Slashdot Mirror
How Open Source Drives Down Startup Costs
prostoalex writes "Reuters Plugged In article (usually syndicated to your local paper's Technology section) talks about the real impact of open source in the technology world -- cutting down startup costs for other developers. New ventures are coming out, where the startup costs range in five-digit numbers, not seven-digit figures, where venture capital financing would be required. The article talks about Project for Open Source Media, Blogger.com, Odeo and Asterisk telephone system."
You should. Even starting a business that fails ultimately is a great learning experience and makes you *way* more attractive to employers. Being in school is a great time as well. Look for something that won't take *too* much time away, that doesn't have much overhead, and could let you pocket some decent spending money.
Is that open-source usually works well with cheap commodity machines, like the ubiquitous PC.
True story: I once did an interview with a very interesting start-up, who designed custom chips for high-speed routers.
During that interview, one of the founders of the company mentioned they were moving all their engineers from Sun machines to 4 CPUs Intel machines running Linux.
He said Linux was already good enough to do 90% of the job, for less than 10% of the cost of a Sun machine. The move, of course, saved '000s of dollars for the company.
The right to offend is far more important than the right not to be offended. (Rowan Atkinson)
Except of course, the opportunity cost of your time. I am working on two different business plans at this moment that have NO startup costs because of open source software. Granted, they are both development projects, however they are both quite viable businesses, and all they require is the spare time of a few dedicated developers after they get home from their "Real" jobs.
Now that a useful machine is less than a thousand dollars, it seems much harder to get training, conferences or other ancillary spending approved.
The expensive or difficult thing, item #0 is the IDEA, the thing that makes the startup actually worth something...
Phaw. Ideas are cheap. There are a million of them, and a good, healthy percentage of them can be quite profitable.
It's the combination of idea, product delivery, legal stability, financial competence, and (most especially) marketing that makes a startup fly.
I remember reading a while back about the "card table" test. The idea goes something like this:
When looking at a startup to invest in, visit their main offices. If they have nice, leather seats and elaborate furniture, take your investment capital elsewhere. They aren't prioritizing their investments on delivery.
On the other hand, if they are using cheap, Costco furniture and/or card tables, they are putting their money where it matters, and are much more likely to succeed.
I have no problem with your religion until you decide it's reason to deprive others of the truth.
The startup costs are one thing; but time-to-market is another huge factor. And what the article missed was the time-to-market factor.
I've done a number of successful embedded projects, and you just can't beat the time-to-market involved with Open Source. I can beat any closed source project hands down if you're talking about new hardware.
What I commonly see is that something unforeseen will arise. With commercial closed-source solutions, I'd be stuck waiting on the Vendor to provide a solution. Often that same solution either already exists, or is easy to implement, in an Open Source implementation.
Recently I saw this on a new motherboard. The ROM BIOS guys (at the mobo company, and at the well-known ROM BIOS company) had lots of problems when we were bringing up new hardware. So we just grabbed LinuxBIOS, and we could diagnose the problem quicker than they could. But I've seen this theme time and time again.
When someone tries to recruit me, I write them off now if they are using something like VxWorks. I really view it as a red flag that they don't know what they are doing.
The bottom line is that I need solutions, not problems. And Open Source either solves it immediately, or allows me to solve it faster than Closed Source. That's why Open Source products are now becoming prevalent.
Closed source software tends to get the "crown jewel" treatment. It starts of with a high value - often being a company's strategic advantage. But, because it is isolated and cost a lot to originally develop it tends to stagnate. Pretty soon your cutting-edge best-in-the-world software falls behind and the company hurts.
Open source software, on the other hand, tends to stay fresher. Because more people are involved, the boundaries are being pushed a lot harder instead of being hampered by internal corporate politics.
Engineering is the art of compromise.
My understanding of the lofty side of the Open Source Movement is a little more esoteric. I may be missing the point. Is the real upside to 'free software' all about the money? I thought the point was the transparency of the information handling method. The obvious benefit is the access to other designers conceptual products. This allows for integration from various sources to facilitate the optimization of each system component. It is the basis of the 'standing on the shoulders' and the ever popular 'don't reinvent the wheel' philosophies. Even more important is the publics access to the code. In an open information society it is fundamental to be able to determine the source, path, and handling methods for data delivery. It's the only way to keep the rascals honest. I love my computer, but I don't want to just take its word that some Bush descendent has once again pulled out a squeaker in Florida. I've developed a sneaky suspicion that it harbors a Republican kernel. I've always thought of Open Source sort of as 'The Great Equalizer' which would finally break through the bullshit barrier of the powers that be. So am I giving the OS evangelists too much credit? Maybe most of them DO just like free stuff.
billy - say it ain't so Linus
The GPL is not evil, it is just something else to watch for. However if you are careful you can manage the GPL just fine. Just make sure you can send out the source code when someone asks. (Which doesn't happen often, most customers won't care) Then make sure that you keep GPL stuff separate from your stuff. We ship a GPL pdf2txt program where I work, not a problem, we just call it as an external program and read the result.
