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.
Notepad or Wordpad? ;-)
Why are we asking him about anything? He should be asking /us/.
On second thought, I do have some questions, which I can wrap into a single bundle:
"Is Microsoft going to pursue a Palladium philosophy in the next 5 years? And, if this the strategy, what guarantees will Microsoft make that protect Free Speech?"
Ratboy.
Just another "Cubible(sic) Joe" 2 17 3061
For what purposes? What was your personal experience with using Linux?
Why doesn't Microsoft incorporate other OSS software into Windows, such as say Firefox or some of the other software listed at http://www.theopencd.org/?
Does Microsoft feel that Linux has any place at all in the IT industry? If so, where?
When Microsoft seems to tout it's desire to facilitate interoperability, do you mean interoperability seamlessly between your operating system and environment with alternative systems (such as Mac OSX, Linux, Sun Solaris, etc...) or do you mean interoperability between Microsoft products?
What applications do you run to protect your Windows boxes from malware (viruses, trojans, spyware, etc.), and what do you pay for this protection per year? How does this cost compare to the cost incurred by other Windows users? How does this cost compare to what you would pay for equivalent protection offered in Debian GNU/Linux?
How do you respond to people who say that the open-source approach and the associated peer-review process inherently create better code?
Q. Do you frequent Slashdot and the other Linux boards to say what your competition's saying about you?
An Indian-American Hindu committed to non-violent thought/speech/action alarmed by the global explosion of radical Islam
- How Linux admins can easily administrate more machines per person-hour, due to the nature of Unix/Linux's remote administration (and don't even get me started on VNC or Terminal Services; they aren't scriptable, they aren't as bandwidth-effective, etc. etc. etc...), than Windows admins?
- 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.
- 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...)
- The "hidden" costs of dealing with "hacked" IIS servers (vs. Apache).
And a further question: Do Linux geeks really pull in that much more money salary-wise than Windows geeks!? find this claim hard to swallow, especially in today's economy. I call BS. Show some proof.Honey, I shrunk the Cygwin
Can you mention 3 areas where you think windows is better than linux and vice versa.
just because I don't care doesn't mean I'm not listening
To what extent are open source applications on Windows helping it to be more competitive versus Linux? For example, I immediately install OpenOffice.org, Firefox, and Thunderbird over a virgin Windows install.
The world will not get better through technology. We must seek to be better people.
One of the biggest criticisms I have heard of .NET is that it is not portable to non Microsoft/Windows platforms. Microsoft has released Rotor, a CLR/BCL system for FreeBSD systems, that with some work can be made to work with MacOS X. Furthermore, there is the Mono system which provides a CLR for Linux.
My question is does Microsoft have any intentions of implementing a CLR and BCL for any other non Microsoft platforms where applications built under one would be (relatively) easily used under another (provided the application does not rely on P/Invokes of course)? If not... why?
Help Brendan pay off his student loans
Can you mention 3 areas where you think windows could learn from linux and vice versa
just because I don't care doesn't mean I'm not listening
Since Linux is likely here to stay, regardless of current quality, where do you feel Linux will be in 20 years. Especially when compared to where you feel Microsoft's OS will be in 20 years.
Just a boy doing unproffesional IT work that's way above his head.
Will there ever be a free (as in beer) version of Windows, stripped bare of everything but IE and without any network server capabilities? That might compete quite nicely with Lunix.
I assume that you must have evaluated Linux to learn its (supposedly) weak points. While doing that, what did you find out about linux that you think is good? Where is Linux challenging MS the most? (except price, of course)
Linux distros allow you to install on multiple machines. With the advent of home networking, why doesn't Microsoft allow a reasonable number of machines to be installed from a single user license (say 5 machines) rather than forcing a home user to purchase multiple copies (or buy inflated license paks).
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?
Hi Martin
I'm an independant contractor with an MCSE that supports a small customer base of companies that mostly run Windows software. I have four development and testing computers at my house, all of which run Linux and free software solutions, this is because I cannot afford to buy Windows 2003 server, Office 2003, dev studio and a lot of other recent releases.
With my cost free Open Source testing platform I have designed and implemented quite a few solutions with software such as Open Office, Open Exchange, Samba etc.
With online activation and licensing restrictions I am not able to run any Microsoft software in a test environment to ensure it is adequately tested and ensure I am able to support it.
This is driving my skill set and support abilities away from Microsoft and squarely into the arms of the Open Source camp. What (if anything) is Microsoft doing to combat this and ensure that the professionals in the field that sell and support your software have access to the resources they require?
Thanks
John the Kiwi
One of the key points in your Windows versus Linux ads so far is total cost of ownership. You point out to prospective Linux customers that switching from Windows to Linux is often more expensive than upgrading to the next version of Windows.
One way to look at this is to say that Windows is more compatible with Windows than Linux, and therefore a better choice. Another way of looking at it is that Microsoft is exceedingly successful at locking in its customers, and that (as a customer) it is best to get out as soon as possible because it will only get worse.
Do you worry that people will take this second point of view rather than the first, and that the campaign might backfire?
I see Microsoft ads in magazines claiming that the TCO a business using Windows is significantly less than using Linux. How can this be?
