Slashdot Mirror
Can Open Source and Commercial Software Coexist?
morrison asks: "In recent years, the Open-Source movement has increased dramatically. Harnessing the power of thousands of developers and testers has proven successful, to varying degrees, in developing operating systems, graphics applications, and web tools, including Linux, POV-Ray, Blender, Gimp, and Apache. In a SIGGRAPH 2005 discussion panel, the questions will be raised as to whether the open-source model is relevant and useful to the graphics community. Does the model of proprietary application research, development, and usage serve the industry better? Or will commercial facilities continue to primarily choose off-the-shelf solutions? Can all models work together? As a large portion of the Slashdot and Open Source community will be at SIGGRAPH, I'd really like to hear some moderated arguments beforehand before stepping up to the microphone."
Just use rpm and force people to use version 7.1 of Redhat. That seems to be the ploy that some vendors (*cough* Cadence and Synopsis *cough*) are using. When I asked one support guy what happens if I want to use a machine that isn't four years old to run their software his response was, "Do you have a solaris box?". I felt like saying, "no, we can't afford one because we spent all our money buying your software."
I Am My Own Worst Enemy
There exist companies like MySQL AB and Trolltech who are commercial, but are "Open Source."
Consider just the Unix companies who sprang into existance in Silicon Valley. First they downloaded BSD 4.1c and developed an OS for their new hardware, then they donated the fixes back to Berkeley for 4.2. When their hardware shipped they went to Western Electric and ought a 32V license.
And this is just one example,and not even a particularly recent one...
--dave
davecb@spamcop.net
Sure they can co-exist until we "replace" them. I don't work for a software house we use software not produce it. Nearly all proprietary software is a royal pain in my butt and the faster it is obsoleted with OSS software the better.
Take for instance the other day we upgrade a piece of software then immediately run into trouble since the vendor decided to make more money so he put some sort of per page processing keys in it and changed the licensing requirements. Two days of production down time while sorting that out. I am now in the process of finding a oss alternative to his product or I will write an new oss alternative....I don't feel bad not one bit for software shops going out of business because of OSS products.
Got Code?
Open source at times forces commercial software to improve. Just look at blender, after you get past the hissy fit that lightwave users have about the interface it has surpassed lightwave in capabilities and speed. We switched all our 3d artists over to it here 1 year ago and after the massive whine-fest over the interface and controls the guys will never go back, and so far has saved us several thousand dollars annually by not having to buy 4 licenses each upgrade cycle. one of the artists has become quite a guru with perl scritping for blender and is doing some amazing things that are almost pixar quality in a free "toy" that the supposed professionals poo-poo as worthless.
they also enjoy using gimp and find it easier to make tileable textures in gimp than in CS... and the biggest thing the guys like is being able to take copies of their software home and use it there. something that is 100% impossible with lightwave and photoshop CS.
They still have the older versions on their machines of the legacy lightwave and CS, but they use them both less and less.
I cant wait to see how the apps continue to shape out over the next 5-10 years... open source never has to add worthless features to entice users to buy the software yet again as is the requirement with commercial software... and that is how it can get better in the long run.
Do not look at laser with remaining good eye.
I think this whole "one or the other" type of argument is a red herring.
The software industry naturally tends towards 'killer aps' which is a nice way of saying that it tends towards monopoly, even more so than traditional industry does. Being able to design a program once, and then produce millions of copies for profit is just too big an incentive to consolidate. It also makes for an incredible economy of scale, so that small companies have a hard time competing.
Look at Adobe's recent purchase of Macromedia to see which way the industry is going.
It's hard for a startup to compete with a readily established killer ap. Take Photoshop, for instance. If someone said "I'm going to start up a company that tries to do what photoshop does" I wouldn't want to invest my money there, unless it pandered to a special niche market - maybe designing 3D skins.
Competition is vital to keep the cost of proprietary systems reasonable. Also, monopolies have a bad history of abusing their customers.
Because OSS software is the most reliable model for giving sustained competition to programs which would otherwise come to monopolize their industry, open source is a vital suppliment to closed source software. It can't be bought out in the same way a private company can, and its low cost puts some downward pressure on the price of closed source systems.
___
It's the end of my comment as I know it and I feel fine.
It seems to me that companies will continue to use what works best. Sometimes OSS will fit the bill, while other times commercial software will best serve the need.
Off the shelf solutions offer many advantages that OSS will find difficult to duplicate. An example of this is technical support. I am well aware that the vast majority of OSS projects have a large and acitve community that is capable of helping with many issues that may arise. However, this is not something that a project manager can look at and assign a cost to. With commercial software, you get commercial support. RadHat has shown that a successful business can be formed around open source software, but I don't know if this business model will form around other OSS projects.
