That's a very interesting perspective. It seems to me that there are two "camps" regarding how the problem of software patents should be handled.
The GPL/BSD camp holds that software patents should be ignored unless a threat emerges. Their reasoning follows: All significant software unknowingly infringes on many patents, regardless of how trivial. Nevertheless, most patents wouldn't hold up in court or could be invalidated by prior art. It isn't worth developers' time and it is in fact dangerous to go looking for possibly applicable patents when writing software. Companies who support software patents realize that a direct challenge to a major open source project could end up going all the way to the Supreme Court and result in software patents being invalided altogether. Holding our own patents in defense would be hypocritical, costly, and may even undermine our arguments against them. Even if we did hold patents for defense, there are so many out there that any patent-grant-covered code/project would still be vulnerable to patents we don't hold. For the last 15-20 years or so, this approach has appearently worked flawlessly, weathering many waves of legal FUD.
The MPL/CDDL/Apache(?) camp holds that software patents are a much bigger threat than commonly perceived and that we must assume they will be around for awhile. Attacks are inevitable as proprietary vendors resist the continued success of OSS. Corporate users are presumed to be leary about patent issues with OSS and thus we need to create protective measures for them. The best defense is a strong offense. We should thus join into the software patent "cold war" by building up our own "arsenals" to deter a real war from actually starting. If projects do not produce their own defensive patents, they should at least use licenses that allow/require contributors to do so, effectively creating a 'patent commons' safety net. Once the stakes are high enough, everyone involved should realize how ridiculous the situation is and collaborate to mutually disarm by lobbying legislators to invalidate software patents altogether. Until that time, we can also use our patent arsenal to enforce compliance with our open specifications and protocols so that there are no proprietary, incompatible forks.
Maybe that's a bit of a simplification, but I think this closely sums up both camps from what I've heard/read over the last few years. I just think it's a shame that it has come to this. Neither camp is a significant majority, so there's little hope of either ideology winning out through dominance. Somehow, both camps need to meet somewhere in the middle. Unfortunately, nobody has figured out how to make this happen yet. This is all the more reason to push hard for patent reform!
But I do have one idea. Suppose that the GPL was modified to include one or both of the following patent provisions:
- "your rights to use software licensed under GPL v3 shall terminate immediately should you engage in a patent lawsuit against any software project covered by an OSI or FSF approved open source license"
- "by distributing, in original or modified form, software covered by this license, you implicitly grant a worldwide, non-exclusive, royalty-free right to use any patents you may hold which are necessary to implement features in this software, or similar features in any software covered by an OSI or FSF approved open source license. **"
** This could also be narrowed to only those licenses which require modifications to be given back to the community upon re-distribution.
Yes, obviously that would all need to be translated into proper legalese.:) But it seems possible to create an "anti-patent commons" if you will, negating the need to build up patent arsenals while still providing extensive protections for both users and developers. If everyone is really that serious about getting rid of software patents, a license with these provisions would seem to be the most potent way to help this happen.
I read your post and blog entry and I understand where you're coming from. Sun does not appear to be in a position to use GPL for OpenSolaris, even if it wanted to. You're right that MPL-class licenses are ideal in this situation. However, I think some of the controversy could be alleviated by using a selective multiple license policy. For example, the Mozilla project now requires that all new source files added to their CVS tree be tri-licensed as MPL/GPL/LGPL to maximize cross-pollenation with other related projects. They are in the process of obtaining permission from authors to re-license existing code under this tri-license, but this does not hinder Mozilla from being distributed today under MPL. I see this as a best of both worlds approach.
So here's my question: Have you folks at Sun considered using a similar approach with OpenSolaris? License everything under CDDL, but gradually dual/tri license certain components under a GPL compatible license as well. (preferably as much as you legally can) The Open Source community has historically welcomed this approach and I think it would boost contribution dramatically.
Regardless of license choice, there is one big fly in the ointment: software patents. Lets just put it on the table.. the primary reason many members of the Open Source community have a distrust of Sun is because of this issue. From Java to the secret settlement terms with MS, there is perhaps legitimate concern that Sun would someday turn on the OSS community -- particularly its competitors (Linux, GNU Classpath, Apache/Jakarta, Mono, etc.) Maybe the current leadership would never even think of such a thing, but leadership can change. Or what if Sun was bought out by a company hostile towards Open Source? These are unknowns, and unknowns breed fear. When Sun says, "here are 500 patents that we're turning loose.. oh.. but.. you can only use them with our CDDL licensed software," that doesn't jive with community ethos. It implies an attempt to maintain a style of proprietary control long deemed unacceptable not just by GPL zealots but by anyone aware of the potential legal risks to developers.
I think it's admirable that Sun has been increasingly moving in the direction of Open Source, but I caution that there are still some barriers. Be careful not to label these as just silly ideological nuances. Some of them are, but others, like software patents, are not. Until the menace of software patenting in the US is destroyed, Sun needs to assure the OSS community that it will have no part whatsoever in any offensive use of its patent portfolio.
