Slashdot Mirror
Can Open Source Companies Stay That Way?
JoeGee writes: "According to this article on ZD Net, more and more companies born from open source projects are beginning to move towards closed source products as a source of revenue. Version 5 of GFS will be closed source, and even SuSE's director of sales Holger Dyroff has a quote that seems to disparage the service model of revenue. The one company that refuses to change its operations is, surprisingly, Red Hat. Red Hat CTO Michael Tiemann says 'We believe the Red Hat brand stands for open-source.'" Yes, this is a dupe. Bad Tim! *whack*
I am aware of a number of medium sized software developers producing specialist systems for specific industries, who are considering 'going open' and charging solely for deployment and support. I come into contact with these systems pretty regularly through my work - and usually call the company to ifnd out what their plans are as much through noseyness as need.
They are considering it for a pretty simple reason - giving software away, and making it open so the client doesn't HAVE to buy your service agreement, gives the client great confidence in YOUR confidence in the quality of your service.
Most users of such systems understand that the service component of the charge is the 'expensive' part anyway. By going open source a company can relaunch, give away the software and offer 'as you need it' support at rates likely to undercut the opposition.
Open means customisable, which opens up another potential revenue stream to the producing company.
It also lets the pain in the arse customers do some modifications themselves. One or two of these can account for 60-90% of ongoing support effort for some of these companies.
You don't have to go 'open source' to follow this business model, but its a great shorthand, and a great differentiator. Anyone work in a firm thinking of following this route?
without having to use a closed-source business model is to sell support and service contracts for the software. I work for a company that offers a free software product (it's closed source BTW). The customer has the option of downloading the software for free, or can purchase the same product and get live tech support (i.e. talk to me on the phone) for a year. I believe Red Hat is doing this already for Red Hat Linux. If Red Hat were to offer free support on every product that they gave away for free, they would not be in business for very long, and the only users of Red Hat Linux would be those who either work with the code or us anti-M$ /.ers who answer tech questions with "RTFM you moron!"
In the end there will probably be a small handful of business models that are really successful. But things are still evolving to quickly to really know what will work. Especially given the current 'irrational negativity' (in contrast to the 'irrational exuberance' of recent memory), it's too early to tell what will work and what won't.
RedHat is, in a way, in the same position as IBM. They've already established a strong reputation and consulting organization and don't need proprietary IP to compete. A small company with no track record can't successfully compete with RedHat.
The same isn't true for a lot of smaller Open Source companies. Small companies can spend a lot of time and money developing an Open Source product, and then find a competitor selling against them using the same product, but with no investment in R&D. The client can't tell the difference, so in the end it comes down to straight marketing, with no points given for actually having developed the product.
My company has faced this situation in the past and now we develop custom proprietary applications on top of an Open Source platform. We still believe in Open Source, but for now we need to keep some stuff to ourselves in order to compete effectively. I'm hopeful that over time we can swing back towards Free software - after all it does help to produce better software. Perhaps after we've had more time in the market and are a little more established. Or perhaps we'll open up certain pieces of the product while keeping more specialized functionality to ourselves. It's hard to say.
It is clear that Open Source/Free Software is here to stay and will take a big chunk of the software market, but individual participants have to find a formula that works.
It is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail. - Abraham Maslow
It's the developers you are paying. Why are we paying for software? It doesn't cost anything to copy software, just 5cents or so for the cd and however much for the printer manuals and box, if you even buy that part. We should be paying for developers time instead, this is where open source fits in.
Let's take an example. The Linux kernel is free, everyone can download it. So how do people make money off of it? Contracts. A company can hire out a person to write a kernel module for their new hardware. The developer gets paid by the hour to write the module and the module can be released open and free. Then later another company can hire someone else to make another module and add it to the kernel. All these contracts may be small, but they keep adding to one big project.
Now this only works with Free Open Source software that can be incrementally improved. A large game for end consumers won't work with this model. What company is going to pay a large group of developers to make a giant game and then release it for free? Maybe as Open Source, but not free.
So Open Source developers should get into contracting to work on open source projects that can help large companies. If a project is Open and Free, it can help everyone and save money for large companies as they don't have to pay for large software packages and then upgrades... and Open Source developers can still be paid.
Outdoor digital photography, mostly in New Engl
The point isn't whether Open Source companies can stay Open Source or not. The companies could all go bankrupt and it wouldn't stop Open Source. Before it had a catchy name, Open Source was just people writing programs that they wanted for their own purposes, and sharing them with other people. The loss of a few trendy business models won't change that.
In fact, it might actually help Open Source in general by sweeping out all the cruft, just like the current slump is cleaning up the dot-com fad. The people who are left will be the ones that develop Open Source software because they just care about having the software, not because they want to capitalize on a freely available army of developers.
Before, developers (or their companies) wouldn't openly release things that they felt really created a competitive edge. (Non-software companies didn't try to sell such systems either. They kept the advantage for themselves.) Now, these Open Source companies are trying to make a profit from creating software that, by definition, is their competitive edge. And they want to release all the source? Not likely. I don't want to sound like one of those people who yammer about how Open Source advocates shouldn't want everything for free, but it doesn't surprise me that these companies are dropping off and selling out. In the end, it doesn't really matter- the heart of Open Source exists outside of these companies.
--WH--
Lets use an example of why I think things are going wrong.
Supplier A "gives" software away and sells support.
When times are good customers pay for the support, well because they can afford it. When times are bad, people cut overheads, one overhead that could be ripe for major trimming is outside IT support. Now if a customer buys a new PC, or needs the software upgraded then they can do it themselves. Supplier A loses out on the support contract, because its argued they are not needed.
Supplier B sells software and support
Supplier B charges people for the software & for support. When times are good people will be convinced to pay for both, when they are bad they cut support. Now if someone buys a new PC, or needs the software upgraded they have to pay Supplier B not only for the software, but also for the cost of installing/updateing the software.
try to make ends meet, you're a slave to money, then you die
The biggest problem with his statements is that he completely ignores why successful OSS projects are successful. If it was the bubble, then OSS would be dead. OSS was around long before and and will be around long after the "bubble". There are plenty of examples of OSS projects which are succeeding despite this "mass-delusion" -> Apache, Tomcat, Linux, MySQL, PostGRES...
...a form of collective geek madness. My favorite, a dismissal of success under the ambiguous "Those silly geeks". On the contrary, history will look back on it as a maturation of software development. OSS projects which serve useful puposes have, since the beginning of the "movement", been successful. Why? Because of they ARE useful and widely available without increased and wasteful overhead expense on "support" and "maintenance" which go largely unused...
The fallacy being spread is that OSS projects cannot, in his words organize and focus enormous human and financial resources on creating and deploying software technology sustained over the long term has been proven false over the lifespan of Linux. To say that great finances are critical also ignores aforementioned success.
A great product sells itself. Otherwise OSS really would be out the door.
Here's where the real difference lies - the strength big players is not (and does not need to be) in their coding,. MS has proven this to be true. But the rest of his statement blurs why popular OSS projects are so - they are so because they have been able to produce high-quality products. Last I checked, coding was pretty key here, as is focus and efficient utilization of resources. Can they maintain it over the long term? Linux is 10 uears old, Apache is up there, BSD...
Computer Science is Applied Philosophy
lets not confuse "open source" and "gpl".
*thousands* of companies working with bsdl code are alive and kicking, and doing well. check out apple for starters.
How many of these companies would do well to adopt a dual GPL/proprietary licencing scheme as did Sleepycat?