Slashdot Mirror
Ask Microsoft's Martin Taylor About Linux vs. Windows
Martin Taylor is Microsoft's global general manager of platform strategy, but he's best-known as the man the company trots out to refute claims of Linux superiority. Here are links to several interviews he's done in the past two years: vnunet.com; CMP; Computerworld; and one on Microsoft's own site. As usual, please submit one question per post. We'll present 10 - 12 of the highest-moderated questions to Mr. Taylor about 24 hours after this post appears, and we expect to publish his answers within the next week.
I see Microsoft ads in magazines claiming that the TCO a business using Windows is significantly less than using Linux. How can this be?
Ask Satan About Good vs. Evil. The top 10 moderated questions will be answered by Lucifer himself.
Give me a break. Do you guys REALLY believe this guy is going to say anything substantive or refrain from spinning the company message? Why do you even bother trying to legitimize this shill?
plz/thnx
Oh, a real question while I'm at it:
Of the Linux distributions you've extensively tested, assuming that you have so that your arguments are based on information rather than conjecture, which do you feel is the most desktop-ready?
perl -e 'print $i=pack(c5, (41*2), sqrt(7056), (unpack(c,H)-2), oct(115), 10)'
If office were priced upon a subscription model (as has been speculated) would it be appropriate to sell and would you try to sell a Linux (any toolkit/window manager of your choice) version in the same way that Office exists for Mac?
:-P)
Does the diversified nature of windowing toolkits for the X Windows system, the current use of both XFree86 and X.org, and the huge effort you have put into you own kernel for MS Windows make it even less likely that a version of any MS software will appear for Linux, no matter how many office desks begin to use Linux? (Is the technical challenge too big for you?
Wow, was that a loaded question or what. The two potential answers that you gave him are both bad. You're right, there is really no excuse, IE should have supported or renounced these things a long time ago.
One of the security principals that was recommended to me at TechEd last year was that if a component wasn't needed on a system, remove it.
The theory behind this was that if it's not installed, it can't pose a threat. And if you don't need it, why have it installed in the first place?
I would like to ask, then, why are Internet Explorer and Outlook Express not removable from Windows Server 2003? On my Active Directory servers I have no need for a fully graphical email client.
Equally, I have no need for Internet Explorer on the servers. I do not use Windows Update on the machines; I instead test the patches on non-production machines before burning them to CD and deploying them manually. I have no need to view HTML help files on the server.
Since we have seen 11 cumulative patches for Internet Explorer in the last two years, this is a concern for me. I'd rather completely remove these applications instead of 'disabling' them.
I can choose to remove these types of applications from my Linux installs, why can I not remove these unneeded applications from my Windows installs?
Most admin tasks on windows can be done via Telnet. Hell, as you well know you can run BASH and SSH if you feel like it. A "good" admin can deal with as many Unix systems as Windows systems.
"The "hidden" costs of lost time due to (A) protecting against adware/spyware/malware/viruses/pop-ups, or (B) actually disinfecting machines that got infected anyhow."
We're talking about servers here. In a well designed domain no one has the rights to the server systems required to infect them with anything.
"The "hidden" costs of downtime due to buggy MS software. Sure, F/OSS stuff has bugs too, but when it does, at least the admin can try to fix them. When MS software is buggy, the admin is 100% at MS's mercy to fix the bug (since, being closed source, MS software is often 100% unfixable to anyone outside MS...)"
Right. Because all Linux admins are realy high level coders that can debug kernel conflicts in their spare time. When software wont work, 99.999% of the time you just wait till the writter fixes it.
"The "hidden" costs of dealing with "hacked" IIS servers (vs. Apache)."
Check the statistics. Apache gets compromised a lot. Whats more there is NOTHING saying you must use IIS if you run Windows. Whats more, all you need to do is delete the cursed front page crap to fix 90% of whats wrong with IIS.
"Have you ever thought about just turning off the TV, sitting down with your kids, and hitting them?"
All these serial number checks, dial-home schemes, registration schemes, digital "rights" management schemes, crippled 'starter' versions of windows, and now all sorts of anti-piracy checks whenever someone wants to patch ther Windows box - Microsoft does spend an awful lot of time and effort deliberately making sure their software doesn't work unless the customer jumps through the appropriate hoops.
Aren't you worried that this continual (and increasingly intrusive) process of deliberately breaking and/or crippling your own software is going to alienate some your customers and make them feel like criminals, particularly since the makers of the 'free software' operating systems that you're now competing against have no need of any of it and can concentrate all of their resources on trying to make their software work?