See this page for details on the Mozilla policy: http://www.mozilla.org/MPL/license-policy .html
This is just what we need, even less fuel efficient modes of transportation.
According to the article, this AirScooter runs for 2 hours on 5 gallons of gasoline. Top speed is listed as 55 knots, or about 63mph. Lets say the average speed you achieve is only 55mph, taking into account takeoff, landing, accel/decel. 55 mi/hr / 2.5 gal/hr = 22 mpg. So the mileage is already better than most SUVs, though not spectacular. But then you take into account the fact that you'll probably save quite a bit of distance by flying in straight lines instead of driving on winding roads. (well, depends on what part of the country..) If the distance saved is say.. 30%, then you're up to about 31 mpg at an equivalent of driving 78mph. Not too shabby for commuting.
"Flying cars" are indeed still feasible. They just don't look like "flying cars." What would be really interesting is some sort of VTOL capable ultralight glider with electric props. (Gliders, of course, being a bit more efficient than helicopters) Once the puzzle of weight-efficient battery technology has been solved, I expect pretty near anything to be possible in this arena. Of course, land vehicles will likewise always remain far more efficient.
Paints under fluorescent lighting will be muchg duller than under daylight. Most people don't care though. They only get depressed after a while, and don't know why.
There are different phosphor compositions available in fluorescent lighting. The oldest produced only 2 or 3 discrete frequencies and were terribly "harsh" looking. Newer fluorescents, especially CFL's, typically have around 5-7, from what I can tell using a cheap diffusion grating spectrometer. You can buy special "daylight" fluorescents with a spectral composition that better approximates sunlight, but I don't think these are actually continuous spectrum.. they just have a spectral distribution closer to sunlight.
White LEDs, unlike fluorescents, are indeed continuous spectrum light sources, as verified by spectrometer. As with any source, there are still stronger and weaker portions of the spectrum.
The whole "full spectrum" lighting as a cure to depression, irritability, skin cancer, your poor stock returns, etc. appears to be a scam as there is no scientific evidence to this date to support the claims being made. Seasonal depression is proven treatable using any bright light source, whether "full spectrum" or not.
Unlike other posters, I'll give you the benefit of the doubt that when you say "Linux" you really mean "Open Source desktop software distributions that may use Linux at their base" I often use the same sloppy terminology, so who cares.
2. Mac OS X and Windows revolve around the interface. On the library level, new interfaces are added, but older ones are still supported for a surprisingly long time (see Carbon / Classic Runtime Environment for Mac OS X, or Win9x Compatibility Mode / Application Compatibility Toolkit for Windows 2000 / XP). Certainly, support is eventually dropped, but the pace is normally quite slow for popular APIs.
If you develop with cross-platform libraries, as you should unless you have a really special need, then all desktop platforms "revolve around the interface." In fact, the added layer of abstraction allows you to leverage the same consistency to all platforms. Qt, Gtk2, and Java/SWT are some good choices.
Now, if you're talking about desktop environment GUI consistency, well.. there is simply no such thing. But do note that the gradual evolution of KDE over the last 10 years has been no more dramatic than Windows95 through XP.
And of course, looking towards the future, the web is the new platform of consistency. "Linux" will not likely beat Windows by replacing it as a desktop platform. The desktop platform will just become increasingly irrelevant and Linux will be left as the defacto choice simply because it is the cheapest way to get cheap hardware on the web reliably. Cross-platform libraries will take care of all the rest -- whether Java or.NET/Mono or Qt or Gtk2 or any combination thereof.
In short, I see all of this as signs that Linux is moving too fast for people to consolidate and work out standards.
On the other hand, standards are emerging nonetheless -- by the fittest surviving. You might see Open Source has having been through a chaotic rush to catch "unix" up to the status quo in desktop software. It's not over yet, but it's showing signs of slowing. The other thing to consider is that some standards are more important to consolidate on than others. This leads me to my next point below...
We're in a transition period where people are held back by old systems, but seek cross-platform compatibility on new systems. It's like watching pressure build for a switchover. Just because there's been no mass exodus you still see them untangle themselves from Windows strangleholds.
You've just hit on a crucial point that the original article missed. "People are untangling themselves from the Windows stranglehold." Or another way to put this: desktop platforms don't matter anymore! Look at what people (companies) are doing to untangle themselves from Windows. They're replacing traditional Win32 client apps with web applications, local or hosted. They're replacing Exchange/Outlook with 3rd party mail software, web based groupware, mail server "applicances", outsourced mail services, etc. Each step along the way breaks away from the very need to have a consistent desktop platform in the first place. The web is the new platform of consistency. It is driven by true standards -- open standards, like the work of W3C. And it matters not what hardware or operating system you run. (Though Linux or BSD will get you on the web for cheaper, both on the server and client side)
It's curious that the article ends with a claim that MacOSX and Java are the only "platform" that can compete with Windows. In fact, Java is a very important piece of the "web" ecosystem -- but it is fully cross platform and runs no better under MacOSX than under Linux. MacOSX is "unix" just as much as Linux and BSD are. The only reason to use MacOSX as a platform is for its really slick graphics / GUI toolkit libraries. And those are only needed for specialized applications not yet suited for the web.
