The Business Case for Open Source Software

An anonymous reader writes "An InfoWorld blog entry makes a business case for open source software, and attempts to explain the business benefits of OSS to management and business owners. The primary benefits the piece uses to argue in favor of OSS include no licensing fees, and no license keys. The article also argues that OSS results in freedom from 'ownership' by software vendors. 'Never again will you fear the BSA (Business Software Alliance) knocking on your door wanting to perform a software audit. The BSA even takes out advertisements on Google search pages for and up to $200,000 reward a disgruntled ex-employee can receive for reporting your company to the BSA! That's quite a powerful motivator...'"

Comments

OSS

[Savage-Rabbit]

The primary benefits the piece uses to argue in favor of OSS include no licensing fees, and no license keys. The thing I like the most about OSS is that I have everything at my fingertips with OSS, they only delay is the time it takes to download stuff and install it. When you are workign with proprietary tools it's the same, you do have everything at your fingertips, except you also have to wait for the license costs to be approved by the bean-counters. Somehow I feel that I get things done quicker with OSS because I can bypass a whole layer of corporate bureaucracy. On the other hand quality of OSS software can be low, documentation often sucks and user friendliness is also an issue although with some proprietary stuff such as certain Oracle products for example user friendliness is nothing to cheer about either and I have seen proprietary software that made me wonder where people get the nerve to demand money for such crap.

Re:OSS

[kebes]

To my mind this is the actual problem with OSS. Accountability is nill. With MS products the same defects are there (though less so as it turns out (predictably)), but in their case at least we know who to blame and can expect the product to be fixed. With OSS I see no way to assure that.

That's not even remotely true. You expound the myth that there is accountability in proprietary software, whereas there is not with OSS. In reality, after you pay for your proprietary software, you have absolutely no guarantee of bug fixes, and no guarantees that changes to the product won't break backwards compatibility (e.g. "mutate"). Don't like it? You can either stick to the current version, or buy the next version, or pay them more money for support contracts that make guarantees.

With OSS, after you freely download the software, you also have no guarantees of bug fixes or interface stability. Don't like it? You can stick to the current version, or freely download other versions, or pay those who make the software for support contracts that make those guarantees, or pay a third party to make those guarantees, or hire people in-house to modify the code to suit your needs, or contract a third party to make those code changes, or port your data to a different software product.

In any case, it's up to a business to evaluate their software needs on a case-by-case basis. But please stop spreading this "because you pay for it there is some guarantee of accountability" myth. Anyone who has tried using phone support for commodity software, or who has read through an EULA, knows this to be a joke.

Re:My experience

[MMC Monster]

I'll take this as sincere and not a troll.

Get a new lawyer for the company. The GPL states that whoever has access to the executable should have access to the source. You said yourself that you only wanted to release the executable within the company, so the GPL doesn't really apply.

As for gcc, if you modify gcc itself and send out copies of gcc outside your company (which is *extremely* unlikely; if you have resources to do that, you are not in the right field.), then you may have to release source. Otherwise: gcc is a tool to compile a program, just as pencil and paper are tools to write the program. You are not bound by the GPL on what you write. Now, if you link to a library that is GPL (not LGPL) licensed, you have to release source to whoever gets the executable.

While here in India...

[jkrise]

the business case for Open Source Software in the enterprise market is already well established. Some reasons:
1. The average IQ of the 'EDP Manager' .... (or should I say IQ of the average EDP Manager) seems much higher than elsewhere... so he can't be fooled forever.
2. Closed source software is so very expensive, enterprises choose to build their own systems; and they mostly choose J2EE and Eclipse. The LAMP stack is packing up with amazing velocity as well. ROI can be seen in a single year, with many apps.
3. Not much of lock-in has occured already - very few companies have data locked in .doc formats... not many firms have BI or Analytics... so leap-frogging ain't a big issue.
4. The hardware specs are roughly 10% in the OSS space.... and that matters a lot as well.
and lately:
5. It is getting more and more cumbersome pirating Closed source s/w - be it OSes, Office, SQL or whatever. Most EDP mgrs over here have been on the same company for a decade on average; and they're pretty amazed at what OSS can do.

A recent Java conference (paid, mind you) had over 10,000 attendees! RedHat is doing very well... not many people know or care about Novell... many state govts. have mandated and stipulated Open Source specs...

