EU Commission Study Finds OSS Saves Money

PS3Penguin writes "Groklaw has up a story about an EU Commission's recent findings on the costs savings available from using Open Source Software. From the article: 'Costs to migrate to an open solution are relevant and an organization needs to consider an extra effort for this. However these costs are temporary and mainly are budgeted in less than one year. The major factor of cost of the new solution - even in the case that the open solution is mixed with closed software - is costs for peer or ad hoc training. These are the best example of intangible costs that often are not foreseen in a transition.'"

Comments

No surprise

[kamochan]

This does not come as a surprise for people having worked in IT and with OSS for some time.

Now, if this report gets public bodies to use and require use of OO/ODF, the large corporations (whose customers or legislators the public bodies tend to be) might move to OO/ODF as well, and then also us small subcontractors could finally junk the P-O-S, all-defaults-are-nonsensical, pay-for-incompatible-upgrades MSOffice. Someone just needs to get the ball rolling...

Damn, it's good to see the EU bureaucracy sometimes produce sensible results!

But

[El Lobo]

One of open source's most touted benefits is its price. Download the software, install it--and don't pay a penny. That's the theory. But to a surprising number of companies, the price tag--or lack of one--is irrelevant. Believe it or not but in my university there are no problems to choose software. We are not looking the philosofical part of the questin (this is OS, this is not). We literally don't care for that. We look at what does the job best. And we buy and use it. And don't care for the price.

Re:But

[Leftist Troll]

But to a surprising number of companies, the price tag--or lack of one--is irrelevant.

Many of those still choose open source software. There's a reason GNU, Linux, BSD and Apache are so widespread, and it has nothing to do with price.

Training cost?

[Realistic_Dragon]

I am a manager with a masters degree in engineering. My charge rate is well past $180/hour.

I spend about 1 hour a day telling other members of staff how things work in Excel. That's Excel 97 by the way, which we have had deployed for over 6 years.

Retraining costs only apply if your staff are trained in the first place. In the world where *everyone* puts "Office expert" on their CV almost no one is trained - at least not to a high enough standard to do anything beyond typing a letter.

With the interface also changing in the next version of Word this cost is even more fictional than ever - but it was never legitimate in the first place.

Re:Training cost?

[donaldm]

I have just recently put Fedora Core 6 on my new HP dual core 64 bit AMD laptop which I purchased (no dual boot ether). I did have a few minor issues which could have been solved with a HP configured install and recovery disk instead of the Microsoft XP professional it came with. So far I can actually do nearly all my work with this OS with the exception of some Microsoft specific solutions that requires I use my company laptop but 95% of my work can be done using FC6.

For a scientist or professional engineer I would strongly suggest a Linux solution (FC6, OpenSUSE, Scientific Linux ... etc) than a Microsoft one even though you can get most of the applications you like for a MS OS (sometimes free as well) at least you will fully own all your data and never have to be dependent on a proprietary Operating System. Actually IMHO LaTeX is actually easier to use than IMHO most word processors and the result is normally very professional. This is especially true if you need to write mathematical papers. You still need to know how to use a test editor though. As far as any type of development that requires maths a good Linux distro can provide everything you need (if you do any type of statistics have you tried "R" since it is like "S Plus") and again for free or cheaply.

If anyone writes to me stating "Oh you had problems with a Linux install on your laptop then there is a problem with Linux". My simple answer is I will give you FC6 or OpenSUSE and Microsoft XP (legitimate copy) and then ask you to install the OS and configure it on a reasonably new laptop (I am being fair here) and I am quite sure you are going to have more problems with the Microsoft OS than with a Linux OS. Since I now have a working Laptop with FC6 (what do you think I am using to type this) I can easily create a recovery disk that could be used to configure all laptops of this type. The first install is always the hardest after that you can easily roll out an OS on equivalent machines, this is how most PC vendors install an MS OS.

