sankyuu writes "After years of rumor and vacillation over fear of patents, the city of Munich has decided to trickle in its first 100 linux terminals. The floodgates are scheduled to fling open by 2008, when 80% of government PCs should be running Linux."
Even if the costs are same, it's better for Europeans for the money to stay in local economy, than to be flushed away to Redmond. But I'm probably just forgetting all the jobs that will be created by Vista;)
You obviously missed the part where they are paying this money to IBM, so your point was?
As far as I can see, its not only not cost efective, its not even going to be complete. The project lead himself admits they can only migrate around 80%, theres also a quick gloss over the 12 month pilot extension because of unspecified 'problems'.
So slipping deadlines, increasing costs, less than complete and beset by problems. looks like a typical software project to me and not the poster child for migration som
You obviously missed the part where they are paying this money to IBM, so your point was?
The point is that the government and the state institutions are the motor behind the adoptions in the private sector and personal use. By adopting open source solutions, Munich is incentivating the creation and growth of a local market for training, supplying and managing open source solutions. And having in mind that quite a few open source is produced in Germany (KDE, for example) then it is obvious that the people of Germany have a lot to win with that migration.
One other aspect to have in mind is that the money which Munich is paying isn't just for installing new software. Munich is paying the price for not being dependent on a certain platforms (windows) and certain software. It's like a drug addict paying for detox treatment. There are quite a few places that certain software was adopted and subsequently their business was built around it. Now, those solutions will have to be rethought and redone, which costs time and money to accomplish. Nevertheless, it does indeed pay off and pays off well.
On a side note, isn't it funny how the exact same FUD directed towards Ernie Ball's migration to Free/Open Source software is being used against Munich? And once again the FUDers will realize that the migration process, although it isn't always smooth, not only is perfectly possible but also ver positive for the organizations which adopt it.
You can hire some guy to "correct this nasty bug in XXX OSS software". You can't hire some guy to correct this problem in IE or in Windows. It is technically more difficult, and legally questionable.
That is what is meant when someone talks about independance in the OSS community.
> Is part of the project to replicate Microsoft's update mechanism?
Yes.
> Are they planning on maintaining their own source tree > for all applications and the OS?
IIRC, sort-of yes. They developed their own Debian based "distribution".
On a 14k clients-scale, there are no out-of-the-box solutions anyway. Not in MSFT-land, not in Novell-land and certainly not in LNX-land. So, you've got to do your own calculations to see what makes sense for you. Munich made theirs, you'd make yours. Big deal. A complete N
Come on! They will be dependent on a different platform (Linux) and different software (OO, etc).
Are you seriously trying to compare the adoption of a standards-compliant, free and open-sourced platform which guarantees backward compatibility and even compatibility with other standards-compliant platforms with the dependance on a platform which is as closed as it gets, has always compatible problems even with it's predecessor and even patched versions, has a history of very dangerous security problems
They're taking a big one-time hit although. Once they've rewritten/replaced all their software and migrated their data the cost to add new units will be significantly lower.
They're taking a big one-time hit although. Once they've rewritten/replaced all their software and migrated their data the cost to add new units will be significantly lower.
I agree with you and I don't understand why so many people assume that a migration from a Windows infrastructure to an OSS one will cost €0.00? If Munich is going ahead and doing this in the first place they might want to make some fundamental changes to their IT infrastructure since they will be ripping the guts out it anyway. Take for example the proposition of replacing dumb Windows PCs that just stand around all day giving users access to a single application (Why pay a Windows XP license for every one of those PCs?) with Linux based thin clients. In this case they might be factoring the replacement of some quantities of computer equipment and infrastructure changes into that figure of €30 million. Then of course there are the costs of testing the whole system, the costs of writing custom software to aid in the migration of entire data bases, websites and other applications previously hosted on Windows 2003+MSSQL+IIS to open source platforms, porting custom made GUI applications/clients to Linux or replacing them with new webapps. I can see why the costs would go up but in the long run I agree with you that their costs should go down as a result of this measure if they handle the project properly which, admittedly, is asking a lot of a German bureaucracy. I would really like to see a financial breakdown and progress report of this project when they are done, this project is really interesting due to it's scale.
