Part III of Desktop Linux – The Hard Truth is published. In it retired World Bank CTO W. McDonald Buck looks at what the CTO needs to consider in making a realistic desktop switch to Linux (-after- having switched to OS software on Windows & Linux in the server room) in a big company and how to avoid ending up toasted.
Part III: Corporate Desktop Linux – The Hard Truth
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
The stuff the author talks about is nothing more than common sense precautions to take when doing any migration of any kind in a large IT infrastructure. Of course there are going to be migration costs and risks. One just has to size them up and then look at the potential long term benefits.
But the author is also blatently fear-mongering. In practically every other sentence, he says “if such-n-such happens, you’re toast”, or “huge risks”, or he tries to over inflate potential costs.
So it’s somewhat hard to take this guy seriously.
Most certainly, doing a Windows to Linux desktop migration is not the right thing for every situation. But in many scenarios, it most certainly is the right thing to do. It all depends on the variables, like current Windows licensing costs, Windows security problems costs, Linux support costs, migration costs, training costs, mission critical application compatibility, and the list goes on.
Any wise CIO or IT manager is going to size all of those factors anyway.
But the article deliberately paints a doom and gloom picture for Windows to Linux desktop migration.
“You can start by migrating the browser, declaring truthfully that you’ve protected the organization from various security risks, though not saving any hard money. That done, you resolve to replace Office with OpenOffice on most of your Windows boxes at pretty low risk, actually saving real money.”
Yes, I can see the migration of the browser going somewhat smoothly; however there are typically internal apps that usually require IE.
Office to OpenOffice: Great soloution to saving costs, however, there are problems with not being 100% compatible. The only way this can change is by getting file formats published. This will have to be initiated though a large number of companies and government agencies. Something like the EU is doing with *.SWX file format.
I am looking forward to the day with file formats and standards compliant applications. Here is the crux of the issue. Once open, innovation will excel. Think about it, if you could switch an application because of poor vendor performance without any penatly of migrating data, would there be a mass exodus. At this point, true innovation will occur. Perhaps:
1)Performance optimizations on the applications.
2)How about usefull features
3)Resonable costs via competition.
Just my humble thoughts.
I kind of agree with the arther. but in a large corp IT shop, it would pretty expensive to do a one shot migration. you would have to do it in phases like so:
phase 1: backbone and web servers
phase 2: print servers and file servers
phase 3: other backend servers if possible
phase 4: new back end servers infrastructures
phase 5: web services and Browser
phase 6: Email and groupware and such
phase 7: access data bases
phase 8: office
phase 9: other software
phase 10: OS
would be the only real way to do it, and along the way, your IT staff that do not know Linux well would have plenty of time to get upto speed.
and also this way, the users would not freak out of one big massive change, they would just see one or two changes at a time.
-Nex6
In my experience most IT directors (or whatever they are called today) do not have a real clue about what is actually running in their company. Sure, thay know what is in the Standard Build (if they have one) but what about all the other stuff (and lots of it) that is installed that allows the users to actually do their job.
<dons-BPE-hat>
If senior managment were to invole the workforce and slowly introduce OSS software OR software that runs on Linux over an 18-24 month period the effect on the business would be minimal. Then you could pull the run so to speak and change the OS. This way the impact on the workfoce and their effectiveness is minimalised. Also, they have had time to get ready for the switch and replaced their favourite IM tool with gaim etc.
The result is $$$$ savings to company and users happy. Tes its all common sense but when you start talking to accountants & lawyers that goes straight down the swanee….
Thats my 0.02tenge worth.
I can see the servers and infrastructure that the users do not see. Those seem to be fairly big easy wins.
As someone who has managed an IT group, the changing of user applications on the users is fraught with danger. Every application you change is going to generate 4-6 weeks of complaining. Good, bad, it doesn’t matter, it is different.
So the cold turkey approach is the only one that works well. In every roll out I have ever done, the users have a problem and immediately want to go back to the old system, regardless of what that system was. The only way to minimize the bloodshed is to say on such and such a date we are changing. That switch requires the buy in from executive management to say “deal with it” to the troops. People will figure out how to do their jobs if you have provided new tools, but be prepared for the whining.
I am looking forward to the day with file formats and standards compliant applications.
Me too, closed standards are just a huge pain in the bum.
I also agree. Is there any movement that is lobbying for this? Does Electronic Frontiers or any other group have this on their agenda? It needs much more impetus from all levels of society.
This is the ONE REAL ISSUE on which we need to work.
Once people (and governements and enlightened companies) require that the formats be completely open (which means documented in all the details), then we won’t need to worry so much about how many people use proprietary software A and how many people use free or open source software B; all those who can’t live without MSOffice could use it without affecting my work with OOo, Kword, Abiword, LaTeX etc….
That is why I really don’t understand how come people, even on technology sites, don’t pay attention to the fact that for example the European Commission request that MS discloses all its software protocols is the one that disturbs MS the most, not just the media player (although the two things are somewhat related).
To Anon:
Regarding lobbying groups see OASIS, http://www.oasis-open.org/home/index.php, although recently there has been some disagreement about patents with EFF.
Hey, if a company wants to keep paying the MS tax, keep funding weekly bug hunts and security patch parties and spyware expeditions, GO FOR IT. Its clear there is absolutely nothing we can do to dissuade you, and its clear that no organization could ever function without Windows XP…its impossible…in fact I am surprised humanity figured out how to make money before XP.
Office to OpenOffice: Great soloution to saving costs, however, there are problems with not being 100% compatible. The only way this can change is by getting file formats published. This will have to be initiated though a large number of companies and government agencies. Something like the EU is doing with *.SWX file format.
That’s not necessarily the case. OpenOffice actually does a better job at translating older versions of the .doc format than Office2003. A switch to OpenOffice might actually be more seamless than a switch to Office2003.
Thanks Klaus.
Abraxas,
I do agree that some of the older versions work better with OO.o. However, large corporations do not typically use Office 97 or Word 6.0. Typically, Office 2000 and Office XP, 2003.
These are the foundations of large business. To potentially open up markets, there has to be compatibility across the board.
Home users and small business may still be using older version of MS Office because there is no need to upgrade. By using an alternative Office Suite or making it possible to use said suit would spawn more competition. Look at the browser market. MS has a huge influence on the standards.
IE: Now enters FF which can mirror most of the w3c standards; and what has happened? 25 Million downloads. MS may have a large influence via market presence but not total control.
Document the file format and you have choice. Ninety percent compatible just does not cut it, it must be 100%. This is a control structure, this is what keeps people upgrading and tied to one product. The products could then be evaluated based on features, needs as well as costs.
This is for public awareness. Open Standards is how the web was built. Look at: HTTP, HTML, TCP/IP, POP’s, SMTP and so on.