The pressure is indeed building and the whole paradigm of how we use computers is changing. With the exception of a possible surge as OpenOffice.org* reaches "acceptability" I don't expect to see a mass exodus. I expect people to wake up one day and say, "hey.. what ever happened to that 'Windows' thing that used to be everywhere?"
* I see OpenOffice.org as a transitionary crutch. The word processing / office suite paradigm itself is quite dated and will not survive the full onset of the web as a true platform.
I'm preaching to the choir here but reverse engineering is a Good Thing for all communities. There is absolutely no reason that we should not support working around what others have obfusticated to make money for themselves.
I completely agree. RE is an indispensible tool.
On the other hand, I think we should be cautious not to extrapolate Linus's views inappropriately. The inference that he is completely against RE in OSS is baseless. The simple fact is that he had a conflict of interest over BitKeeper and was forced to be political about the whole situation. It is quite possible that his statements were intended to de-fuse McVoy and/or others in attempt to mitigate any impending bad-mouthing of the community he helps to shepherd. I do not think Linus was trying to make any sort of 'moral statement' whatsoever. He even admitted that it was only a matter of time before the Linux community would move away from BK to something open source.
People mostly do things for their own benefit, as they should. I don't think it's good to encourage decent people to sacrifice themselves for the benefit of people who give nothing back. That just leeches the resources of decent, generous people and gives more power to the other sort.
Which is why most of the good Open Source software out there is commercialized in one form or fashion.. Of course, there are other projects that thrive on the sheer fun of doing something really new and interesting.. like Blender. But most projects should be heavily commercialized. I personally am a big fan of direct user feedback through pledging to support desired feature development -- at least for smaller projects.
What you describe is certainly the ideal. However, only larger companies typically set up that much security infrastructure. They can afford to do it right, manually, because they have enough highly trained IT staff. Smaller organizations typically go to far fewer extremes. And, after all, they are much smaller / less interesting targets. So for them, the effort is better invested in an out-of-the-box firewall solution and having money left over for improving physical security.
Webmin is a total joke -- well, not necessarily Webmin itself, but the quality of the modules available. Furthermore, nothing is integratable. On most commercial firewall solutions, changes to one component make proper changes to others. (Ex. turning on a transparent HTTP proxy would modify firewall rules for you after asking) I'm not saying this is an absolute necessity, but most people who install these commercial firewall don't have time to figure everything out for themselves. Webmin is an admin framework for people who are already *nix experts and just want a convenient non-SSH way to do remote admin. Then again, for such people, it's usually faster to just use SSH, so the existance of Webmin is hardly more than a "checklist" item. (ie. "yeah, of course we've got a web admin console..")
Astaro isn't free. By the time you build a machine and get an Astaro license, it costs just as much as any of the pre-built "security appliance" devices. Also, last time I did a comparison, it's not even very competitive with regards to features.
Almost all of the commercial firewall solutions are already based on Linux or BSD. The point is to create a free (and superior) replacement for the likes of Astaro. (and, if you will.. the "distros" that come with security appliances) I am 100% against proprietary Linux/BSD distributions. There's no need for them. The only reason they exist is that the Open Source community has dropped the ball just short of the goal. The advantage of free distros is you have full control and there's no reliance on any company for updates.
One particular scenario was a firewall. I suggested a Linux firewall due to the lower upfront cost. Now, there were a Microsoft shop, but a firewall is not something that has to be administered everyday (when it is working properly). Instead they decided to go with a Checkpoint firewall that cost them a hell of a lot more.
Here's the problem: A firewall today is not just about "Linux kernel + iptables." Those Checkpoint boxes (and others) are full featured "security appliances" as they call them. They have features such as:
- application proxies to filter / virus-scan / monitor content: HTTP, FTP, SMTP, etc. (so you can say.. stop employees in the sales dept. from playing games on Pogo during 9-5.. to give an example of the granularity of control available)
- network monitors and various intrusion detection / prevention methods
- complete mobile VPN services, including dynamic firewalling rules
- user authentication services (used for VPN, proxies, replication to other network services, etc.)
- very complete GUI admin tool / management console. (and multiple security devices can be linked together throughout the company..)
Can you do that all with free Linux distros and available OSS tools.. Mostly. Will you be saving any money by the time you've got all the raw materials kludged together into a working solution? Nope.
The Open Source community has failed miserably at producing real-world solutions. It has produced an enourmous amount of quality raw material. (And if you examine the commercial firewall solutions, you'll find much OSS being used internally!)
I think there is a good solution to this: The major free Linux/BSD distros need to have subprojects focused on specific needs. For example, there should be a "Debian/Firewall" sub-distro. (note: not a fork) It should provide a more or less ready-out-of-the-box firewall solution using pre-integrated "best of breed" components from the base Debian distro. If there are shortcomings discovered, the improvements can be fed back into the base distro using standard processes. If there are flaws found in the raw materials, this is a perfect way to make sure that OSS meets real world needs through user feedback.