Take for example the proposition of replacing dumb Windows PCs that just stand around all day giving users access to a single application (Why pay a Windows XP license for every one of those PCs?)
Because it's the only thing the application will run on? Unfortunatley it's not always possible to use an alternative.
That's exactly what some FUD-spreading people don't get. What Munich is paying for isn't simply F/OSS. Munich is paying for the total replacement of it's software infrastructure. Munich is paying to get rid of a solution which costs them on software alone more than 500 per platform per upgrade cycle (which, by the way, is dictated by the software company) and replace it with a platform which is uncomparably cheaper. Tha is exactlly the point of Munich's migration and what every german tax payer should be ha
If Munich is going ahead and doing this in the first place they might want to make some fundamental changes to their IT infrastructure since they will be ripping the guts out it anyway.
If restructuring a complete IT workflow system is at all as difficult as platform porting and restructuring a complicated computer program, then you need to resist the temptation of just "fixing things while we're in there anyway".
If things don't work later, then it's too hard to track down the responsible fix.
I wonder how many of the custom apps they're building on Linux will also be open source, and therefore available to the next government looking to switch. It could be that Munich is taking a _really_ big hit, but each organisation which follows the same path will find it progressively easier to switch.
I've often thought that commercial software vendors are taking an immense risk in not porting to Linux, thereby allowing the whole FOSS application stack on the platform to be developed without commercial-grade competition.
This sort of migration could start a cascade effect, where each successful adoption catalyses the next, and there are damn few commercial software houses prepared to take advantage of that.
That's upwards of $3k per computer, how in hell did it get to this ridiculous figure? Being an european myself and in-the-know about the ways of social-democracy I have some nice theories indeed, but being less than IT--related I'll spare the rest of you.
I, for one, think a migration to windows Vista would not be much cheaper; licenses, new hardware, 3 praty and home grown application rewrites, and a lot of tests. however, the license costs for OS and many desktop apps will vanish in the future with the linux solution. also, it looks like they are going for linux terminals, so i guess a lot of maintenance work will be saved on the new clients as well.
You forget that this will leave Munich with suppliers who are able to do mass Linux migrations to government specifications. This unique ability was hard fought for in administrative and legal nightmares over the past years, and is a major part of what the 35 million buy.
Munich will be quick to offer practical migration services to other cities all over Germany. If even a few see the chance to save some money over the M$ option (many German cities, most importantly Berlin, are in big financial trouble), Mu
They fail to mention how many government servers have been running linux behind-the-scenes for years. Changes are, at least a handful. I realize the excitement of this is in bringing Linux to the desktop, but people aren't always aware of its presence, even when quietly surrounded by it. Now I don't have any facts on Munich's server architecture, so I could very well be wrong.
Snarkiness aside, I think this is a cool project. It'll be interesting to see who else follows Munich's lead, and what ol' Ballmer aims to do about it. Maybe he'll chuck a chair (doh.. there's that snarkiness again... time for me to creep back into my hidey hole).
I checked this news out yesterday, and read on the german project website that they already used linux on their servers since 1995. The idea to change came when it was apparent that windows NT would be stopped and new MS software would require them to start a contract forcing them to renew this contract constantly. Then, if I remember correctly, the major tested if his wife could manage to work on openoffice, which turned out pretty nicely.
With the major and of course a majority in the city council backing this, they started a very gradual and careful way to change, with a halt since 2004 because they needed a risk analysis in the case that software patents would be installed EU-wide. The cost risk turned out to be pretty small, as for every patent there can be a workaround eventually, linux is based on code that is already known since the 60s, and some other reasons. In the mean time they made sure they had automated software install systems working, and other practical issues resolved. The big news now is that they will actually start with the first linux machines for office employees. First ones will be for office work that requires interchangeable software (word processor, etc), then more complicated office work will follow.
