This is a short write up of the event. Alain Williams wrote it & received comments from Phil Hands and Luke Kenneth Casson Leighton.
The MS announcement http://www.microsoft.co.uk/events/Ms EStdEventDisp. asp?params=fHx8fDB8bXNldmVudHN8MHww&EventID=80 61
****
All that you ever wanted to know about Linux but were afraid to ask.
Officially called:
20:20 Seminar Series: Microsoft Windows and Linux An open and honest technology discussion
How is MicroSoft presenting Linux to its customers ? We need to know so that we can be ready for the challenge. MicroSoft is an important competitor, we cannot afford to ignore it, we cannot just dismiss it with a smile of smug superiority
The undersigned were at the above MicroSoft presentation in London, England on 10 June 2004. This is not a literal report, more an attempt to extract the ideas of what was said as sound bites.
There were some 300 delegates, about 90% wearing suits. MicroSoft was well aware that many Linux types were present.
This was a carefully scripted event with someone acting like a TV chat show host. The banter and 'off the cuff' jokes (it is a good idea to keep your contact list of girlfriends protected from your wife) were well rehearsed.
First part: presentations
Philip Dawson - Senior Program Director - Meta Group
. Open Office is incomplete and incompatible. . Have to repackage when the kernel changes. . Difficult to replace MS support with Linux equivalent . Cost of ongoing integration & support . Desktop:
* lacks ecosystem (exchange, active directory, office, 3rd party divers & apps)
* lack of admin tools
* requires ITO to do something
* desktop is about breadth, Linux is narrow . Much of Linux uptake is Unix -> Linux migration, little Windows -> Linux . The costs between Linux and Windows balance out when you buy RedHat/SuSE (Debian is
not suitable for the enterprise because there is no support). . Should focus on services . Problems with the different Open Sources licenses - if you want to base an app on Linux
you need to understand all the different licenses otherwise you will get into trouble. . The Operating System is not comoditised, the battle is on the application stack, this
is where the focus is. . Beware the corporate IP threat:
* You may loose control of your own written applications
* Liabilities from use of open source (eg SCO) (I think he said this) . If you deploy Active Directory do not deploy Samba . Moving shell scripts Unix -> Linux is difficult (ie so why not move to Windows) . There is no hardware saving if you deploy Windows or Linux - ie the same number of boxes needed. . He dismissed, as largely irrelevant, all hardware platforms other than Intel compatible ones. . There were several other cheap jibes that showed ignorance but which would be taken
as true by many who are not familiar with Linux. . Don't look at the TCO (Total Cost of Ownership), look at the ROI (Return On Investment). . MySql is incomplete and does not scale. Don't compare MySql to MS-SQL, but it would be rude to MS Access
to compare MySql to it.
Nick Barley - Directory of Marketing - Microsoft UK
. Boardroom Boredom. Most boards don't really care about IT, they regard it as a cost that
never really delivers on what it promises. . Why MS ? : "We make the complex simple" . MS provides simple packages apps. . "Its free v MS" is a not true headline, you need to look at the TCO. . Linux cost has moved to the same as the MS model - RedHat charges now . Ecosystem buzzword was used again, MS has a lot of partners: integrators, ISVs,... . With MS you get the software all from one place, with Linux if comes from all over,
he quoted Larry Ellison (I think) ''if you saw an airplane with wings made by different
manufacturers would you fly on it ?'' - it is not safe to have components from different
vendors. . Need to get away from being too interested in the technology - need to be less wrapped up with the latest feature
Nick McGraith - Head of Platform Strategy - Microsoft UK
. Went to show file versioning at the OS level & how this integrated into MS Word. . He spent time showing off the features of the latest MS XP service pack.
Coffee.
Second part: presentations from 'users'.
Most of these were MS Windows Eulogies, there was little about Linux in there - which is what I thought we were there to learn about.
Basil Shall, Systems Director - the Grosvenor Group (a real end user)
. Said that he had problems getting MS to take an interest in them, so he wrote to MS head in the UK . get all of s/ware from a single place/supplier . integration issue [ie making different components play] taken care of by MS into the future.
Paul Hartigan, Chief Executive - PharmiWeb Soluitions (an ISV)
. They looked at Linux as a possibility but plumped with MS. . Important items:
* RAD (Rapid Application Development) with Visual Studio - too much learning with Eclipse
* Low training (people already used to MS technologies)
* Stable and low maintenance post implementation
* MS was seen as safe & low risk
* Technical/business support from MS is good
Most of what he talked about was sales pitch for his company.
Anthony Leaper, Chief tech directory - Siebel Systems (CRM provider)
Most of what he talked about was sales pitch for his company 95%, a tiny bit about MS, I don't think that he even mentioned Linux.