One of the myths about Windows is that there is a company behind it you can hold responsible for flaws that impact an organization. If you read the EULA of any MS product, even an update, it disclaims any responsibility whatever. They specifically avow that they are not fit for any purpose.
So what's up with that?
Open source licenses usually have the same thing, but those are generally free products. You guys have taken in a couple hundred billion. Plus, we can use the code as we like. So you can't claim any kind of equivalence.
My vote: "against". A yes/no question giving no real insight and trying to trick the opponent into saying something that will sound silly. The problem is the difference of platform quality will show when the administrators are both equally skilled on their respective platforms. And schools of administration of Linux and Windos are so different, that it's impossible to compare skills of the two, it's impossible to tell whether they are "equally skilled" - the factors you CAN measure are compound ones - i.e. how smoothly the systems run (downtimes, reboots, intrusions etc) and if you take two system-administrator sets that run equally well, it's still impossible to tell if that's because of the admins or because of the software.
So, answer 1) "I agree admin skill is essential" - "so why do you claim Windows is better than Windows then?"
Answer 2) "I disagree, software is the ultimate factor" - "so you say Windows administered by a crappy admin will run better than Linux with a good admin? What a piece of crap!"
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
Why does Microsoft regularly seem to decide to break with open standards and impliment their own version of them?
Silly rabbit
questions should be genuine. it shouldn't be used to prove a point, or lack of a point, or to push an agenda or to reinforce what we already know. nor should questions be used to try to push the answerer into a corner to "admit" something. we know it won't happen, it's self-serving and frankly, useless.
I can answer that one.
Apache and Php were largely written and optimized for (GNU/Li|U)nix. (I can't say, but I expect the same for MySql but would have to see benchmarks.) This makes sense, as there are still more LAMP than WAMP users.
Were they written with Windows in mind and ported to Linux, you would likely see the same result.
questions should be genuine. it shouldn't be used to prove a point, or lack of a point, or to push an agenda or to reinforce what we already know. nor should questions be used to try to push the answerer into a corner to "admit" something. we know it won't happen, it's self-serving and frankly, useless.
For the first time, I wish I had mod points. I've never used any before. I would rate this insightful.
To the person who posed the original question, I ask, "A colleague of mine told me that you're mean-spirited and vindictive. I think you're just not very bright. Which one of us is right?"
There are plenty of PHB's who are militant about MS products. They don't post on message boards and the like though. They stick to company memos and purchasing decisions.
People make a really big deal out of free. But if one product is better than the other -and I won't argue whether win. or linux is better- than why not pay for it. People pay more for a Porche than a Taurus because a Porche is better. If there was ever a car which was entirely free, people would still pay for the Porche.
The argument shouldn't be: Linux is better because it is free. It should be: Microsoft's higher price is not enough to justify the additional features one gets from it. If someone paid you a hundred dollars to use an operating system and it was really bad, most people probably wouldn't use it. The price, whether it be $-100, $0, $100, or $1000 is meaningless. However weighing that price to the given feature set is what is important.
Microsoft is completly capable of competing against Linux in the long term and writing them off as some ageing dinosaur is not accurate.
No I dont work for MS, Yes I'm running Linux on my laptop, m6811 fedora core 3... hot
Fast Federal Court and I.T.C. updates
Slightly less loaded question...
Many people have suggested that the TCO studies referenced by Microsoft are inherently biased towards Windows - the infamous example being "Windows on Xeon vs. Linux on a huge IBM mainframe". Do you believe that the studies in the "Get the facts" campaign are impartial?
You ARE aware that linux zealots are probably the noisiest faction in the IT world, right? You're also aware that a vast, vast majority of that noise is focused as an anti-Microsoft beam, right?
So, as an obvious linux zealot yourself, why don't you answer the question? Why doesn't linux just speak for itself instead of relying upon a few thousand nasal whines?
-
Inventor of the term 'pardon my French'.
Microsoft's "Get the facts" campaing has been highly publicised in a wide variety of places, and frequently cites figures from studies that seem to show Windows at an advantage over Linux, yet on careful examination of these studies there are often methodological flaws in them.
I recall reading the details of one from the downloadable report on the Microsoft web site recently that compared the throughput of Windows + IIS to Linux + Apache for serving static web pages. The figures showed Windows in a clear lead, yet on closer examination it appears that the Windows installation had been thoroughly optimised (by, e.g., turning off the collection of last access information on the file system and increasing the default filesystem block size, see pages 30 & 33 of the document linked) whereas similar optimisations had not been applied to the Linux system for the test (with default configurations suggested by the distribution installer accepted for filesystem parameters, see pages 30 - 32 of the document).