In some cases we will even release source code. pdf is not our core ability, so if we find a bug in pdf2txt we are likely to send it in so everyone can use our fix. We won't let you see the parts where we have added value, but those parts are carefully not GPL, and they are what you pay for. (Though admitidly you could do everything we do yourself given a year)
The above assumes that you are a developer trying to sell a product. In many cases you are selling your services to setup GPL software to someone else. In that case you don't care about IP and GPL because anything you write is paid for already by the customer.
As I've been researching opening my own small business, the cost of implementing a Linux solution is significantly lower than implementing a Windows solution. First, no CAL's to deal with, I can perform multiple installs of the OS, and the Opensource versions of the applications I need off all the functionality and reliablity to accomplish my mission. In laymens terms, it was a "no-brainer"
They came for the Communists, and I didn't object - For I wasn't a Communist; They came for the Socialists, and I didn'
right now and open source has saved us untold thousands. We needed a real phone system so that we would have voice mail and "sound" bigger than we were: Asterisk to the rescue. For a couple of $39 cards and an old pc we've got a pbx. Instead of using Oracle or Sql Server we're going with mysql and php. We need to print bar codes. $800 active X library? Nope, a php library I found on the web for zero.
However, I think it really depends on the kind of company you're going to start. In our case, we're developing software for our own use. I can't imagine trying to bundle some of these pieces and sell them. Surely it would work with some things but, imho, no the majority of OSS projects. I don't say this to belittle the efforts of those who have done so much for me. The reason I say it is because the mindset of the vast majority of users runs completely counter to that necessary to effectively utilize much of the open source software available.
Take Asterisk for example. I had to reboot my phone 4 times on Friday to fix various problems. I'm not crying, just pointing out that the first time you tell a user to reboot his/her phone, they are going to look at you like you have 2 heads.
Me and a bunch of friends started up an ISP on nothing more than a few Slackware Linux 100MHz Linux boxes and about 16 14.4K modems. Our starting budget was about a few grand. Talk about shoe-string. We had no employees, completely run by the owners, doing work mostly for free on a promise of getting a big pay-off in the future. Later, after about 4 years and around 1000 customers, we sold the whole operation to a larger ISP. Although, we didn't get that much for our efforts... live and learn.
Meh.
Your "IP" is not threatened. If you use GPL code in your program and then sell it without releasing the source, you are guilty of copyright violation. There is no precedence for somebody being guilty of copyright violation having to give up the rights to their own work.
What will happen is you will be required to cease distribution and pay a monetary fine, if anybody catches you and sues you. If you wish to redistribute the work you will have to remove the GPL portions and replace them with other code that you do have a license to redistribute.
It is true that many people are happy if you release your source code, and that could be used to avoid a lawsuit. But there is no requirement to do that, and it does not even get you out of being liable for the copyright violation (otherwise you could just release the source code after it is worthless due to age).
Furthermore, there is very little money-making opportunities on Linux that require you to use GPL code, in fact it is difficult to figure out something that could make money but requires GPL code, perhaps heavily modified kernels that can't be done with loadable modules?. Linux allows any program to run on it, and the vast majority of the libraries needed to make a working program are LGPL or BSD or similar (due to the fact that the library writers want their stuff used as much as possible).
ClarisWorks was developed with a four-figure startup cost, starting in 1989, and became a top-selling product with millions of users. It's true that that success was not achieved until the program was sold to Claris, and additional development resources were added.
But from the point of view of the original developers - myself and Scott Holdaway - our startup costs were very small. We bought two computers, rented a house together, and hacked. Details here:
http://www.swiss.ai.mit.edu/~bob/clarisworks.php
I should add that from my personal point of view, the open source meme has made it much harder to figure out how to make a buck selling software. In the old days it was simple. OK, call me clueless. I gave up and went back to school.
That's what I think every time I see a post about "OSS as a business or used in business."
But I get flamed for saying so...
Is Capitalism Good for the Poor?
Exactly the same with closed software. You can buy compilers that have, deep in the 5th layer of the EULA, a clause that states you don't own the software you produce, or can't distribute it, or have to pay patent licensing fees or something.
I guarentee you that extending GPL code is no different than extending microsoft's code. You have to contact the copyright holder (e.g. trolltech in the case of QT, Microsoft in the case of whatever they make) and they decide if they want to sell you a license or not.