Now apply this principle to all major areas of network services.. mail servers, file servers, web servers, etc. As long as there is a decent web based admin interface, there will be no problem getting organizations full of Windows-only IT staff to use more OSS. (And meanwhile all the old-school Unix folks are squirming in their seats.. Sorry folks, I don't like it either, but sometimes pragmatism is required. There simply aren't enough smart Unix people to go around. So we either compromise or we let proprietary software continue to dominate the industry.)
Yeah.. try that one on your boss.. "Sorry, you didn't get that extremely important email, but it's more important that I continue my ineffective vigiliante attack on spam by the overzealous mail server configuration I've chosen."
The issue is that you're only obligated to give the source to the people to whom you give/sell binaries. If you give the binaries to five customers, you have to release the source to those five customers.
Technically, if you do not disclose source code, you must provide, in writing, an offer valid for 3 years to disclose source code to those customers upon demand. The cost may not exceed the physical distribution cost of the source code.
As others have pointed out, the customers receiving the binaries and source are free to redistribute them, and probably cannot be constrained from doing so by any non-disclosure agreement..
Correct. NDA's cannot legally add limitations to the distribution of GPL-covered software. You may enter into an NDA with customers such that you agree to not distribute the modified software to any other party but your customers. But you cannot enter into any NDA that limits what your customers can do with the software once they've received it. And an NDA cannot trump your customers' rights to demand a copy of the source. Furthermore, if a customer was to redistribute a binary, the party receiving it would have the right to demand source from your customer. To meet that demand, the customer would be obligated to demand the source from you.
So when you get down to it, the GPL guards pretty well against this sort of thing by ensuring that any party involved can legally spoil this unethical business model. What would be really interesting is if an employee who was not under NDA with his employer disclosed the modified GPL software. In this case it would come down to whether the modified software is covered by trade secret law.
Is Mozilla actually more secure? Or is it just as bad as any other piece of software?
In terms of design decisions, you might easily say that Mozilla is more secure than IE. (not being integrated with OS and all..) In terms of coding bugs, Mozilla is no different than any other super complex piece of software. But there's another way to look at it. Because the Mozilla code is open, we might expect an ugly rash of bugs to be found near the beginning of its rise to popularity. But we might also expect this to rapidly taper off as all the major bugs are found and squashed. So you might say that now is a relatively dangerous time to use Mozilla (instead of say.. Konqueror or Safari). But, on the other hand, it's still not quite popular enough to attract the volume of real-world attacks that IE has received. Honestly, if you're some jerk running a malicious website, are you going to target this quirky bug in Mozilla or the myriad of IE exploits that are sure to pay off?
What does bother me is that the Mozilla folks haven't taken automated updates seriously enough. I cringe to think of how many Firefox early adopters have no clue what that little red arrow at the top of their screen is. Or if they do, how many dial-up users will be patient enough to wait for the update to download.. which isn't really an update at all but a full copy of the latest version.
If you are good, you don't need the labels anymore
Herein lies the problem. There are plenty of unsigned bands willing to try new things to get noticed -- including internet-centric schemes. But the problem is that most of them aren't very good. Or if they are good, they're not different enough from existing mainstream hits. When you get down to it, instrumentalists are a dime a dozen. I have several musician friends who can perfectly imitate and they're not even career musicians. (note: It's even easier today with all the fancy DSP gear available.. turn the knob to "Hendrix" and your cheap solid state amp all of a sudden sounds like a custom-rigged tube rack.) Hence we have the "starving artist" syndrome -- the hundreds of thousands of talented musicians etc. who will never make it but keep waiting for their "big break" as they continue playing at local venues and working crappy day jobs to survive. But it's no surprise why they will never make it. They don't know how to develop their own styles. They all sound alike.. plain vanilla white-boy rock.. fomulaic, dull, uninspiring. Meanwhile, the musicians who decided to try something even slightly new were scooped up by labels before they could even realize what was happening to them. (And 90% were probably subsequently dropped and may have left music forever..)
All it will take is for ONE band with an original "breakthrough" sound to make it big online using the "free music is free advertising" model and the RIAA will soon be history. You can't stop money from talking. Very few concert venues will refuse a profitable opportunity. If the audience demands, they must provide or somebody else will. And as we all know, live performance is where the real money in music is.
After years of complaining that the RIAA and MPAA were missing the boat, and should have embraced things like Napster instead of supressing them, we got iTunes and the like.
..Except that this is not really what P2P technology was intended to accomplish. The original purpose of P2P was to enable free and efficient distribution of media content so that artists would not need to sign all their rights away to publishers in exchange for mass distribution and so that consumers could access a much larger and more diverse selection of media. Of course, before this could even be attempted, the whole idea was hi-jacked by a bunch of lamers who just wanted free crap-pop/rap major label music. (aka. original Napster, etc.) It was all massively popular because, well.. the music being shared was already popular -- thanks to radio airtime, MTV, etc. If there were any independent artists legitimately trying to make it on the old Napster, they were surely drowned out.