The article keeps emphasizing how hard it has been so far (to move over 100 computers, and 200 by the end of the year).
After reading all that, this seems like a lie: "Schiessl said it would be impossible to migrate all users to open source, but that 80 percent would move across by between late-2008 and mid-2009."
Most of those computers are probably used for almost exactly the same. The first 200 computers are probably representative for the 80%. So they use the first 100-200 to learn and to develop deployment procedures etc. When that works, they roll it out to all the similar computers. Just like in some smaller places, they use days to test something on one or two computers. When it works they spend an hour putting it on all 500 company desktops, most of the time just waiting for network transfers and rebooting.
And another reasons would be to get the users time to switch over to some of the news apps in a timely fashion (i.e. not just having them "airlifted" into a whole new set of apps and OS). That's why they also have an interrim stage where the users run OpenOffice on Windows. First the apps, then the OS... makes sense doesn't it?
It should be noted that Mayor Christian Ude's PC is slated to be among the first batch of systems to run the Debian-based Linux-desktop Munich will be using.
It should be noted that Mayor Christian Ude's PC is slated to be among the first batch of systems to run the Debian-based Linux-desktop Munich will be using.
Why Debian? Not that I'm implying that Debian is a bad distribution but isn't SuSE HQ practically in their back yard (Nürnberg) ?? Or has Novell uprooted SuSE development and moved the entire outfit to the USA ??
that site has general information about the Linux-Project and a link to this site:
Firstly, thanks for a very German answer:D
Secondly, while that is a nice a site and I say that because this project interests me and I did take a look at that Wiki, I was hoping for a more detailed business and financial oriented explanation than "They will be deploying not SuSE but Debian GNU/Linux, the freest of the Linux distributions." The word 'Debian' is mentioned only once on the pace you linked to.
They don't say, but since the support and maintenance contract was won by Softcon and Gonicus, they obviously don't want to pay Suse for doing nothing. Debian was their choice.
Just one more reason to admire this rethink. You might expect that they would move from the big name vendor to the biggest name Linux vendor they could get, but in the end the name doesn't buy you anything. What matters is that you're supported.
the standard configuration will be Debian GNU/Linux 3.1, KDE 3.5 and OpenOffice 2. however, the main reason for the delays and the slow roll-out are that a lot of custom applications had to be ported and for some existing client/server apps interfaces had to be created from scratch.
They tried. They gave them every chance to come up with a better operating system. They even delayed the switch to Linux by many years to give them a chance. Even now, they're giving them until 2008 to get at least some share of the cake.
But Microsoft just couldn't get Longhorn ready in time.
... some more (French, sorry) 400,000 PCs are to swith to Open Office in 2007 in the oublic sector, folowing a successfull move in the Gendarmerie (rural police, 90,000 PCs).
- A summary here [europa.eu] or in the official French annouce [adele.gouv.fr].
Some Open Source headways in Europe, indeed, can clearly be seen in EU site [europa.eu].
Quite heartening indeed! Maybe the big conservative companies will finaly notice this trend. I am sure Microsoft did.
The cost of switching would apply both in both directions of an OS migration. What Linux has in it's favour is that it's support for Microsoft closed formats (e.g. via Openoffice) is far better than the reverse. Once documents are in open formats it's hard to make a case to back out.
I do wonder whether we'll start to see Microsoft supporting these Open Standards as a way to ease the migration path back - supported of course by heavy subsidies on licensing.
The sentence should read, "Munich Finally Starts Implementing Linux."
The embrace happened a few years ago. It's (Linux) implementation is what has just happened. By the way...does anyone know whether it's KDE or GNOME at the forefront here?
Because KDE is a (mostly)German project, whereas GNOME is distinctly American.
It's perceived as being a German project, but with the number of people who are not Germans involved and on KDE eV it isn't any longer.