How would you answer those who are concerned that by presenting these "independent" tests where the testers have followed precise instructions from Microsoft on how to optimise their products but have not (apparently) consulted Linux experts on how to optimise Linux systems as authoritative that you are unfairly distorting the truth and painting a poor picture of Linux? Is it just that you're doing your job the only way you can, because on a level playing field Linux would win? Or is the picture of these reports as unfair to Linux in some way wrong?
It seems apparent to me that Windows Server 2003 and SQL Server are propped up by third party ISVs. For example, I've done work with small credit unions which use a software system that will only interact with SQL Server, requiring of course, a Windows Server OS as well.
In this situation the software costs for a new server far exceed the hardware costs. Most of these applications have no need for anything beyond a basic SQL db server, yet were written before MySQL, Postrgres, and other OSS db's became viable alternatives to SQL Server. ISVs typically refuse to incur the costs of re-writing their software to no longer use specific SQL Server hooks. This gives Microsoft a huge inertia advantage over other platforms as far as small businesses go. I've spoken with many small business owners and the ISVs and generally speaking they agree that using a cheaper GNU/Linux solution would be preferable.
Given that such flagship products as Windows Server 2003 Small Business and SQL Server 2000 compete with Linux not on it's own merits, but on what ISVs allow you to choose, how do you expect to compete with OSS once the ISVs begin to change their strategy?
Slackware, what else when it must be secure, stable, and easy?
Have you ever taken part in an OSS project? If yes, which one? If no, how can you truely understand the merits of this development model?
It *DOES* have a militant following.
Work in a large server farm which offers both Windows and UNIX hosting. There will be Windows Guys and UNIX Guys, and they will constantly be at each others throats over things which are easy in THEIR environment, but a pain in the ass in the other's.
I moved off windows in 1997 when a virus ate my master's thesis. AV vendor was no help. MS was no help. Basically nobody could help me. I had to retype 130+ pages from old printouts. I have yet to have a single problem on Linux, since I moved in 1997. No issuues with malware of any kind. Its been close to 7 years now. All my windows using friends have constant problems with malware - even several of the MCSE sys-admins have such problems. Why should I ever come back to windows?
Rob Enderle's excellent new book: Everything I needed to know about Computer Science I learned in Marketing School
Currently most people use 3rd party apps to keep their machine virus-free, yet recent moves by Microsoft have led many to believe it is considering a non-free MS branded anti-virus application. Don't you think that if there are holes in the OS that allow viruses to get on the machine, the company should be responsible to fix them and remove any infections for free, as your product is therefore defective?
That's a compatibility thing from the mid-90's. Lynx might get one type of page, "mozilla" browsers would get another, etc. "Mozilla" in the header string has nothing at all to do with Mozilla.org. It means "this browser is mozilla compatible"... in fact, that's what the word "compatible" in there is for as well.
How have you approached investigating/researching Linux and other FOSS and their pros/cons?
Did you just install a RedHat box and play with it for a few hours? Have you installed multiple different distros (and/or applications) and worked with them each for (at least) a few days? Somewhere in the middle? None of the above?
Thanks for taking the time to answer our questions. Please encourage others to do so as well.
If Microsoft products are truly superior, then why is it that high volume websites such as the new Microsoft search engine, for example, are running on Linux and/or Apache. See netcraft results.
We've been hearing a lot about the new command-line features that Longhorn
is planned to have. What other geek-appeal features is Microsoft working on,
to help Windows compete with Linux on the tech-savvy user's desktop? Will
Longhorn finally ship with a better text editor than Notepad? DOS used to
come with BASIC and later QBasic; are there any plans to include a flexible
and powerful general-purpose scripting language, such as ActivePerl? What
else does Microsoft have up its sleeve to appeal to people who might
otherwise seek alternatives?
Cut that out, or I will ship you to Norilsk in a box.
Actually, the question is misleading. The .doc
file format is documented on MSDN[1], and is just
as open as PDF. The two also serve different needs.
PDF is effectively a page description language,
albeit one with some nice interactivity features
like forms and even animations (although few
people use them). The .doc format is intented
for editable documents, and stores various metadata along with the content. PDF is not and
doesn't.
But it does lead nicely to another file format related question. Last week, Bill Gates claimed:
Common file formats are the contract by which office applications can exchange data with each other. Given Bill's commitment to interoperability, when can we expect the Visio file format to be documented so that other diagram editors such as Dia of Kivio can interoperate with Visio, as Bill desires?
Similarly, the Exchange wire protocol is the contract by which mail clients communicate and exchange data with the MS Exchange mail server. I take it that we can look forward to documentation for that, too, so that the myriad email clients in use today can talk to an Exchange server?