The fact is, iTunes and others have not really changed the music distribution landscape that significantly. They're just the new middlemen. Despite iTunes' relative popularity among other upstart online music ventures, it has made a fairly small dent in the overall market. More to the point, these new pay services do not yet offer the promise of P2P. Sure, the barrier to new artists has been lowered and that's a step in the right direction. But these services do not enable either the "sell tickets, give away music" or "give away some, sell the rest direct" business models. And why would they? That would cut them out of the game.. and there's big money in being a middleman! Of course the price of middlemen is paid by all other parties. Consumers get DRM, less-than-CD quality downloads, and higher prices. Artists lose a large percentage of their sales. The public gets an ever shrinking percentage of public domain or otherwise freely re-usable material.
If we really want to change the content industry for the better, we need to start thinking more strategically. Many good components are already out there.. Creative Commons licensing, FLAC and other lossless audio codecs, Bittorrent and copious other P2P technologies, companies like Magnatune and CD Baby, etc. But nothing combines them all into a single powerful tool!
What we really need is a true P2P service designed from the ground up to be a channel only for freely licensed content (of all types). It ought to have an indexing and searching service like the original Napster, but which only tracks legit free content and which uses hashes to validate original, unmodified content for those who want it. Then, add a user-feedback ratings and review system so that the inevitable crappy content does not waste everyone's time. Finally, make the whole thing commercially supported via artist services like online ticket and merchandise sales (linked through the index / artist database system.) All the software can be Open Source because there's no DRM involved, but don't allow contributions to the official project which add support for exterior searching. Trademark the name so that knock-off projects can't dilute the marketing. Make sure the original is so high-quality that nobody wants to fork it otherwise.
Incidentally, the whole indexing / artist services system could be run by a not-for-profit organization. Think of it as an "RIAA" for independent, freely-licensed content. Artist services could certainly be provided by outside businesses as well.
Another possibility is that I might convert them to Linux and KOffice (or whatever I end up using).
I gave KOffice a fair shot and there are definitely nice things about it (speed and memory usage for instance..) But KOffice simply doesn't have the file format support that OpenOffice does. The ugly reality is that people still need to read/write MSO document types. Save yourself some time by using OO.org for now and switching to something else later when it becomes available. (hopefully that'll be some sorta rich web app that makes word processing obsolete anyhow!:-)
Or, I'm a company with 5000+ developers and if I suddenly decided to put them all into Linux application development, I would go bankrupt.
With that many developers, you'd have quite a bit of negotiating power to lower the price. Of course you probably wouldn't be writing strictly desktop apps anyhow.
Suddenly jumping on to become completely dependent on the mercy of Trolltech is not good business sense, even if it is a fact on life in Windows world.
If you are writing proprietary apps, you are creating vendor lockin for your own customers. Pot calling the kettle black, eh?
What if Trolltech suddenly decided to go evil and put on a price tag of, say, $10000?
They'd go bankrupt. Somebody else would continue the GPL project and another would probably re-implement the Qt API's with a LGPL licensed project. (Which incidentally is always an option anyhow..)
On the other hand, it would be a complete disaster if Qt suddenly became the "default" toolkit for Linux.
Market demand would then allow Trolltech to lower their developer license prices. And the rest of us would have a superior and consistent GUI toolkit. (-:
I still maintain that the reason OpenOffice is moving forward so slowly is because it's so damn complex, and the number one priority of everybody involved with OpenOffice should be to make it simple enough that more hackers can help out.
I've seen the code and it's hideous (though slowly improving). It would probably be easier to bring the KOffice suite up to the feature level of OpenOffice.org than to completely re-write OO.org. But OO.org has momentum and corporate backing and KOffice sadly has virtually none. Incidentally, remember how long it took to turn Netscape Navigator into Mozilla Firefox?:-)
Say what you want, but a large percentage of the folks that are gutsy enough to be rolling out OpenOffice.org are doing so at least in part because they are Free Software advocates. In short, they are pushing OpenOffice.org for ethical reasons, and not for practical reasons.
Reality check: Most people who are experimenting with rolling out OO.org are doing so on Windows desktops, whether or not it's for ethical or practical reasons. Most people who are doing really huge OO.org rollouts in large companies are doing it entirely for practical reasons.
Sun is just being stupid on this front. Java has already fragmented into several mostly compatible forks. IBM has their own JVM, as does Apple, Oracle, Borland, and there are a wide range of Free Software Java-alike systems.
I agree that Sun is being stupid by not releasing more of their JVM and class libraries as Free Software. On the other hand, it is inaccurate to say that the forks of Java are "mostly incompatible." The situation is generally better than C/C++ compilers, so put it in perspective.
Freeing Java would completely kill all of the non-Sun Java toolkits, and it would give Sun the Free Software allies it needs to compete against Microsoft's.NET.
