SuSE OpenExchange [novell.com] (now Netline OpenExchange [openexchange.com]). Similar Products. Exchange is cheaper out of the box until you add Spam Control, Virus Control, etc... Also, Exchange counts licenses by CAL connection, OpenExchange is Licensed by concurrent connections - much cheaper. If you want you can even download [open-xchange.org] the Netline Open-Xchange for free with no license restrictions.
Martin Taylor is correct on many points. Unfortunately his logic breaks down because those points are universal and not specific to OSS.
utter bullshit (Score:2, Interesting)
And another thing, PC unix doesn't have such a bad case of 'bit rot' - once you have it configured and running it's the same year after year (other than slowly becomming obsolete - I have 233Mhz notebooks with RH6.2 from years ago that still work fine that I use for a serial terminal or other low speed functions). Windows just gets slower and crappier with time untill you're forced to upgrade or do something.
Buffer overruns... who knew?? (Score:5, Interesting)
When you look at the issue of buffer overruns, eight to 10 years ago in software development, you did not know how much space you might need for something so you just create a big buffer zone to allow things to happen. Who knew that people could go exploit that and use that buffer space to do malicious things?
I'm speechless. I have no words. Except... W... T... F! is he blathering on about?!?
Like sound/audio (Score:3, Interesting)
As much as I love and use linux, jwz is right. Sound and audio are a broken mess. Why can't all desktops/distributions/etc use the same damn audio server interface, like they all use X as a video server interface. It drives me nuts!
Re:What exactly breaks? (Score:2, Interesting)
Saturday I lost the ability to use ttfonts in PHP after I upgraded PHP. Now I'm regretting making all those graphic banners and image links on my company intranet using PHP. But that's because of my ignorance, and it will take me a while to figure out what I need to do to fix that particular issue.
Other than that, that particular server has been heavily used 24/7 for 5 years, only coming down to replace a failing UPS, a failed RAID drive, and to max out the RAM. How is that for reliability, security, and stability?
I would say Mr. Taylor's quote about the brittle nature of Linux certainly applies to older servers like some of mine that cannot be taken down and upgraded (Or shouldn't have, in my case). I am definitely regretting not trying out Debian years ago for that very reason.
I seem to remember WinXP SP2 breaking a lot of things recently, and I can probably come up with a pretty large list of things that are broken in Windows by adding software. Of course, Microsoft would say that they can't control what is broken if you install software that isn't theirs. Apparently in their opinion the same excuse cannot be used by OSS.
Re:Hmmm (Score:3, Interesting)
Like every Windows server I've worked with? Not to mention the expectation with Windows clients that one must wipe and reload the OS annually because of how it falls apart and becomes increasingly unreliable?
(due to) the brittle nature of the platform, when you do that, other things break.'
I've never thought of Hardened Linux (PaX & Grsecurity, or SELinux) or OpenBSD that way. I'd have to believe most other hardened systems administrators do as well. The solution for "hardening" a Windows server is to front it with as much protection as possible, given the understanding you cannot lock it down enough for public IP exposure.
Just looking at Network World will show one there's an entire industry in making appliances to help keep the bad guys out of a fragile Windows server. Realistically, many Apache advocates would probably acknowledge that the strength of hardened Linux and BSd is why Apache is so popular - you can inexpensively deploy your webserver without all the defenses.
Try this taste test:
1. Take a small DMZ segment and insert an up-to-date sniffer in passive mode (with the sniffer having its active IP on a completely separate segment isolated from this segment, and also isolated from any internal LAN as this test will have risk). Get it configured to alert you 24x7 when bad things happen; e.g. email to text messaging script.
2. Take a current production Windows server load, apply all the available service patches and packs to be fair, add Microsoft DNS and run it on a public IP on the DMZ segment, with no third party host, firewall, ACL-enabled filtering router.