Besides, I'm uncomfortable with that as an explanation. I'd like to see a comparison of what they need from a desktop environment, what worked in one and didn't work in the other, what users needs and what admins need, and make a reasoned comparison on that.
But seriously, I am surprised more governments don't move to Linux as a terminal/front end solution anyway. With more services being handled on central computers far away from the actual customer access points, it doesn't really make sense to have full-featured OSes put in place only to hobble them with security software. It is better to create simple remote terminals which can run programs remotely over the network, saving hardware costs and reducing IT headaches.
>I honestly think it has something to do with them having been convinced once already to move away from that type of architecture
I remember us smugly installing x-terminals everywhere to replace the PCs we had. It's the future! It's open! It's economical to run!
Within 3 years, several million pounds worth of x-terminals were junked and we were back with PCs. So much easier to get the software! People know how to use them! Easier to find support staff!
Heh. Misread that summary as "Linux urinals". I was wondering what sort of benefit Linux could bring to the porcelain pots
"Welcome to UrinalXP!
Your ToiletFlush(tm); license has expired. Please supply your Credit card details using morse code on the ToiletFlush(tm); button. The ToiletDoor(pat.pend.) will stay shut for safety reasons until you comply."
80% By 2008? (Score:2)
Amazing.
holy not cost effective, batman! (Score:4, Interesting)
The current projected costs are 35 Million Euros (up from 30 Million) to convert 14,000 computers.
2,500 Euros per computer.
Re: (Score:3, Interesting)
Re:holy not cost effective, batman! (Score:5, Informative)
Parent
Re: (Score:2, Insightful)
As far as I can see, its not only not cost efective, its not even going to be complete. The project lead himself admits they can only migrate around 80%, theres also a quick gloss over the 12 month pilot extension because of unspecified 'problems'.
So slipping deadlines, increasing costs, less than complete and beset by problems. looks like a typical software project to me and not the poster child for migration som
Re: (Score:3, Insightful)
But most/all of the cost is consulting, of which a significant percentage will go to salaries to people locally.
Re:holy not cost effective, batman! (Score:4, Interesting)
The point is that the government and the state institutions are the motor behind the adoptions in the private sector and personal use. By adopting open source solutions, Munich is incentivating the creation and growth of a local market for training, supplying and managing open source solutions. And having in mind that quite a few open source is produced in Germany (KDE, for example) then it is obvious that the people of Germany have a lot to win with that migration.
One other aspect to have in mind is that the money which Munich is paying isn't just for installing new software. Munich is paying the price for not being dependent on a certain platforms (windows) and certain software. It's like a drug addict paying for detox treatment. There are quite a few places that certain software was adopted and subsequently their business was built around it. Now, those solutions will have to be rethought and redone, which costs time and money to accomplish. Nevertheless, it does indeed pay off and pays off well.
On a side note, isn't it funny how the exact same FUD directed towards Ernie Ball's migration to Free/Open Source software is being used against Munich? And once again the FUDers will realize that the migration process, although it isn't always smooth, not only is perfectly possible but also ver positive for the organizations which adopt it.
Parent
Re:holy not cost effective, batman! (Score:4, Insightful)
Parent
Re: (Score:3, Interesting)
Yes.
> Are they planning on maintaining their own source tree
> for all applications and the OS?
IIRC, sort-of yes.
They developed their own Debian based "distribution".
On a 14k clients-scale, there are no out-of-the-box solutions anyway. Not in MSFT-land, not in Novell-land and certainly not in LNX-land.
So, you've got to do your own calculations to see what makes sense for you.
Munich made theirs, you'd make yours. Big deal.
