We think we have the OSS business model. See the link below. For those of you who hate Marc, try to pretend that this did not come from him and read it with an open mind.

I?m sure, big vendors will benefit from OS and OS will increase monopolization in software industry. Just think who pays a salary for "key" people in Apache or Eclipse projects.
I do not understand why music industry is fighting till the last man against "free distribution" of songs and software industry doing exact opposite?

Best,
Igor.

Why do you think that removing "money interest" from software vendors will make any good? All my previous experience from socialism tells me exact opposite. For now I can see that OS will kill many small vendors just because people will prefer "free" to "good" software. And, I

I have oftern found that open source software is better supported than the commercial equivalent. I reported a bug on an open source project yesterday and receive acknowlegment and a fixed source file today.

Last year I found a bug in a commercial database, provided scripts that could reproduce the problem and included build data and the exact version that the bug was introduced. The automated reply took one day. After a week I got an email suggesting that I try the latest hot fix.

I may be missing something, but unless this article has been written in response to some over-enthusiastic open source propaganda, it is completely pointless. The main message seems to be that 'open source' does not equal 'fantastic'. What a surprise! If this isn't common sense, I don't know what is. Personally, I am very happy with the proportion of open source output that is of excellent quality and there is a lot more of it than any single developer could consume in a lifetime. Quality aside, it is the underlying ethos of openness and sharing that is unbeatable. And just in case you wonder, I do write commercial software.

Sorry to say, but:
This is yet another irrelevant article.
None of the statements about open source go beyond what common sense tells people anyway. Some statements are pure assumptions derived from an insignificent amount of people and projected on the mass of developers.
Actually, most programmers would deny to be as naive as you try to indicate.

Especially as most have seem both: free rubbish and free gold.
And the ones who have browsed through the tons of open source projects have also encountered that some are dead, dead, dead.

So what's left after subtracting all this from your article:
- A speculation about what can make open source projects successful.
- A bunch of unfullfilled expectations about hard facts and problems companies face when using open source software.

I work with a bunch of open source enthusiasts that really believe commerical software is evil and open source is some kind of a salvation.

At times they refuse to work with some commercial tools and waste time looking for some open source alternative.

They simply don't have a clue about writing commercial software, all they do is whine about re-inventing the wheel and I (among a few others) spend time with a lawyer to find out that we can really use these open source offerings without turning our own product into open source.

I'm really beginning to think OS is the choice of bad programmers, the ones that have lost the will to create something of their own.

Exceptions make the rule, I do admit a few OS projects really deserve all the respect, but these are the ones that have taken care of their licensing as well.

When in doubt, leave it out!

In 20 years of experience I have
observed that management in commercial software production , both in-house IS departments and software houses have
very little regard for software engineering principles or practice. From
looking at and using much open source code , it is completely different , it
seems as if the authors have read all the books and put it into practice and
they care that the software is properly
structured. This makes an enormous difference in quality and changeability.
The only real downside is an endemic lack of documentation!

You do have to evaluate the open source software you are going to use and make sure it has had enough people behind it and that it has proven itself. You can't just take any open source project, and assume it's high quality.

We switched from a very popular and expensive app server six months ago to Open Source, and our problems have been solved. We ported the application without changes, and it just runs. We don't have to worry about license fees, maintenance and/or additional overhead of application loadings that come with at least some of the commercial products. Our stability has been phenomenal.

Why has JDJ in the last few editions blatantly put articles that are nothing more than ads for Oracle App Server, PeopleSoft etc.? No wonder, there is another aginst open source - who paid for this one?

Opensource is a gift to programmers and developers. It unites programmers all around the world. If anyone says commerial products are better than OS products, there could be only one reason. Everyone who work on a commercial product gets their daily bread and satisfaction. OS developers build the product part-time 'caz they don't get money from what they doing. The only satisfaction to OS developers is the outcome of the product and its success. Because Linux Trovalds had that satisfaction he gave Linux and many others like him..

Nobody funds projects developed at sourceforge!!

Don't even compare mySql and Oracle for determining the whether OS or a commercial product is better. Because everyone here knows how many millions were spent for developing the Oracle database to what it is now..Don't compare any OS products with commercial for that matter..

OS is great and unique.. It will dominate the industry in one-day..My only request to all developers on OS, do not lose hope in the effort you put in. Stay united and stay with OS and improve how projects on opensource can get organized.

Regards
Gopal

The conclusion of this article is missing : "Every architect should keep in touch with open source."

The author is right, OSS is not a silver bullet. OSS comes in many different qualities. Management should give their software architects time to stay in touch with the OSS projects. This avoids wrong architectural decisions on the one hand and allows much more widespread usage of OSS-projects on the other hand. Too many managers select an expensive commercial offering without knowing their is a viable OSS-alternative.

Regards,
Tom Baeyens
Founder of http://jbpm.org

If they did, I'd load up my magic revolver with magic "clue" bullets and shoot you in the head.

--Kt

Open source may not be a "magic" bullet, but it is a bullet just the same...and what could a "magic" bullet do that a regular bullet couldn't do anyhow (aside from being free)?