3. Take a second server and load OpenBSD or hardened Linux (hardened Gentoo, with PaX/GrSecurity is easy to do and well documented and supported. [gentoo.org])
4. Run until you get compromised.
If you're on a well travelled public subnet, you'll start seeing scans almost immediately. It took me six days to have someone return to the windows box and start attacking it. I killed the project at that point by dropping the subnet altogether. The hardened Linux host was repeatedly scanned by numerous hosts and ignored. Granted, it's not a scientific approach in that the bad guys just were not interested in the hardened host, but the real world value is the knowledge of which system they feel is easy enough to break. The black hats know which OS is the brittle one...
And they call themselves techies (Score:1, Interesting)
Bah.
I've been using Windows back to 3.1 with DOS 6 and straight on through 95, 98, NT4, ME, to XP Home and Pro and throughout that time, I've gotten MAYBE all of ten BSODs.
Windows is, and there is NO arguing this point in my book, leaps and bounds easier to use and configure than any version of Linux. Installshield and its competitors are mature and make installation of all sorts of apps easy. More and more, Windows coders are starting to code properly as Windows changes its architecture to "strongly encourage" it.
The primary reasons for instability on Windows boxes are willy-nilly isntallation of software of unknown origin, inane web usage without regard to security and contractions of viralware, incompetent driver installation by amatures and by techs who should definitely know better (rule of thumb for USB related installs has been since Windows 95B to install the driver software first, THEN connect the device, I see this basic rule violated by A+ and MS certified people constantly), farking with the registry when there is no concrete reason to be in there, and installation of system software such as anti-spyware and anti-virus without paying heed to KNOWN conflicts (such as installing McAfee on top of Norton or vice versa, saw this on Windows 95 ten years ago, STILL seeing this today). On that rgistry point, I still see people installing spyware that is advertised as a "registry cleaner".
I find it amazing that so many people who think they know Windows complain endlessly about it yet in nearly every case I personally look into, USER ERROR was the singular culprit. NOT Microsoft, NOT Bill Gates, NOT closed source, but USER IDIOCY.
Linux and other Unix variants are so difficult to use they rule out MOST user error by way of shutting out the most incompetent people but in no way, shape, or form does it stop it. The help forums for Fedora, Suse, etc. all bear testament to this. Easily, Linux is reloaded in a nuke and pave by the average Linux user maybe five to ten times more often than Windows.
I've had the same Windows XP install running nicely without a nuke and pave for four years now. Any problems that crop up, I fix. I also don't reload my Linux boxes very often either and have become pretty good at scrubbing failed builds and fixing problems. And I do it without complaining, without blaming Linus or anyone else... I take responsibility for my own boxes. So should all these supposedly "experienced Windows users".
Re:Microsoft don't need to spread FUD about OSS (Score:3, Interesting)
This is the stupidest goddamn thing I've ever heard. Look, I'm a Gentoo dev. The simple truth is that drobbins hasn't been involved in Gentoo development for more than a year. There was no "theft" or "hiring away" here. He was already gone.
Article Translation (Score:3, Interesting)
Notice he did not say more investment into the technology than Linux.
But I think now, two to three years into this, we're seeing these issues around cost and reliability coming up such that, we now know we need to go back to the basics on how we evaluate a platform and choose it.
Is he talking about their customers, or Microsoft?
We continue to run our lab where we analyze and look at open-source software to understand and ensure we're still building the right things from a short-term or long-term basis.
Read: If there is some code in Linux that we can use in Windows to make it more competitive, we'll use it.
Apples and oranges: Save money applied more people? Yes, people cost money, but in my experience, there is usually a higher ratio of servers to admins for Windows than Linux. Did these customers use their Windows admins for the Linux boxes?
The systems were running fine until the company had a huge spike in traffic, and there were all kinds of downtime issues. So they did the upgrades, added a few servers, some hardware, some memory and new technologies around the Web site to do more customer relationship database tracking. It was all very complex, and some of the seams of the Linux architecture were beginning to show.
Scratch the word "Linux," since this statement can be applied to any architecture using any OS (Win, HP-UX, Solaris, etc...)