Yes. And this is even more true considering that.NET works quite well alongside Java and that MS is actively trying to woo Java developers to the Longhorn platform, with all it's proprietary but dazzling client-side APIs.
And I might have gotten OpenOffice made an acceptable alternative to MSOffice. But I'm not going to try. I won't recommend products that I don't use.
So this hypothetical company is going to continue filling MS's coffers because of your silly purism? Do you quite see how ironic this is? That's like being outnumbered in battle but shooting one of your fellow soldiers because you didn't like the joke he made.
There is almost no question that Java will be fully Open Source within the next 5 years -- either by action of Sun or "Java-like" OSS projects. There will be plenty of time for purism after the MSO monopoly tumbles.
Free Software hackers want to like Sun, and they want to like Java, but the licensing issue is a big deal to them. Unfortunately for Sun, it absolutely needs the Free Software hackers to jump on board.
Free Software hackers with any common sense and/or pragmatism like Java for jobs where it is the best tool, but they don't necessarily like Sun. The two are not really tied together -- especially considering that much of the Java software developed by the Apache/Jakarta project (and even Eclipse) is starting to run on Free JVM's and GNU/ClassPath. Does Sun need the F/OSS community? You bet.. and they know it by now. If it wasn't for OSS alternatives to some of the awful marketecture specs coming out of the JCP, Java would not likely be in the position it is today.
Heck, I develop in Java for a living (web development), and yet I still don't use a single desktop Java application outside of Eclipse and it isn't pure Java.
Same here. I don't see Java as a language well suited for lightweight desktop app development and I think it's silly that the OO.org developers went with it due to memory requirements alone. On the other hand, Mono/.NET is currently way behind Java for serious business applications. (at least on the backend / middleware..)
Yeah? What if Microsoft gobbles up Sun at some point in the future?
Then we'd all switch to IBM's JVM, silly. And everyone would start working overtime to get the various Open Source JVM's and GNU/Classpath up to where Sun's reference implementation had been. You act as if "Java" is a program rather than a language. You do realize that there are literally dozens of JVM's and Java compilers right?
Obviously the optical challenges would be pretty huge.. which is why I meant the post primarily as a joke about silly design patenting.
On the other hand, some sort of extreme wide-angle lens (fresnel?) combined with perspective correction in software might be able to sorta make it work.. at least with more distant objects behind the screen. It'd be interesting eye candy regardless..
Slashdot Mirror
That's a very interesting perspective. It seems to me that there are two "camps" regarding how the problem of software patents should be handled.
:) But it seems possible to create an "anti-patent commons" if you will, negating the need to build up patent arsenals while still providing extensive protections for both users and developers. If everyone is really that serious about getting rid of software patents, a license with these provisions would seem to be the most potent way to help this happen.
The GPL/BSD camp holds that software patents should be ignored unless a threat emerges. Their reasoning follows: All significant software unknowingly infringes on many patents, regardless of how trivial. Nevertheless, most patents wouldn't hold up in court or could be invalidated by prior art. It isn't worth developers' time and it is in fact dangerous to go looking for possibly applicable patents when writing software. Companies who support software patents realize that a direct challenge to a major open source project could end up going all the way to the Supreme Court and result in software patents being invalided altogether. Holding our own patents in defense would be hypocritical, costly, and may even undermine our arguments against them. Even if we did hold patents for defense, there are so many out there that any patent-grant-covered code/project would still be vulnerable to patents we don't hold. For the last 15-20 years or so, this approach has appearently worked flawlessly, weathering many waves of legal FUD.
The MPL/CDDL/Apache(?) camp holds that software patents are a much bigger threat than commonly perceived and that we must assume they will be around for awhile. Attacks are inevitable as proprietary vendors resist the continued success of OSS. Corporate users are presumed to be leary about patent issues with OSS and thus we need to create protective measures for them. The best defense is a strong offense. We should thus join into the software patent "cold war" by building up our own "arsenals" to deter a real war from actually starting. If projects do not produce their own defensive patents, they should at least use licenses that allow/require contributors to do so, effectively creating a 'patent commons' safety net. Once the stakes are high enough, everyone involved should realize how ridiculous the situation is and collaborate to mutually disarm by lobbying legislators to invalidate software patents altogether. Until that time, we can also use our patent arsenal to enforce compliance with our open specifications and protocols so that there are no proprietary, incompatible forks.
Maybe that's a bit of a simplification, but I think this closely sums up both camps from what I've heard/read over the last few years. I just think it's a shame that it has come to this. Neither camp is a significant majority, so there's little hope of either ideology winning out through dominance. Somehow, both camps need to meet somewhere in the middle. Unfortunately, nobody has figured out how to make this happen yet. This is all the more reason to push hard for patent reform!
But I do have one idea. Suppose that the GPL was modified to include one or both of the following patent provisions:
- "your rights to use software licensed under GPL v3 shall terminate immediately should you engage in a patent lawsuit against any software project covered by an OSI or FSF approved open source license"
- "by distributing, in original or modified form, software covered by this license, you implicitly grant a worldwide, non-exclusive, royalty-free right to use any patents you may hold which are necessary to implement features in this software, or similar features in any software covered by an OSI or FSF approved open source license. **"
** This could also be narrowed to only those licenses which require modifications to be given back to the community upon re-distribution.