I don't agree with the alternative term for "open source" as being "commercial sw". It should be open source and non open source. All successful sw products are basically commercial in nature. Even TCP/IP can't be used without some investment in infrastructure. As for the value of having source (and free speach and beer (maybe) for that matter) it certainly diminishes completely at some point. For example, consider a 100% Java ORDBMS. If it was written in C open source would only allow you to freely port it around. Taking on the task of learning and supporting the source is certainly nothing anyone would do lightly, even if they had just won the lottery.

Open Source Software started as a response to the poor (and expensive) software coming out of Microsoft and other large software houses. Compared to what Bill Gates has shoved down the throats of the computing public, OSS is much better and less expensive. But a lot of CIOs don't seem to get this. They will continue to hemorage money in Microsoft's direction because they don't want to know any better.

OSS is not a magic bullet to MS's gaping wound, but it is a bandage. If only Bill would take his meds...

Just my .01 cents:
I have worked in many projects in many companies, and I have never seen the code shared as it is on SourceForge.
How many times inside the same company a group will retain control on the source code of one app?
How many useless meetings are held every day to convince other developers to share their source and knowledge?

Even if the project is dead, at least they are available to the community to learn from it.
Even a dead project will tell you that someone else have already encounter this problem, and made an attempt to solve it.
In a commercial world, failures are silent buried and forgotten..

Yes, OS it is no silver bullet. But IMHO is a step forward.

No developer think (it's my thought) open source is better than commercial software. But, many developers ask ourselves what software component is better to solve our problem. The answer sometimes drive us to open source and sometimes to commercial software. So my conclusion is: open source is not worst than commercial sofware.

Both the article and the responses so far make a number of great points. I do certainly agree with the premise that Open Source is not a magic bullet, but OSS has a number of advantages in addition to what has been discussed so far.

Using resources like the Trove, I can examine the amount of mindshare and attention a project gets -- has it reached or will it reach the critical mass discussed above? With commercial software this is not always as easy. The ecosystem of less well-known software can be opaque to say the least.

I can download the source and design documents and within a short time get a first impression of how solid the engineering is. How often did it happen to you that only after some serious use you find that the commercial package you adopted has been badly engineered?

Finally, having the source is a huge advantage when you encounter the inevitable problems. My experience of commercial support is frankly abysmal; it usually takes me less time to find a bug in source than it takes me to prod a customer support department into action.

Many books on software development that I've read claim that upwards of 70% of all software projects fail. There's just no way to have stats on commercial projects that fail because what company would want that kind of press.

Yes there is no silver bullet, but what open source gives the software community is a chance to produce better software (higher quality & more useful to more people) and the fact that there are dead projects is simply part of open source. It's software evolution. Those are worth pursuing so no one does. On the other hand how much money is spent on commercial software projects, that are doomed to failure but are forced to continue, because of political pressure?

I like and agree with the artical as well, yet open source doesn't compare to socialism well. The open source model has many more similarities to the University model of open and collegiate research than it does to socialism. In socialism, there is a tendency towards forced redistribution of profit, but open source is specifically non-profit.

Rather than open source being opposed to free enterprise and capitalism, it bolsters it by enabling smaller vendors to enter the market by having cheaper tools, and it promotes the advancement and betterment of software because of it's open and transparent nature.

Just like a free and open system of scientific researchers in universities (and to a lesser extent in government and corporate research when their results have been made public) has contributed much of the science foundation for the technology based capitalist organizations that are around today.

True that not all opensource software equates to quality code, but surely one of the beauties is that when a project gets it right, it becomes very popular and the "average" developers like myself can take advantage of it (saves time and money, plus you learn a few thnigs about good design). Look at things like JBoss, Struts, Hibernate to name a few). I doubt even the MS money machine can afford 70,000 trial and errors to produce the same number of successful projects.

I agree entirely with this article. Personally I think it highlights the benefits of a community source project development model.

One interesting notion is how effectively open source plays in a capitalistic software market, and what that means to the market. It appears that a largely socialistic mechanism (open source) is being used to battle the monopolistic practices used by the large software vendors. The small vendors are left to uphold the capitalistic endeavors and are struggling somewhat in this down economy.

I think the open source community has been a direct response to the control of software being dished out by the software monopolies. It is interesting that all of the cited successfull open source projects also have some community or company generating income indirectly to those efforts.

It still takes revenue to sustain momentum. But I do like to see companies finding more innovative ways to compete with the monopolies. Given the total failure of government in dealing with them, open source may be the only way to get new innovations into the market.

It's not so much of a silver bullet as it is a free bullet. Sourceforge is a model of Darwinian evolution. It has 70,000 ideas that evolved or becamse extinct. We just know about the extinction rate because the fossils are presented in a nice searchable GUI.

I don't think many people are so naiive as to assume open source = better. Anyone who tried to run netatalk, a Mac file service for Linux, found out fast that that equation is flawed. Linux itself is a work in progress, as is Windows, Oracle, or any other software in use.

For me, open source is about control. Linux and Apache are configured with text files, no secrets, no broken registry. Java setup is a pain in the neck, but you know what you have and how to work around it. This is in direct opposition to Microsoft and most other Windows software providers, who hide and obfuscate their products to make money.

> How many of the projects on SourceForge
> are dead or dying?

Although I agree that the "Open Source Is Not a Magic Bullet", I didn't like the approach with the number of the idle projects in Sourceforge. If that's the success criterium, we should consider "How many commercial projects are dying or being unsuccessful?". I'm pretty sure that the number is lot more than the idle-open-source ones.