You can build it, design it, and it will work great. The trouble begins when you want to add things to it, add some services and things like that. Because of the brittle nature of the platform, when you do that, other things break. We see that in the labs all the time, and our customers see that as well. So that has a (total) cost of ownership impact on it.
As so many other people have pointed out, this happens with Windows too. I think the big difference here, though, is that an application issue on Linux does not hose the entire OS, whereas on Windows, there is that possibility.
It is also more of a commercial discussion now.
Yes, this whole interview is nothing more than a Microsoft commercial
So we're not an open-source company ... we have projects available today that make Microsoft technology open source.
Huh?
When you license technology as a consumer or business, you should be comfortable that you're protected from patent (or) copyright...claims from anyone. That should be a core fundamental principle of licensing software.
Is he aware that SCO lost the lawsuit?
So if I'm an ISV (independent software vendor), I should be able to take the technology that I've licensed, build something on top of it, and sell it.
Hmmmm... sounds kinda like what Apple's doing with BSD.
So this ability to patent your technology and have some level of protection against it, and in the course be able to build on top of that and innovate on top of that, is exciting.
Software patents are exciting for them, I'm sure. Other than that, I have no clue what he means
From a software perspective, we don't think the patent system is perfect. We had put forward some recommended restructuring to patent laws in the United States
Oh yes, more money changing hands in Washington to benefit the "legal" person.
We have a feature called Configure Your Server Wizard, which allows you to go in and choose a server role so you can take a file server and (reb
Re:Ironic Isn't It (Score:3, Interesting)
On the contrary; the linux gang has a strong anti-Microsoft strategy, and it's one of the most insidious, subversive strategies of all.
They've been providing a cheap, reliable system with no licensing or other legal hassles, which does much of what its users want it to do. It doesn't provide easy entry to viruses, spyware, or other evil stuff, and you aren't tricked into needless upgrades.
Can you imagine the effect on the corporate world if this sort of thing became widespread? It's clear why they'd want to stop it now, before people get the idea that everything might be like this.
It's not the first time such things have happened. Remember back in the 1970's, when the Japanese auto makers started rejecting the traditional "planned obsolescence" scheme that the auto industry depended on for their income. The result was an economic disaster to the rest of the industry, especially American auto companies. Many companies never recovered; others now have profit margins that are a fraction of what they were before this attack (taking into account inflation, of course). American roads are now filled with vehicles that often keep going for 250,000 miles or more before they need to be replaced. In 1960, you had to replace most cars after 50,000 miles. This huge drop in sales was and still is a disaster to the auto industry.
Linux is a similar threat to the computer industry. It runs very nicely on old, obsolete computers, eliminating much of the profit of selling replacement computers every 2 or 3 years. I personally have a linux gateway/firewall running very smoothly on 6-year-old hardware, and see no reason to replace it.
This sort of thing is already starting to impact computer sales in the US. If linux (or *BSD) were to become widespread in industry, it would be a financial disaster comparable to what happened to the American auto industry. It's easy to understand why the folks at Microsoft feel that they must stop it at any cost.
(That's at any cost to the customer of course.)
Re:Hmmm (Score:3, Interesting)
You have services running, and I can guarantee you that at an given time, there will be some service with a potential weakness that can be exploited.
In the single server setup that means that you cannot protect any of your services fully because your server will always be as week as the weakest service running on it.
When distributing functions over servers, the compromise will be limited to the one server running a vulnerable service.
Also, when you have seperate machines for different functions, failure will be far less catastrophic because it means only one service is affected.
It does cause a little bit more work overall, but it significantly reduces the amount of work needed to deal with things going wrong.
I have heard many people following your reasoning, probably because it makes sense at first glance. I have seen very few keeping to this opinion AND being seriously interested in securing their network, after having had a decade or so experience with it.
Re:*yawn* (Score:1, Interesting)
Again with the MS bashing to prove a point about Linux.
Hint: try reading some documentation, taking your time. Also, typing the following three magic lines seems to work for me:
make
make install
And when it doesn't magically (your words) work what do you do. Slack, despite the criticism is probably oldest distro around. It has proven itself to me time and time again and I remember running Slackware
Sorry, it is the topic. See the line above 'pot calling the kettle black', etc.