A complete N
Re: (Score:3, Insightful)
Are you seriously trying to compare the adoption of a standards-compliant, free and open-sourced platform which guarantees backward compatibility and even compatibility with other standards-compliant platforms with the dependance on a platform which is as closed as it gets, has always compatible problems even with it's predecessor and even patched versions, has a history of very dangerous security problems
Re:holy not cost effective, batman! (Score:5, Interesting)
Parent
Costs: €0.00 (Score:4, Insightful)
I agree with you and I don't understand why so many people assume that a migration from a Windows infrastructure to an OSS one will cost €0.00? If Munich is going ahead and doing this in the first place they might want to make some fundamental changes to their IT infrastructure since they will be ripping the guts out it anyway. Take for example the proposition of replacing dumb Windows PCs that just stand around all day giving users access to a single application (Why pay a Windows XP license for every one of those PCs?) with Linux based thin clients. In this case they might be factoring the replacement of some quantities of computer equipment and infrastructure changes into that figure of €30 million. Then of course there are the costs of testing the whole system, the costs of writing custom software to aid in the migration of entire data bases, websites and other applications previously hosted on Windows 2003+MSSQL+IIS to open source platforms, porting custom made GUI applications/clients to Linux or replacing them with new webapps. I can see why the costs would go up but in the long run I agree with you that their costs should go down as a result of this measure if they handle the project properly which, admittedly, is asking a lot of a German bureaucracy. I would really like to see a financial breakdown and progress report of this project when they are done, this project is really interesting due to it's scale.
Parent
Re:Costs: €0.00 (Score:3, Insightful)
Because it's the only thing the application will run on? Unfortunatley it's not always possible to use an alternative.
Re:Costs: €0.00 (Score:3, Insightful)
Re:Costs: €0.00 (Score:3, Insightful)
If restructuring a complete IT workflow system is at all as difficult as platform porting and restructuring a complicated computer program, then you need to resist the temptation of just "fixing things while we're in there anyway".
If things don't work later, then it's too hard to track down the responsible fix.
Re:holy not cost effective, batman! (Score:5, Insightful)
I wonder how many of the custom apps they're building on Linux will also be open source, and therefore available to the next government looking to switch. It could be that Munich is taking a _really_ big hit, but each organisation which follows the same path will find it progressively easier to switch.
I've often thought that commercial software vendors are taking an immense risk in not porting to Linux, thereby allowing the whole FOSS application stack on the platform to be developed without commercial-grade competition.
This sort of migration could start a cascade effect, where each successful adoption catalyses the next, and there are damn few commercial software houses prepared to take advantage of that.
Parent
Re: (Score:2, Insightful)
Being an european myself and in-the-know about the ways of social-democracy I have some nice theories indeed, but being less than IT--related I'll spare the rest of you.
Thats one big incentive... (Score:2, Insightful)
Re: (Score:2, Insightful)
Re: (Score:3, Interesting)
Munich will be quick to offer practical migration services to other cities all over Germany. If even a few see the chance to save some money over the M$ option (many German cities, most importantly Berlin, are in big financial trouble), Mu
Behind the scenes... (Score:3, Insightful)
Snarkiness aside, I think this is a cool project. It'll be interesting to see who else follows Munich's lead, and what ol' Ballmer aims to do about it. Maybe he'll chuck a chair (doh.. there's that snarkiness again... time for me to creep back into my hidey hole).
Re:Behind the scenes... (Score:5, Informative)
With the major and of course a majority in the city council backing this, they started a very gradual and careful way to change, with a halt since 2004 because they needed a risk analysis in the case that software patents would be installed EU-wide. The cost risk turned out to be pretty small, as for every patent there can be a workaround eventually, linux is based on code that is already known since the 60s, and some other reasons. In the mean time they made sure they had automated software install systems working, and other practical issues resolved. The big news now is that they will actually start with the first linux machines for office employees. First ones will be for office work that requires interchangeable software (word processor, etc), then more complicated office work will follow.
Parent
Re: (Score:3, Funny)
80% in 2 years? (Score:2, Interesting)
After reading all that, this seems like a lie:
"Schiessl said it would be impossible to migrate all users to open source, but that 80 percent would move across by between late-2008 and mid-2009."