Yes, obviously that would all need to be translated into proper legalese.
I read your post and blog entry and I understand where you're coming from. Sun does not appear to be in a position to use GPL for OpenSolaris, even if it wanted to. You're right that MPL-class licenses are ideal in this situation. However, I think some of the controversy could be alleviated by using a selective multiple license policy. For example, the Mozilla project now requires that all new source files added to their CVS tree be tri-licensed as MPL/GPL/LGPL to maximize cross-pollenation with other related projects. They are in the process of obtaining permission from authors to re-license existing code under this tri-license, but this does not hinder Mozilla from being distributed today under MPL. I see this as a best of both worlds approach.
y .html
So here's my question: Have you folks at Sun considered using a similar approach with OpenSolaris? License everything under CDDL, but gradually dual/tri license certain components under a GPL compatible license as well. (preferably as much as you legally can) The Open Source community has historically welcomed this approach and I think it would boost contribution dramatically.
Regardless of license choice, there is one big fly in the ointment: software patents. Lets just put it on the table.. the primary reason many members of the Open Source community have a distrust of Sun is because of this issue. From Java to the secret settlement terms with MS, there is perhaps legitimate concern that Sun would someday turn on the OSS community -- particularly its competitors (Linux, GNU Classpath, Apache/Jakarta, Mono, etc.) Maybe the current leadership would never even think of such a thing, but leadership can change. Or what if Sun was bought out by a company hostile towards Open Source? These are unknowns, and unknowns breed fear. When Sun says, "here are 500 patents that we're turning loose.. oh.. but.. you can only use them with our CDDL licensed software," that doesn't jive with community ethos. It implies an attempt to maintain a style of proprietary control long deemed unacceptable not just by GPL zealots but by anyone aware of the potential legal risks to developers.
I think it's admirable that Sun has been increasingly moving in the direction of Open Source, but I caution that there are still some barriers. Be careful not to label these as just silly ideological nuances. Some of them are, but others, like software patents, are not. Until the menace of software patenting in the US is destroyed, Sun needs to assure the OSS community that it will have no part whatsoever in any offensive use of its patent portfolio.
See this page for details on the Mozilla policy:
http://www.mozilla.org/MPL/license-polic
This is just what we need, even less fuel efficient modes of transportation.
According to the article, this AirScooter runs for 2 hours on 5 gallons of gasoline. Top speed is listed as 55 knots, or about 63mph. Lets say the average speed you achieve is only 55mph, taking into account takeoff, landing, accel/decel. 55 mi/hr / 2.5 gal/hr = 22 mpg. So the mileage is already better than most SUVs, though not spectacular. But then you take into account the fact that you'll probably save quite a bit of distance by flying in straight lines instead of driving on winding roads. (well, depends on what part of the country..) If the distance saved is say.. 30%, then you're up to about 31 mpg at an equivalent of driving 78mph. Not too shabby for commuting.
"Flying cars" are indeed still feasible. They just don't look like "flying cars." What would be really interesting is some sort of VTOL capable ultralight glider with electric props. (Gliders, of course, being a bit more efficient than helicopters) Once the puzzle of weight-efficient battery technology has been solved, I expect pretty near anything to be possible in this arena. Of course, land vehicles will likewise always remain far more efficient.
Paints under fluorescent lighting will be muchg duller than under daylight. Most people don't care though. They only get depressed after a while, and don't know why.
There are different phosphor compositions available in fluorescent lighting. The oldest produced only 2 or 3 discrete frequencies and were terribly "harsh" looking. Newer fluorescents, especially CFL's, typically have around 5-7, from what I can tell using a cheap diffusion grating spectrometer. You can buy special "daylight" fluorescents with a spectral composition that better approximates sunlight, but I don't think these are actually continuous spectrum.. they just have a spectral distribution closer to sunlight.
White LEDs, unlike fluorescents, are indeed continuous spectrum light sources, as verified by spectrometer. As with any source, there are still stronger and weaker portions of the spectrum.
The whole "full spectrum" lighting as a cure to depression, irritability, skin cancer, your poor stock returns, etc. appears to be a scam as there is no scientific evidence to this date to support the claims being made. Seasonal depression is proven treatable using any bright light source, whether "full spectrum" or not.
Unlike other posters, I'll give you the benefit of the doubt that when you say "Linux" you really mean "Open Source desktop software distributions that may use Linux at their base" I often use the same sloppy terminology, so who cares.
.NET/Mono or Qt or Gtk2 or any combination thereof.
2. Mac OS X and Windows revolve around the interface. On the library level, new interfaces are added, but older ones are still supported for a surprisingly long time (see Carbon / Classic Runtime Environment for Mac OS X, or Win9x Compatibility Mode / Application Compatibility Toolkit for Windows 2000 / XP). Certainly, support is eventually dropped, but the pace is normally quite slow for popular APIs.