Now back on topic. If you are a manager and it has been put to you that you need to spend vast amounts of money to retrain your staff to the switch from MS Office to Open Office, then I would suggest firing people and I am not just speaking as a professional engineer I am speaking as a manager. Most MS documents can be imported into Open Office (including many with macros) with little if any changes needed. The only problem you have is when you try to read an OO document back into MS Office. That in itself should tell you how standards compliant Microsoft is.

The biggest problem an organisation is going to have making a switch to Open Standards (note I did not say Open Source) are the managers who will most likely say "Oh it is not like MS Windows" or who have made bad business decisions although to be fair to them they may have made the right business decision at the time, that have locked the company into proprietary solutions.

Sometimes you have to force change (the engineer in me speaking) otherwise things will never change since most organisations are very conservative and won't change unless a decision comes down from the top but sometimes the top managers are even more conservative or love to organise committee's, which usually means nothing changes.

I've Been Saying This For Some Time

[Master of Transhuman]

No mattter WHAT it costs to transition your people, those costs can be amortized over time. Whereas paying proprietary software license fees is FOREVER. By definition, sooner or later OSS HAS to cost you less - not even taking the intangibles of avoiding lock-in, flexibility, etc. into account.

The only issue is whether you can afford the upfront costs - and that has to be decided on a case-by-case basis. And you solve that issue by doing your migration over time according to a PLAN.

Planning? A novel idea for most IT management who are usually locked in to a crisis management mode...

Re:I've Been Saying This For Some Time

[Volante3192]

Have tou ever heard someone in the next office over going click, click, clock rapidly about 50 times? I'll bet you they were using Excel.

If that was me, it's Minesweeper.

Where is the control group?

[Anonymous Coward]

I always see the studies about the costs of migrating to Linux. But they never adequately explain the control group.
To be of any real value, you have to compare the Linux migration costs to some control group.

Here are some possible control groups:
1. Group transitioning from Windows95/98 to Window XP to Windows Vista
2. Group transitioning from Windows95/98/XP to Mac
3. Group transitioning from Mac to Windows Vista
4. Group transitioning from Windows95/98/XP to LTSP
5. Group transitioning from Linux to... Linux?
6. Group transitioning from Windows NT to Windows 2003 to Windows Vista

It seems that the control group in most of these studies is only imaginary: Windows XP with no transition.
That control group doesn't exist. It is never actually included in the studies. It is only conjectured.

What is the value of a study that uses an imaginary control group?

Re:Expect a response saying the exact opposite

[Amazing Quantum Man]

[SARCASM]
Don't you know Vista is Free with purchase of a PC?
[/SARCASM]

I guess they never got the memo

[oOo Shiva oOo]

Wasn't it well established in an open letter that open source is dangerous and could derail the European economy? :)

At this point, I would somewhat dissagree

[pembo13]

The forces who do not want to see OSS succeed for their own financial reasons will do what ever it takes to make sure your costs go up. If Linux usage spikes next month (for example) I would except to see a rise in underhanded tactics as well.

That loud clapping noise you hear..

[Rob T Firefly]

..it's the sound of millions of Slashdotters facepalming at once.

Price doesn't matter

[duffbeer703]

Vista will make cost irrelevant.

Lots of companies and most governments are going to be mandated to use whole-disk encryption for laptops and desktops in the next year or so. The easiest way to do this is to get your hands on Vista Ultimate or Vista Enterprise.

This is a problem.
Vista Ultimate is a consumer product and you cannot get it via a volume license agreement, so that's out.
Vista Enterprise is available via volume & enterprise agreements but you must have software assurance agreement in place.

To get software assurance, you pay Microsoft a "seat fee" equal to the number of computers that you have that aren't:
  - Servers
  - Applicances (VPN devices, Google Search boxes, etc)
  - Kiosks (ATM's, POS terminals, etc)
  - Embedded devices (Treos, Blackberries, etc)

That means that you'll pay Microsoft for Macs, Linux machines, FreeDOS machines... anything that is a workstation. So switching to Linux won't save a time, because you'll pay Microsoft anyway!

OSS might save money...

[Schraegstrichpunkt]

... but I still prefer ALSA.