Somehow, people this part of the world do not seem to wait for Gartner reports or NYT articles before experimenting with OSS.

Re:My experience

[Zonk (troll)]

Although we met several technical challenges along the way (specifically, Linux's lack of Token Ring support and the fact that we were unable to defrag its ext2 file system), all in all the process went smoothly. You've used Windows too much. ext2/ext3 do not need defragging. If you insist of defragging, then use XFS and run xfs_fsr to defrag.

So you can imagine our suprise when we were informed by a lawyer that we would be required to publish our source code for others to use. You need to get a lawyer that has decent reading comprehension. You only have to distribute the source if you distribute the binaries outside of the organization.

It was brought to our attention that Linux is copyrighted under something called the GPL, or the Gnu Protective License. GNU General Public License.

Part of this license states that any changes to the kernel are to be made freely available. The GPL is only a distribution license, not an EULA. If you don't distribute it outside of the organization it doesn't apply.

Furthermore, after reviewing this GPL our lawyers advised us that any products compiled with GPL'ed tools - such as gcc - would also have to its source code released. This was simply unacceptable. You seriously need lawyers that can read. Take a look at OS X. It's compiled with GCC and it's not GPL.

I think the biggest thing keeping Linux from being truly competitive with Microsoft is this GPL. Its draconian requirements virtually guarentee that no business will ever be able to use it. Read the full text of the GPL. Read the full text of the Windows XP EULA and the EULAs on all of the updates you have to apply to not get 0wn3d. Which is draconian? Hint, it's not the GPL.

Same here, and more...

[mangu]

When you are working with proprietary tools it's the same, you do have everything at your fingertips, except you also have to wait for the license costs to be approved by the bean-counters

This is one of the two main points for OSS that I have experienced. The second important point is that with OSS your system is able to survive the vendor. Where I work we have a 400000 lines VAX-FORTRAN software that we are struggling to migrate. Although we do have the Fortran source code, migrating it to any other Fortran is very costly, we have the choice of doing it ourselves or pay about $250k to outsource the job.

I think our experience shows the importance of going all the way in OSS, the operating system, utilities, compilers, etc are just as important as the applications. That's why we are migrating our system to g77 on Linux, instead of using one of the several commercial Fortran compilers whose vendors claim VAX compatibility.

Re:Same here, and more...

[senatorpjt]

Where I work we have a 400000 lines VAX-FORTRAN software that we are struggling to migrate.

Yeah. And, I bet that when those 400,000 lines of code were written, the idea of DEC folding was about as plausible as the idea of Microsoft folding.

Re:Business Case? How about home case?

[Technician]

It is more noticable in XP then in Vista. It seems like every 2 weeks I am installing WGA, in Vista it must be happening in the background because I havent noticed anything yet.

In XP it is a add on patch. In Vista, it's built in on the ground floor. Do a google search for Vista false positive. Pick any item on the first google page. They all relate to WGA problems on Vista.

Flamebait?

[cgenman]

It's quite different over here. It's not just a question of experimenting with OSS, it's a combination of seeing the job as deciding between presentations from different vendors and being averse to taking personal risks.

As I mentioned, it seems like people have stopped doing their own research and now mainly choose between different schpiels from different vendors. Vendor 1 selling you something for 100,000 dollars, and vendor 2 selling it to you for 50,000? Clearly if you go with vendor 2, you've saved the company 50,000 dollars a year. No need to point out vendor 3, who doesn't have a substantial sales team but who sells something identical for 5,000 dollars per year, or an OSS solution which might need 1,000 dollars per year worth of tweaks. Or maybe it makes sense for you to write your own. We generally have a 2-vendor solution, and nobody can fault you for choosing the better of the two, right?

The risk-aversion deepens. In corporate US if you create a product that everyone else is making, your job is reasonably safe even if it tanks. And, in fact, simply because everyone else is making it, it's likely to tank. On the other hand, if you create something original (i.e. something with an open market) and it tanks, it's more likely that your career will bear the brunt of that mistake. OSS is currently viewed a lot like that. Taking risks is largely regarded as a negative, and certainly regarded as a dangerous career move.

One does not experiment with OSS, because one does not experiment. One either knows for sure, or one contracts a vendor who will bear the brunt of the responsibility when things might go wrong.