No it really isn't the topic at all. The topic at hand is how Linux is easy to add third party applications to, not how Windows is broken. Nice try to save face though.
Surely, Sir, you jest? An Anonymous Coward having all these responsibilities? But I'll humor you, here are two reasons:
I don't jest and please stop calling me Shirley. Logging into a
If you don't want to compile anything, why not install Debian, and reap the wondrous benefits of apt-get install? Warning: it's addictive.
I have used apt get. It has failed me as many times as it has worked.
Four words for you: Free Download. Unlimited Licenses.
Two words in reply: Federal Money.
If we don't spend it then it goes to another state.
I rest my case.
Ditto.
Re:Joel on software (Score:3, Interesting)
"Why do so many Linux developers like developing for Linux" is a stupid question - "Why have so many non-hackers suddenly started getting excited about (and defecting to) Linux" is an interesting one.
That said, the interviewer's hardly grilling the MS press flack so it's entirely possible you're right.
However, he still tries to redefine the ideals and approach of "open source" a few lines later, and those were explicitly defined in the question.
Re:This is true... (Score:5, Interesting)
So instead of smearing Linux like they used to, there recommending that IT managers actually use metrics and eveluate the platforms. WTF?
Let me help you: The main reasons are:
Show us the money! This is an easy claim to make...
Bait and switch? "Don't use Linux it's brittle." ... but ... "It's about issues of cost and vendor now."
Has he ever read one of Microsoft's EULAs???! What a dick. That's the choice of the author. Microsoft will choose differently than RedHat. His implying that in europe, software has been hampered because software patents exists, but are unenforceable. It's a shame it doesn't show in the products.Don't give any evidence that Linux actually IS brittle. And it's nonsense. Linux is more agile than any Windows OS.
Bwa Ha Ha Ha Ha (FUD) (Score:5, Interesting)
A little while ago I was called in to teach a Solaris course. I asked the lab admins to install the Solaris Community CD. They were like "Oh, no. We've got a system that works. We don't want to change anything". The fear in their voice was palpable.
I was dumfounded for a second. All I was asking them to do was add a CD's worth of random software. Nothing was even being enabled... then it dawned on me. "Oh. You're used to Windows aren't you? This is Unix. It's actually stable when you add software to it.
Ultimately I had my students add in the software. It was easier. I just mounted the CD image and made it available by NFS. They installed the software and all was well.
The fact that people are so scared of making changes to Windows disgusts me, but I don't think it's going to change. It's part of their FUD campaign. "If WIndows is so bad, what's it going to be like to go to a new system?"
Re:This is true... (Score:4, Interesting)
A few weeks ago we had an interview from Steve Ballmer [slashdot.org] saying that Oracle didn't innovate. Seems that MS needs to coordinate their FUD better.
Re:Joel on software (Score:3, Interesting)
Oversimplification. And yes, designing a general purpose computer and it's OS around the idea that the user will require a bit of training makes for a more usable computer. If you want consumer electronics, buy consumer electronics. Tivo is a good example of Linux made user friendly for the consumer electronics masses. GNOME, KDE and emacs aren't, but are more useful for those willing to invest the effort.
> If our developers and/or users really think with their heads this far
> up their asses, the platform is dead.
Calculus is very useful, but will never be made 'user friendly'. Does this make it 'dead' also? No, the answer is to follow the Unix way and make the hard things possible even if it makes the really easy things a little harder.
Re:Hmmm (Score:3, Interesting)
To me, it shows Microsoft's arrogance that no matter what the facts are, "we're" better and we'll have the best solution attitude is far from what the best solution actually is.
Re:Buffer overruns... who knew?? (Score:2, Interesting)
I think that's very revealing. The sheer number of buffer overflow vulnerabilities in Microsoft software seems to indicate that MIcrosoft has always programmed this way. Looks like this is just official confirmation that only recently has Microsoft even tried to care about security.
Go figure.