Another example would be the W3C standards, the contract by which a web developer sends markup information to an end user for viewing in a browser. The rest of the world is happily using CSS to provide rich presentation of information to end users. Yet as developers, we are forced to break that contract because Microsoft's IE browser doesn't honour the contract, and our web sites don't display in the intended manner. Will MS commit to bringing IE up to scratch so that it interoperates with the rest of the world?
Will MS start making versions of Word that use standard UTF-8 character encoding, rather than a Microsoft specific one that produces output that doesn't interoperate with non-Microsoft platforms (and even, as we found out this week, with newer versions of IE, which correctly ignore the MS character set!)
Or was he merely referring to making Microsoft applications interoperable with each other, a move which reduces customer choice, and prevents them from picking the best solution available for the task because it may not interoperate correctly with existing Microsoft products?
[1] At least, it was. I don't know if that documentation has been kept up to date with the latest versions of .doc
"The invisible and the non-existent look very much alike." -- Delos B. McKown
.. but I have to say - neither. I don't like both.
And this is a reasonable answer regardless of
whether one's affiliated with Microsoft or not.
3.243F6A8885A308D313
For experienced computer users, the clear separation between the different parts of the operating system, the hardware drivers, and the applications makes a system running Linux feel cleaner and more organized. The inherant transparency in the system makes it easier to secure a Linux machine and keep it secure, but in addition, it is reassuring to the user, who can feel a greater degree of control over their computer.
In Windows, there are user-level applications and drivers intertwined with the base operating system. In what ways do you think this is beneficial or detrimental to the user and the system as a whole? Regarding software design at Microsoft, what is the importance of creating or maintaining system transparency and modularity relative to other things like useability, speed, interface homogeneity, and the like?
Many would argue that the latest incarnations of Windows are not as modular or transparent as a GNU/Linux system. Is Microsoft planning to place more of an emphasis on transparency and modularity in a future version of Windows? If so, in what ways?
Light is filtering down from above. Would you like to use DIVE?
In your October 2004 interview with VNUnet.com you deny that Microsoft must compete with Linux in your operating systems, going so far as to say that 'nothing could be further from the truth.' With this in mind, why does Microsoft not aid in the development of API's designed to enable Linux-based operating systems to run Microsoft applications? If the Linux community and Microsoft are not competitors, as you claim, it seems to follow that collaboration on such projects as Wine (Win32) and Cedega (DirectX) would do nothing but benefit the community overall, and do much for future interoperability.
We are a medium sized school with approximately 250 windows 2000 or XP workstations (OEM licensing + office XP) and 100 or so staff and student laptops.
We're looking at migrating off our aging NT servers to new backend logon and file servers. We already have several linux 'edge' or special purpose servers; firewalls, backup, web, email, pxe+dhcp, dns etc, and we have a decent amount of in-house experience in both windows and linux.
Given our desktops must remain on windows because of office and windows-only education software we priced up both windows server 2003 and linux replacements. Sticking to the same hardware for both costings, we came to some worrying conclusions.
Redhat Enterprise ES would set us back £700 a server, with free client access and 3 years of upgrades, and we've also got the option of a completely free system like debian. We'd use samba+ldap to largely replicate our existing setup, but with beefier hardware and security updates.
Windows 2003, at £30 a seat for new Client Access Licences, would set us back nearly £22,000 for current and next year projected licence requirements, just for authentication and file sharing, with extra costs in the thousands for every extra server we might add later.
Given that implementation will be done in-house regardless of our chosen solution, I'd like to know if:
a) I'm missing something obvious with regards the licensing costs for windows server 2003?
b) If I'm not, whether you expect such a large mismatch on up-front costs to seriously impact on Microsoft's server business in the education and other cash-strapped areas?
Even assuming the windows implementation is more efficient and quicker to deploy in itself, the linux system would be far simpler to integrate our existing single-purpose servers with (direct access to the ldap user database, for a start). With the CAL licencing savings alone, we could buy an extra server and 20 workstations.
Remember kids, it's all fun and games until someone commits wholesale galactic genocide.
Hello. Since you are being asked about Linux vs Windows, should we take for granted that you are partially a linux user, or at least you have tried Linux enough to reliably back up your arguments? If yes,what distributions have you tried? Do you have a personal preference?
Has Microsoft management ever considered creating an open source Windows which could take advantage of this resource? Restated, has Microsoft considered cultivating a "community" of volunteer developers with access to Windows source code.
It's not an "open question". They put "Mozilla Compatible" in their identification string so that pages that scanned this for "Mozilla" (because they were written for Netscape) would find it and give them the content, rather than returning a page that said it only worked with Netscape.