. Integration was a buzz word that has been used a lot by him & the previous speakers
and finally registered in my mind. . It was a good thing that they have MS engineers at their head office and some of their
engineers in Redmond - this allows them to get the best out of MS products [[presumably
they need this as they don't have access to the source code]]. . The message was benefits of partnership with MS.
Colin Bradford, Technology Leader - Computacenter
This guy really did talk about the Linux/Windows choice, this was not just a presentation of his own company and deserves credit for that and for what I feel was a well balanced talk.
* Integration technology - this is something that MS does well. * Linux is being used to leverage better deals from MS * has seen much migration of Oracle Databases from HP, Sun to Linux. * Hardware diver support could be better. * Seen a growth of Linux databases - Mysql.
Panel discussion.
I can't remember most of this.
One answer was that MS is using standards and can interoperate with other systems.
Q. I have a lot of legacy email (10 years old) that I would like to read, but can't, how do I ? A. Ask the tech people offline - "rocket scientist"
Q. I'm glad to see that MS is using open standards and can so interoperate with Open Systems,
when will MS freely document it's file formats and protocols so that Open Systems can
interoperate with MS systems ? A. I can't remember that they answered this in a meaningful way.
There was some 'honest' admissions from MS that they had made mistakes in the past but were trying to get it right; eg: pricing; support; bug fixes.
Finally: we all ate an excellent free lunch at Microsoft's expense:-)
Commentary by the authors of this report: *****
The fact that MS had this seminar shows that they are concerned about Linux, they were looking at our response as we must look at what they said.
MicroSoft has some good technology, there are some things that it does better the OSS community does. Perceptions by potential decision makers are important.
Areas where (we believe) that Linux does need to improve: * Admin tools. Yes we can all shell script it, but there should be more 'off the shelf' GUIs * Interoperability. MS is good at this, eg you can copy a picture from the web browser and
paste it into your MS Word document - to do that in Linux you have to save in a file.
This is something that must be made to work. * MS makes it simple to do the simple everyday things. With many Linux tools it is hard to
get started, very often this could be fixed with:
* better default setup;
* better documentation - in some cases *any* documentation. Documentation that does not
assume that the reader understands as much about the problem as the program authors do;
* good & complete examples. * Consistent user interfaces between applications. Better designed UIs where it is easy to find
what you want to do. * Linux has it's own 'ecosystem', there are many projects out there. Many of them are crap,
some are good but not 'enterprise ready', some are world class. It is not always easy
to find and evaluate these programs and we know where to look. This is a marketing
exercise part of which means not talking about those projects that are not good enough.
Even when programs come as part of a distribution it is not always obvious what they do
(e.g. cervisia and GIMP are names that mean nothing if you don't already know).
In many ways this is an ease of use issue. * Marketing. This is as important as all of the technical issues.
Linux markets best to geeks, it needs to be marketed to CEOs.
Linux marketing has mainly been about:
* Cost of purchase. This needs to be expanded to: cost to implementation and cost to maintain
in the long term - as business needs change.
* Risk. We see risk reduction in terms of computer reliability & security; CIOs/CTOs also see
long term support as important RedHat/SuSE understand this.
* ISVs with specialist packages need to sell for Linux. Many ISVs see this as cost of porting
to a new platform with little advantage (just make the customer but an XP box).
Another lesson to be learnt is that Microsoft has so hooked everybody to have no choice, that they can no longer get off "the Titanic". In order to help them get off the titanic, it is necessary for open source to provide "easy transition" points.
That means tackling the underlying infrastructure for them:
Wine, NT Domains, Exchange,.NET (including integrating with NT security).
In technology terms, that means going after Win32 api compatibility, DCE/RPC compatibility, MAPI compatibility, Active Directory compatibility. Without those core technologies there is literally zero chance that any of those companies will even CONSIDER putting money towards open source.
The wrong response to this report is to find holes to pick and so dismiss the MS viewpoint as wrong. * This will allow you to ignore the valid points that MS does make. We need to address these. * Many of those that the above is aimed at will not have the knowledge to dismiss the points, we need
to find ways for them to do so.
At the end of one of the breaks I briefly spoke to a non techie senior management type. I asked if he was going to deploy Linux in his organisation: ``No, it is not secure enough, Windows is much better.'' I did not have the time to explore this with him, I think that we need to understand that decision makers like him believe this sort of thing; our message is not getting through. Unless we can get through to these people MicroSoft will prevail.
If someone does not understand our message; it is not because they are stupid but because we have not explained properly.
Although you make many valid and interesting points, many big companies (can't say, sorry) are already deploying Linux in both the desktop and the back office.
Why? Because the geeks are tired of Windows. And some geeks eventually wear suits.
That is why MS is on tour and is reduced to name calling and using pupets (SCO) to gain as much time as possible.
MS knows that the writing is on the wall (they can't be that stupid), there is nothing they can do to stop people sharing their own stuff (testament to how
Business will be either better or worse.