These studies typically assume that the status quo is Windows, the workforce is already skilled with Windows but not Linux, Windows is currently installed on the machines, etc. and what is being compared is the cost of sticking with Windows vs. switching operating systems, retraining the workforce, and similar expenses. These transition costs make Linux appear more expensive than Windows, even when the Linux solution itself is cheaper to run.
So, to make this into a question for Mr. Taylor, is this an accurate summary of the studies, and can you point to any that are conducted from a more neutral perspective, without making assumptions of one OS or the other enjoying a comfortable incumbent position?
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.
We've had x86-64 for a while now, but we're hearing that we may have to wait until the first half of next year for 64 bit Windows. It seems as if Microsoft is missing the party here. I can get Linux 64 bit binaries right now. Can this be seen as an example of the open-source Linux out manuvering it's competitors?
"I do a grep for shit, bollocks, and tits before checking in code. I'm professional..." -RECURSIVE_META_JOKE, reddit.com
Sitting here on the Linux side of the fence, and as a part time Open Source developer, I can tell you the things I admire about Windows, both as a platform for development and as a workstation or server. Specifically, the painstakingly preserved backwards compatibility, and the pervasive integration of system are the envy of anyone who has had to use or develop for a wide range of Linux distributions.
What I don't see is the other side - specifically, what does Microsoft see in Linux? What does Linux offer that Windows does not, and what does Linux offer that Windows doesn't do as well, from a Microsoft point of view? Just as important, where is Microsoft headed to close those gaps?
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
should software be sold as a product or a service ?
Is there any particular type of software ( op sysems/enterprise apps/utilities/research software/etc ) that should be open source ? why ?
http://validator.w3.org/check?uri=http%3A%2F%2Fwww.slashdot.org Errors found while checking this document as HTML5!
I understand that Enterprise customers prefer large updates on a long timetable, but consumers tend to want new features now - I don't want to wait three years for a feature that Gnome, KDE or Apple has to show up in Windows. How do you plan on preventing the Windows brand from becoming "stale" when viewed in relation to a community with a much more rapid and dynamic release schedule?
There is another option, which would fit with Microsoft's history. They might just not agree with the transparent png standard and implementation of CSS and be silently protesting their adoption. It wouldn't be the first time they've done this. They've done it with J++, Outlook, etc. Choosing to ignore protocol or industry standards simply because they didn't agree with them and they are an 900 lb gorilla.
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?
Martin,
Many major companies such as IBM and Apple have learned that they can benefit from OSS software by taking an existing OSS product, refining it to fit their needs, and then redistributing the finished product and giving some source back to the community. In this manner, for instance, Apple was able to produce Safari, which is IMO a high-quality and stable web browser, that was produced much faster than it would have taken Apple to write an equivilent product from scratch.
Why has Microsoft not taken similar approaches to software development? I guess that in the past, OSS code has been used in Windows (TCP stack, for starters), but why does Microsoft insist on resisting innovation rather than contributing to it?
"Quoting famous computer scientists out of context is the root of all evil (or at least most of it) in programming." - K
Given that TCO is significantly less for windows than linux[1], are the folks at google morons for using linux?
They use a LOT of computers, and TCO has got to be important in that enviroment.
[1] See MS advertising and "Get the Facts" literature.
What do you see as being "wrong" with linux on the desktop? Are the problems inherent to the licensing regime or is it a technical issue? Is it something like the fragmented environment? (kde v. gnome) Is it poor applications? Is it an underdeveloped user experience? Lack of commercial software development? I have a feeling that your insight on this topic would be very interesting to many in this community.
People who think they know everything really piss off those of us that actually do.
http://www.microsoft.com/windows/sfu/
Why not fork?
I wish I could mod you up. There are a TON of MS only people out there. Heck just post a job and put one Microsoft product in the "like you to know" category and you will get a 100 resume's of MS ONLY people. The job could have 5 hard core non MS requirements and it won't make a difference. Then "if" you interview them just ask them what other technology they like besides Microsoft and why. You will probably get a puzzled look on their face.
I honestly wish I had a dime for all the times I have seen Microsoft people start a holy war when you mention replacing one of their products with a non "Windows" option. I generally ask them the question above "What other products have you looked at or worked with?" I personally don't care who makes a product as long as it makes my life easier and lets me focus on the business, however I will say that I try and avoid companies that lock me in to their technology, so I find myself trying to avoid Microsoft in some ways...
and as I have always said, Once free software is "Good Enough" then the proprietary software vendor is going to be dead. It is almost impossible to compete with free when free is "Good Enough". You can do it but you won't have 50Billion in the bank and have one of the highest profit margins in the world.
Lastly, my question.
At what point in marketshare would Linux need on the client before Microsoft would start porting their applications over to it?
I ask this because it won't be long before the Linux client marketshare will be greater than the Macintosh...
The more I learn about science, the more my faith in God increases.
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
I work in a technical environment (chip design) on
linux, and the only reason I use Windows is to work
with documents from Microsoft Office. Will Microsoft release a version of Office for linux,
and if not, why not? It seems to me a major opportunity for consumer choice was lost when Microsoft was not broken up into independent OS and Applications companies.
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.