Where OSS shines is research and development type work. The large code base created by OSS projects are useful to people who just want to try something out. I don't want to pay money to fool around with speech recognition in my new app, so ViaVoice or Dragon Naturally Speaking SDKs are out of the question. However, the CMU Sphinx project offers a speech recognition system that I can play with for free.
If the shoe fits, wear it.
My take is that open source software is great at implementing things that are already commodities. Web servers became a commodity while Apache was being written. HTML rendering was becoming commonplace when Netscape decided to open Gecko. Same thing with MP3 coding and LAME and a hundred other examples I can't think of right now.
Proprietary stuff, on the other hand, tends to be newer and more cutting-edge than open source stuff. Photoshop vs. GIMP, Microsoft Office vs. OpenOffice. Nero vs. x264 coding. The Mac's UI vs. Gnome/KDE. Where commercial software leads the way, open source follows (and usually does so exceedingly well).
Commercial software, then, is where innovation happens. Open source software excels at development and commoditization. They not only exist, but complement each other.
Can Open Source and Commercial Software Coexist?
What a horrible question. Reminds me of another one:
Is Duke Nukem Forever going to come out?
Everyone's got an opinion on it, and yet you expect reasonable responses are going to be the ones modded up? No, instead, you'll get about four "+5, Funny" comments and maybe one "+5, Informative" with some scattered "+5, Interesting"'s that are really about different topics, like how cool BSD's license is and some classroom examples where no one knew what open source software was anyway.
If you want legitimate discussion, ask the question in a context. Like this: "Can Commercial Software roadblocks still allow Open Source developers to provide sufficient products in the near and long term?"
Another good one: "Is Open Source development keying in on certain specific applications (Apache, PHP, MySQL), causing stagnation in development of other equally-important and every-growing more challenging OS softwares (Samba, Wine, PERL)?"
the idea is that GAIM and GIMP can coexist and use the same GTK package, and these days the idea works well in my experience, you just need to make sure that you use the most up to date GTK installer, be it from GIMP-win or from GAIM-win32.
in theory GIMP 2.2.8 will run with as old a GTK version as 2.4.14, though i'm not sure if that's the case or not, it tends to be GAIM that breaks if you don't use their GTK version.
Seriously though, if you're not totally braindead you can probably install GAIM and GIMP and keep them running harmoniously. Inkscape and Gnumeric of course have their own copies of GTK in the app directories, but that does make for heavy downloads and DLL bloat.
And, as it happens, we make graphics software. We're a small company with a product that has considerably more features, and more power, than either the Gimp or Photoshop, and we do very well with it. There is no problem (for us) having the Gimp, at no cost, and Photoshop, at relatively high cost, marketing to the same group of people. With a moderate price model, we can (and do) convince owners of other products to give ours a shot without any particular problem.
The only trouble we've had is when we set our prices too low -- below $99.00, no one will take the product seriously. We've tried multiple times to set it lower, as we're well down the ROI curve, but it just won't sell below $99.95. We did find a workaround, though... we have an offer where we'll give it to you "at a discount" (for $49.95) if you say you have a Corel, JASC or Adobe product. We really don't care if you do or not; we don't even check. :-) But people will buy that even though they won't buy it if we actually price it at $49.95. The lesson? People are funny.
Aside from the in-your-face issue of price, commercial development, large or small, by its very nature brings something else to the table that open source doesn't, and that is a constant drive to work on the product without distraction or interruption. It does this by virtue of funding the development. This ensures that the developers can be secure in the knowledge they can go home at night and get the cat fed, pay the XM bill, and so on. They don't have to work at night (though of course they can, and if the company is smart they'll reward such behavior.) They can have a rich social life. Still, they get to spend many hours a day pushing pixels, and as a graphics developer, I can tell that is a significant pleasure.
All in all, I see no reason for commercial graphics development to be concerned about open source. Certainly there is no reason for open source to be concerned about commercial graphics development, per se.
Frankly the risk/danger (to everyone) is not other developers. The danger is software patents. The danger comes from the legislature. You can -- without ever intending to -- run afoul of someone's invention and be in a world of financial hurt as you try to defend yourself and protect the time and energy (and money, if you're commercial) you've put into your legitimate development, and the legal system can crash your progress as sure as if they were the on-coming train in the tunnel. In my opinion, that is the problem that needs addressing, and that is what will cause the most disruption(s) to any project, be it commercial or open source.
I've fallen off your lawn, and I can't get up.