-- Calvin Coolidge
Report from those who were there (Score:4, Informative)
received comments from Phil Hands and Luke Kenneth Casson Leighton
The MS announcement
http://www.microsoft.co.uk/events/M
****
All that you ever wanted to know about Linux but were afraid to ask.
Officially called:
20:20 Seminar Series: Microsoft Windows and Linux An open and honest technology discussion
How is MicroSoft presenting Linux to its customers ? We need to know so that we can be ready for
the challenge. MicroSoft is an important competitor, we cannot afford to ignore it, we cannot
just dismiss it with a smile of smug superiority
The undersigned were at the above MicroSoft presentation in London, England on 10 June 2004.
This is not a literal report, more an attempt to extract the ideas of what was said as sound bites.
There were some 300 delegates, about 90% wearing suits. MicroSoft was well aware that many
Linux types were present.
This was a carefully scripted event with someone acting like a TV chat show host. The banter and
'off the cuff' jokes (it is a good idea to keep your contact list of girlfriends protected from your wife)
were well rehearsed.
First part: presentations
Philip Dawson - Senior Program Director - Meta Group
. Open Office is incomplete and incompatible.
. Have to repackage when the kernel changes.
. Difficult to replace MS support with Linux equivalent
. Cost of ongoing integration & support
. Desktop:
* lacks ecosystem (exchange, active directory, office, 3rd party divers & apps)
* lack of admin tools
* requires ITO to do something
* desktop is about breadth, Linux is narrow
. Much of Linux uptake is Unix -> Linux migration, little Windows -> Linux
. The costs between Linux and Windows balance out when you buy RedHat/SuSE (Debian is
not suitable for the enterprise because there is no support).
. Should focus on services
. Problems with the different Open Sources licenses - if you want to base an app on Linux
you need to understand all the different licenses otherwise you will get into trouble.
. The Operating System is not comoditised, the battle is on the application stack, this
is where the focus is.
. Beware the corporate IP threat:
* You may loose control of your own written applications
* Liabilities from use of open source (eg SCO) (I think he said this)
. If you deploy Active Directory do not deploy Samba
. Moving shell scripts Unix -> Linux is difficult (ie so why not move to Windows)
. There is no hardware saving if you deploy Windows or Linux - ie the same number of boxes needed.
. He dismissed, as largely irrelevant, all hardware platforms other than Intel compatible ones.
. There were several other cheap jibes that showed ignorance but which would be taken
as true by many who are not familiar with Linux.
. Don't look at the TCO (Total Cost of Ownership), look at the ROI (Return On Investment).
. MySql is incomplete and does not scale. Don't compare MySql to MS-SQL, but it would be rude to MS Access
to compare MySql to it.
Nick Barley - Directory of Marketing - Microsoft UK
. Boardroom Boredom. Most boards don't really care about IT, they regard it as a cost that
never really delivers on what it promises.
. Why MS ? : "We make the complex simple"
. MS provides simple packages apps.
. "Its free v MS" is a not true headline, you need to look at the TCO.
. Linux cost has moved to the same as the MS model - RedHat charges now
. Ecosystem buzzword was used again, MS has a lot of partners: integrators, ISVs,
. With MS you get the software all from one place, with Linux if comes from all over,
he quoted Larry Ellison (I think) ''if you saw an airplane with wings made by different
manufacturers would you fly on it ?'' - it is not safe to have components from different
vendors.
. Need to get away from being too interested in the technology - need to be less wrapped up with the latest feature
Nick McGraith - Head of Platform Strategy - Microsoft UK
. Went to show file versioning at the OS level & how this integrated into MS Word.
. He spent time showing off the features of the latest MS XP service pack.
Coffee.
Second part: presentations from 'users'.
Most of these were MS Windows Eulogies, there was little about Linux in there - which is what
I thought we were there to learn about.
Basil Shall, Systems Director - the Grosvenor Group (a real end user)
. Said that he had problems getting MS to take an interest in them, so he wrote to MS head in the UK
. get all of s/ware from a single place/supplier
. integration issue [ie making different components play] taken care of by MS into the future.
Paul Hartigan, Chief Executive - PharmiWeb Soluitions (an ISV)
. They looked at Linux as a possibility but plumped with MS.
. Important items:
* RAD (Rapid Application Development) with Visual Studio - too much learning with Eclipse
* Low training (people already used to MS technologies)
* Stable and low maintenance post implementation
* MS was seen as safe & low risk
* Technical/business support from MS is good
Most of what he talked about was sales pitch for his company.
Anthony Leaper, Chief tech directory - Siebel Systems (CRM provider)
Most of what he talked about was sales pitch for his company 95%, a tiny bit about MS,
I don't think that he even mentioned Linux.
. Integration was a buzz word that has been used a lot by him & the previous speakers
and finally registered in my mind.