Re: (Score:2, Insightful)
So they use the first 100-200 to learn and to develop deployment procedures etc. When that works, they roll it out to all the similar computers.
Just like in some smaller places, they use days to test something on one or two computers. When it works they spend an hour putting it on all 500 company desktops, most of the time just waiting for network transfers and rebooting.
Re: (Score:2)
Mayor's PC among the first (Score:5, Informative)
Just curious (Score:2)
Why Debian? Not that I'm implying that Debian is a bad distribution but isn't SuSE HQ practically in their back yard (Nürnberg) ?? Or has Novell uprooted SuSE development and moved the entire outfit to the USA ??
Re:Just curious (Score:5, Informative)
See here:
http://www.muenchen.de/Rathaus/dir/limux/english/
OK?
Parent
Re: (Score:2, Informative)
Re:Just curious (Score:4, Informative)
that site has general information about the Linux-Project and a link to this site:
http://www.ssrc.org/wiki/POSA/index.php?title=LiM
Parent
Re: (Score:2)
Firstly, thanks for a very German answer :D
Secondly, while that is a nice a site and I say that because this project interests me and I did take a look at that Wiki, I was hoping for a more detailed business and financial oriented explanation than "They will be deploying not SuSE but Debian GNU/Linux, the freest of the Linux distributions." The word 'Debian' is mentioned only once on the pace you linked to.
Re: (Score:2)
At the time, there was no OpenSuSE like now.
cheers,
Rainer
Re: (Score:2)
Just one more reason to admire this rethink. You might expect that they would move from the big name vendor to the biggest name Linux vendor they could get, but in the end the name doesn't buy you anything. What matters is that you're supported.
additional info (Score:5, Informative)
however, the main reason for the delays and the slow roll-out are that a lot of custom applications had to be ported and for some existing client/server apps interfaces had to be created from scratch.
cheers from Munich,
Andreas
They Tried (Score:5, Funny)
But Microsoft just couldn't get Longhorn ready in time.
Meanwhile, on the other side of the Rhine... (Score:5, Informative)
Some Open Source headways in Europe, indeed, can clearly be seen in EU site [europa.eu].
Quite heartening indeed! Maybe the big conservative companies will finaly notice this trend. I am sure Microsoft did.
Re: (Score:2)
(Nice link to the EU site BTW!)
Difficult Switch Benefits Linux? (Score:2)
I do wonder whether we'll start to see Microsoft supporting these Open Standards as a way to ease the migration path back - supported of course by heavy subsidies on licensing.
The headline is mis-leading! (Score:3, Informative)
The sentence should read, "Munich Finally Starts Implementing Linux."
The embrace happened a few years ago. It's (Linux) implementation is what has just happened. By the way...does anyone know whether it's KDE or GNOME at the forefront here?
Re: (Score:2)
Re:The headline is mis-leading! (Score:5, Informative)
Parent
Re:The headline is mis-leading! (Score:4, Interesting)
Besides, I'm uncomfortable with that as an explanation. I'd like to see a comparison of what they need from a desktop environment, what worked in one and didn't work in the other, what users needs and what admins need, and make a reasoned comparison on that.
Parent
Octoberfest (Score:5, Funny)
Can I be the first to suggest... (Score:4, Funny)
Chairs of Mass Destruction (Score:4, Funny)
Re: (Score:2, Insightful)
I honestly think it has something to do
Re: (Score:2)
I remember us smugly installing x-terminals everywhere to replace the PCs we had. It's the future! It's open! It's economical to run!
Within 3 years, several million pounds worth of x-terminals were junked and we were back with PCs. So much easier to get the software! People know how to use them! Easier to find support staff!
Re:Linux urinals? (Score:4, Funny)
Your ToiletFlush(tm); license has expired. Please supply your Credit card details using morse code on the ToiletFlush(tm); button. The ToiletDoor(pat.pend.) will stay shut for safety reasons until you comply."
Parent
Re: (Score:2)
The current Windows-based urinals are susceptible to the Piss of Death attack.