Can open source software compete against commercial software? Depends on what type of software.
Gaming software strikes me as one market that will never be anything but commercial. Most of the fun of games would be ruined if you had helped design them: knowing all the twists and turns in the plot, exactly how to beat every boss, and knowing how the story ends before you take your first step ina run-though.
When designing games, you're entertaining others but not yourself-- something you'll certainly always want to be paid for. Open source OSes and the like is giving increased usability and productivity to everyone, yourself included.
So while I see promise for the viability of open-source software competing against commercial software in things like instant messaging clients, I just wanted to point out that I don't think it has a chance in gaming software.
Not if the Open Source community has anything to say about it...
Don't get me wrong. I like open source software and I try to promote it as much as I can. But there are things the open source community does that I feel is basically shooting themselves in the foot.
I think my biggest pet pieve with open source software is not the concept itself, but how OS zealots treat proprietary software. Getting to my point, take for example Linux. (I think this is where my karma goes to hell...) In particular, how Linux treats loadable kernel modules. If you load a kernel module that does not explicity state that it is GPLed you start to see messages in the log like "AHH! Proprietary software! The kernel is TAINTED! I'm melting!!!!"
Everyone complains that hardware vendor X doesn't support Linux. But then if they finally build support, and don't open their IP up to the world, they get bitch slapped for it. "Ah! you're tainting my kernel!" You know, I haven't seen whiners like that since elementary school.
Yeah, running in kernel context grants 3rd party software access that can potentially change how the kernel functions. However, most of the time the LKMs simply add the necessary support for the hardware and leave everything else untouched. The only noticable difference is that the hardware _actually_ works. (which is also sometimes debatable...)
I know, technically, according to the verbage of the GPL, all LKMs should be GPLed. I really feel that is too restricting if you want the support from 3rd party vendors. Which is why I'm guessing, that non-GPLed LKMs are still able to be loaded. But if you want the support, quit whining about it!
My question is, how long is it going to be until necessary kernel symbols are no longer exported to proprietary LKMs? When is the final bullet going to be fired into your foot where no commercial company is going to be able (or rather willing) to support Linux at all?
I build drivers for both Linux and Windows. I have YET to see Windows complain about whether or not my source code was GPLed or proprietary!
If you want open source and commercial software to coexist, we really need to get with the program here!
Ok, rant mode off... goodbye sweet karma! it was nice knowing you...
Ascalante: Your bride is over 3,000 years old.
Kull: She told me she was 19!
The Open Source movement is just another source of competition for software developers. Just because this software is "given away" doesn't mean it turns commercial software into a commodity item. For the lower value items such as a basic text editor, I can see how a commercial developer may freak out because an Open Source text editor alternative may be viewed as a viable replacement to the commercial version. When you get to the higher priced items though, you (generally) get what you pay for. For example, people are still shelling out hundreds and hundreds of dollars for Photoshop, FinalCut Pro, other specialized apps that require some real programming know-how. Is there an alternative to these apps? Possibly. But it's the commercial developers that went through the painstaking process to write the code, make sure it fulfilled specific requirements, and delivered it to the marketplace giving end users a quality product and a sense of security that this product will work on their machine. Maybe there are some Open Source alternatives to these high priced items, but until they work as flawlessly and seemlessly as the commercial versions, the commercial developers have nothing to worry about. Until then, they just need to keep looking over their shoulders and making sure they are staying ahead of all the developments the Open Source projects are making. It's a healthy form of competition that ultimately gives us all better products and more choices to choose from.
"He uses statistics as a drunken man uses lampposts...for support rather than illumination." - Andrew Lang
If you want commercial drivers in your kernel, you should be running an OS that doesn't apply the GPL to the kernel. If you want to write drivers for an open-source kernel, and you don't want to GPL the drivers, then you should write them for an OS that doesn't apply the GPL to the kernel.
What the world needs is an open source emulator that lets you run non-Linux drivers (BSD, Solaris, even Windows) in the Linux kernel. Since the interface you're emulating is not GPLed, your drivers won't need to be GPLed, even if the shim itself is GPLed.
Yes, I know this would be horribly difficult, but it only needs to be done once.
I am an amature web designer, I am self taught and have been doing more and more commercial work on the side, because people are wanting to use the cheap linux web space out there and need people familiar with software like Drupal.
Recently I have been working a bit with a graphics designer that uses a Mac and Adobe to do her work. She was surprised at some of the stuff I was able to do with open source software and admitted that if the software she used was available on Linux, she would seriously consider switching.
One last thought, with some corprate backing the Mozilla SVG project would probably take the web by storm.
once more into the breach