Open Source: The Model Is Broken | Hacker News Reader

Open Source: The Model Is Broken

businessweek.com

19 pointsstillmotion14 comments

What a poor article.

It's pretty much completely devoid of any evidence to substantiate any of his claims.

He states: "The open-source business model is broken." and makes vague references to companies that have faltered but gives no hard numbers, e.g how do the numbers of companies using an open source business model failing relate to companies with a proprietary model?

The article's filled with other assertions that are backed up by vague statements. I mean:

"And therein lies the great paradox: Open-source code is generally great code, not requiring much support. So open-source companies that rely on support and service alone are not long for this world."

A rampant generalisation, followed by a complete lack of any evidence.

Yeah, it's not that good. I voted it up just because I think open source business is a fascinating subject and interesting thing to discuss, and if nothing else, the guy was quite involved with it at OSDL.

My take is that 1) open source works very, very well, but that 2) "opens source business" is not really a solved problem, and that the economics of the whole thing are weird and interesting.

My impression is that "sell consulting support for open-source products" is a good business model for the individual consultant (in the sense that if you're good at it you can make a decent living at it). It's not such a good business model for an investor looking for growth, because the consultants don't leave behind much profit for the investor, and the only way the consulting firm can grow its revenues is by hiring more consultants.

That particular problem would imply that consultancies are not a scalable business. There are large consulting firms. So that seems off in one way or another.

You can build a small consultancy into a large one by hiring more bodies, and some firms have done that. But you don't get much economy of scale.

The article states that "The open-source business model is broken".

This is wrong, because there is no one "open source business model"; there are lots of them. ESR lists 9 in The Magic Cauldron ( http://www.catb.org/~esr/writings/cathedral-bazaar/magic-cau... )

The article does have a grain of truth in that open source does close off some possibilities for making money from a product.

BusinessWeek is one of the few magazines out there that is able to reliably trigger the "that's not even wrong" reaction from me.

Open-Source is not a business model; it's an element in the business environment sure; but in and of itself it has only the most tangential relationship to the flow of money.

And yes, some companies, Sun in particular have badly botched their open-source efforts, but that's mostly because Sun is in trouble in their hardware business and still thinking that it's the mid-90's. But for every Sun having a loud and spectacular failure; there's a Google or a Facebook building their success on open-source infrastructure.

//BW is just bad, they get get facts, quotes and names wrong too often to be an accident.

"For anyone who hasn't been paying attention to the software industry lately, I have some bad news. The open-source business model is broken."

Yes, if you haven't heard that open source as a business strategy is broken, it's because you aren't paying attention. Or perhaps the real reason you haven't heard that is because not many people are reaching this conclusion anymore after so many astonishing open source exits in the last year alone.

"This will require many to adopt a new mindset, viewing open source more as a means than an end in itself."

Agreed, giving away software and source code for free does not directly translate into revenue. Point scored for knocking straw man on his ass.

"The traditional open-source business model that relies solely on support and service revenue streams is failing to meet the expectations of investors."

I think the only thing disappointing investors about open source business is missing out on opportunities like:

"Consider Sun MicroSystems' (JAVA) $1 billion acquisition of open-source database software vendor MySQL."

Now I know it wouldn't have supported the position about how much open source business sucks, but it's worth mentioning some more examples here:

January 2008 - Nokia buys Trolltech for $150M

One from my own industry, open source network security:

March 2007 - Sourcefire raises $75M in IPO

I'm not sure exactly how many other software companies went public in 2007, but it wasn't very many. 5? 10?

Call me naive, but I think it would be great for OSS to have a license that is free (as in freedom), but free as in beer only as long as a certain revenue threshold is not passed, basically, a "pay when you can" model.

Why shouldn't a company that makes 100k/month using linux, etc. spend 0.5% or whatever of their revenues supporting that infrastructure. It could go a long way to improve OSS project funding, quality and long-term maintenance. (I guess it would be akin to a tax).

Another interesting idea with so much hosted software today would be to adopt a sort of "Affero LGPL" (require hosted software to publish the source & modifications of the code, but don't force the whole code base to become GPL). Perhaps allow a fee for keeping modifications proprietary. Not for ideological purists, but pragmatic imho.

Just trying to throw some ideas out there for discussion.

The problem is that it makes things very, very complex. What if GE has 10 linux servers - how much do they pay? How about a little startup (that makes 100K per month) that is entirely based on doing custom Linux work? What if I'm in charge of a "skunkworks" project at IBM, with just me and two people working on a new project - do we have to pay?

Basically, you take away one of the critical freedoms of free software with that kind of license, as much as it would be very, very tempting to get some of that money recycled into developer salaries.

And further, who would actually receive the payments? The programmers who wrote the code? The folks who packaged up the distributions?

Would it be useful to have some sort of open source payment organization, a la ASCAP for composers, such that a general fund comes from donations from happy software users, and based on some formula, members of the organization get payouts based on the frequency of the usage of their software?

Why mix the business model with the license? If you are the sole owner of the code, first release it for pay/donations whatever, then once you cross a revenue threshold, release it as open source?

What happens in the case of forks? If I decide to fork some open source software under this license, make a bunch of changes and start selling it under this license, what do I owe the original authors? Now how about if the changes were minimal or amounted to a complete rewrite?

I'm not really sure that I can think of a way to have a license like that which doesn't rapidly become overly complex when forks are thrown in. And if the freedom to fork is taken away, I don't think that I would consider the license to be free as in freedom.

Heh. so the article basically says 'open source is really good for software users, therefore it is bad for software vendors' - perhaps they are right, but that seems more like a positive than a negative to me.

Personally, I wouldn't want to buy software from someone who wanted to lock me into something broken.

Reply on news.ycombinator.com