Before you throw wild accusations around, you ought to learn at least a little bit about the history of what you're talking about.
Rarely have I seen such a high profile interview that gave any kind of meaningful answers or insights as to what the interviewee is actually doing/thinking/feeling. It's just a game of lipservice and wordsmithing: just like politicians, when asked a difficult question, the answers are always vague and nonspecific. God forbid any recognizable figure actually speak their mind!
As interesting as this interview could potentially be, I'm afraid that the OSS/Linux community really won't gain anything from it. But Microsoft will benefit: they'll get to go around talking about how they are "cooperating" with the Open Source community.
I have always thought it would be interesting to have live, face-to-face debates between Microsoft (represented by Gates, Ballmer, senior management and engineers, etc) and some prominent OSS figures (who to send would be a debate in and of itself, but you get the point). The debates should be broken up by topic, i.e. economics, security, innovation, marketing, etc.
These PR types are pretty much all the same. Their job is just to make their company/product look good---they don't even need to know anything about it. Take the top PR guys from ten random major corporations, and shuffle them around to different companies. I bet they only need one day to resume their pace. Why? Their job is totally formulaic: when asked a difficult question, (1) restate the question, but downplay its significance, (2) cite an example of where you did something good -or- talk about some initiative to stop a (minor, unrelated) problem, (3) make really general statements that are true, but don't really have any meaning. Repeat steps (1) through (3) until the issue has been clouded enough to talk about how great your company/product is, and how your high-level goals/initiatives support this right behavior.
Back to my original point: given that I think we ought to be able to predict Mr. Taylor's answers, here's my crystal ball:
Slashdot: If Microsoft does not see Linux as a threat, why has such an investment made to discredit it?
Taylor: Laughing. Microsoft has a commitment to its customers and stockholders. In that regard, it is our obligation to constantly assess market conditions and provide the best value to our costomers. We listen to our customers. And a lot of customers have been asking about Linux! We cannot affect customer curiosity, and we certainly encourage competition. Therefore, all we can do is educate our customers, to make sure they make the best decisions. Numerous case studies have shown that Windows has a lower total cost of ownership than Linux; in many cases it is also more secure and better performing. We simply want to dispel the myth that Linux is a legitimate alternative.
Slashdot: Microsoft has recently made statements suggesting that it is an interoperability champion; even moreso than Linux. How can you achieve such a great deal of interoperability with closed, proprietary standards, and threats of patent lawsuits?
Taylor: Interoperability is an increasingly important aspect of technology. Microsoft backs its products with a greater degree of customer support than any other software company, guaranteeing the greatest deal of interoperability. We believe strongly in intellectual property protection because we've made huge investments in the innovation we've brought to the market. The open source community does not respect intellectual property rights, and as such, our products must be protected accordingly. However, we have the most competitive and exhaustive collection of licensing options for third party developers who wish to use our technology in their own product. The same cannot be said for open source, which offers neither support nor idemnification.
Slashdot: With the proliferatio
By garnering the attention of the Linux community and focusing that voice, stripping it of its many facets, into a 1 on 1 interview with an MS PR person, can only mean that MS will be able to characterize and make "official" said voice from the OSS community.
This is merely an opportunity to allow MS to respond and refute in public the "true voice of the OSS crowd" or whatever they may paint any of your responses as.
First of all, you're expecting answers from the wrong guy. Second of all, you're expecting answers and not PR.
This guy's a spin artist, and you're giving him legitimacy, validation, and a voice.
Instead of fighting Linux, Java, etc., Microsoft should embrace them. Instead of trying to dictate what the customer uses, foloow their lead.
Face it - Microsoft will never kill Linux. But, there's potentially a lot of money to be made supporting it. Imagine a Microsoft bundle of Linux - with Microsoft support. You'd get a lot of business you'll never see otherwise.
Then, consider the concept of cross-platform software. There's really no reason why Word isn't available on more platforms than just Windows and OS X - why not Linux as well? If you write your core software platform-neutral and wrapper the UI nicely, you'd gain market share for very little extra costs.
If Microsoft was more open to this sort of thing, I think people would be as hostile to the company. Right now, every thing you guys do is so tighly bound to Windows you alienate many folks who otherwise might be customers.
I personally don't like Microsoft or Windows - until I got my new box with hyperthreading, it was possible to a single program to lock up the box by maxing out the CPU, and I'm not impressed with the stability or security - but I DO have a Microsoft keyboard - because it's the best I've ever used.
People WILL buy a product if it's simply the best. But if it's shoved on them as a result of market monopoly (and the quality is only moderate), they'll look elsewhere.