. It was a good thing that they have MS engineers at their head office and some of their
engineers in Redmond - this allows them to get the best out of MS products [[presumably
they need this as they don't have access to the source code]].
. The message was benefits of partnership with MS.
Colin Bradford, Technology Leader - Computacenter
This guy really did talk about the Linux/Windows choice, this was not just a presentation of his
own company and deserves credit for that and for what I feel was a well balanced talk.
* Integration technology - this is something that MS does well.
* Linux is being used to leverage better deals from MS
* has seen much migration of Oracle Databases from HP, Sun to Linux.
* Hardware diver support could be better.
* Seen a growth of Linux databases - Mysql.
Panel discussion.
I can't remember most of this.
One answer was that MS is using standards and can interoperate with other systems.
Q. I have a lot of legacy email (10 years old) that I would like to read, but can't, how do I ?
A. Ask the tech people offline - "rocket scientist"
Q. I'm glad to see that MS is using open standards and can so interoperate with Open Systems,
when will MS freely document it's file formats and protocols so that Open Systems can
interoperate with MS systems ?
A. I can't remember that they answered this in a meaningful way.
There was some 'honest' admissions from MS that they had made mistakes in the past but
were trying to get it right; eg: pricing; support; bug fixes.
Finally: we all ate an excellent free lunch at Microsoft's expense
Commentary by the authors of this report:
*****
The fact that MS had this seminar shows that they are concerned about Linux, they were looking at
our response as we must look at what they said.
MicroSoft has some good technology, there are some things that it does better the OSS community does.
Perceptions by potential decision makers are important.
Areas where (we believe) that Linux does need to improve:
* Admin tools. Yes we can all shell script it, but there should be more 'off the shelf' GUIs
* Interoperability. MS is good at this, eg you can copy a picture from the web browser and
paste it into your MS Word document - to do that in Linux you have to save in a file.
This is something that must be made to work.
* MS makes it simple to do the simple everyday things. With many Linux tools it is hard to
get started, very often this could be fixed with:
* better default setup;
* better documentation - in some cases *any* documentation. Documentation that does not
assume that the reader understands as much about the problem as the program authors do;
* good & complete examples.
* Consistent user interfaces between applications. Better designed UIs where it is easy to find
what you want to do.
* Linux has it's own 'ecosystem', there are many projects out there. Many of them are crap,
some are good but not 'enterprise ready', some are world class. It is not always easy
to find and evaluate these programs and we know where to look. This is a marketing
exercise part of which means not talking about those projects that are not good enough.
Even when programs come as part of a distribution it is not always obvious what they do
(e.g. cervisia and GIMP are names that mean nothing if you don't already know).
In many ways this is an ease of use issue.
* Marketing. This is as important as all of the technical issues.
Linux markets best to geeks, it needs to be marketed to CEOs.
Linux marketing has mainly been about:
* Cost of purchase. This needs to be expanded to: cost to implementation and cost to maintain
in the long term - as business needs change.
* Risk. We see risk reduction in terms of computer reliability & security; CIOs/CTOs also see
long term support as important RedHat/SuSE understand this.
* ISVs with specialist packages need to sell for Linux. Many ISVs see this as cost of porting
to a new platform with little advantage (just make the customer but an XP box).
Another lesson to be learnt is that Microsoft has so hooked everybody
to have no choice, that they can no longer get off "the Titanic".
In order to help them get off the titanic, it is necessary for
open source to provide "easy transition" points.
That means tackling the underlying infrastructure for them:
Wine, NT Domains, Exchange,
In technology terms, that means going after Win32 api compatibility,
DCE/RPC compatibility, MAPI compatibility, Active Directory compatibility.
Without those core technologies there is literally zero chance that any of those
companies will even CONSIDER putting money towards open source.
The wrong response to this report is to find holes to pick and so dismiss the MS viewpoint as wrong.
* This will allow you to ignore the valid points that MS does make. We need to address these.
* Many of those that the above is aimed at will not have the knowledge to dismiss the points, we need
to find ways for them to do so.
At the end of one of the breaks I briefly spoke to a non techie senior management type. I asked
if he was going to deploy Linux in his organisation: ``No, it is not secure enough, Windows is
much better.'' I did not have the time to explore this with him, I think that we need to
understand that decision makers like him believe this sort of thing; our message is not
getting through. Unless we can get through to these people MicroSoft will prevail.
If someone does not understand our message; it is not because they are stupid but because
we have not explained properly.
Big comapnies are ready to use Linux. (Score:2)
Why? Because the geeks are tired of Windows. And some geeks eventually wear suits.
That is why MS is on tour and is reduced to name calling and using pupets (SCO) to gain as much time as possible.
MS knows that the writing is on the wall (they can't be that stupid), there is nothing they can do to stop people sharing their own stuff (testament to how