If you develop with cross-platform libraries, as you should unless you have a really special need, then all desktop platforms "revolve around the interface." In fact, the added layer of abstraction allows you to leverage the same consistency to all platforms. Qt, Gtk2, and Java/SWT are some good choices.
Now, if you're talking about desktop environment GUI consistency, well.. there is simply no such thing. But do note that the gradual evolution of KDE over the last 10 years has been no more dramatic than Windows95 through XP.
And of course, looking towards the future, the web is the new platform of consistency. "Linux" will not likely beat Windows by replacing it as a desktop platform. The desktop platform will just become increasingly irrelevant and Linux will be left as the defacto choice simply because it is the cheapest way to get cheap hardware on the web reliably. Cross-platform libraries will take care of all the rest -- whether Java or
In short, I see all of this as signs that Linux is moving too fast for people to consolidate and work out standards.
On the other hand, standards are emerging nonetheless -- by the fittest surviving. You might see Open Source has having been through a chaotic rush to catch "unix" up to the status quo in desktop software. It's not over yet, but it's showing signs of slowing. The other thing to consider is that some standards are more important to consolidate on than others. This leads me to my next point below...
We're in a transition period where people are held back by old systems, but seek cross-platform compatibility on new systems. It's like watching pressure build for a switchover. Just because there's been no mass exodus you still see them untangle themselves from Windows strangleholds.
You've just hit on a crucial point that the original article missed. "People are untangling themselves from the Windows stranglehold." Or another way to put this: desktop platforms don't matter anymore! Look at what people (companies) are doing to untangle themselves from Windows. They're replacing traditional Win32 client apps with web applications, local or hosted. They're replacing Exchange/Outlook with 3rd party mail software, web based groupware, mail server "applicances", outsourced mail services, etc. Each step along the way breaks away from the very need to have a consistent desktop platform in the first place. The web is the new platform of consistency. It is driven by true standards -- open standards, like the work of W3C. And it matters not what hardware or operating system you run. (Though Linux or BSD will get you on the web for cheaper, both on the server and client side)
It's curious that the article ends with a claim that MacOSX and Java are the only "platform" that can compete with Windows. In fact, Java is a very important piece of the "web" ecosystem -- but it is fully cross platform and runs no better under MacOSX than under Linux. MacOSX is "unix" just as much as Linux and BSD are. The only reason to use MacOSX as a platform is for its really slick graphics / GUI toolkit libraries. And those are only needed for specialized applications not yet suited for the web.
The pressure is indeed building and the whole paradigm of how we use computers is changing. With the exception of a possible surge as OpenOffice.org* reaches "acceptability" I don't expect to see a mass exodus. I expect people to wake up one day and say, "hey.. what ever happened to that 'Windows' thing that used to be everywhere?"
* I see OpenOffice.org as a transitionary crutch. The word processing / office suite paradigm itself is quite dated and will not survive the full onset of the web as a true platform.
I'm preaching to the choir here but reverse engineering is a Good Thing for all communities. There is absolutely no reason that we should not support working around what others have obfusticated to make money for themselves.
I completely agree. RE is an indispensible tool.
On the other hand, I think we should be cautious not to extrapolate Linus's views inappropriately. The inference that he is completely against RE in OSS is baseless. The simple fact is that he had a conflict of interest over BitKeeper and was forced to be political about the whole situation. It is quite possible that his statements were intended to de-fuse McVoy and/or others in attempt to mitigate any impending bad-mouthing of the community he helps to shepherd. I do not think Linus was trying to make any sort of 'moral statement' whatsoever. He even admitted that it was only a matter of time before the Linux community would move away from BK to something open source.
People mostly do things for their own benefit, as they should. I don't think it's good to encourage decent people to sacrifice themselves for the benefit of people who give nothing back. That just leeches the resources of decent, generous people and gives more power to the other sort.
Which is why most of the good Open Source software out there is commercialized in one form or fashion.. Of course, there are other projects that thrive on the sheer fun of doing something really new and interesting.. like Blender. But most projects should be heavily commercialized. I personally am a big fan of direct user feedback through pledging to support desired feature development -- at least for smaller projects.
What you describe is certainly the ideal. However, only larger companies typically set up that much security infrastructure. They can afford to do it right, manually, because they have enough highly trained IT staff. Smaller organizations typically go to far fewer extremes. And, after all, they are much smaller / less interesting targets. So for them, the effort is better invested in an out-of-the-box firewall solution and having money left over for improving physical security.
Have you looked into webmin?
Webmin is a total joke -- well, not necessarily Webmin itself, but the quality of the modules available. Furthermore, nothing is integratable. On most commercial firewall solutions, changes to one component make proper changes to others. (Ex. turning on a transparent HTTP proxy would modify firewall rules for you after asking) I'm not saying this is an absolute necessity, but most people who install these commercial firewall don't have time to figure everything out for themselves. Webmin is an admin framework for people who are already *nix experts and just want a convenient non-SSH way to do remote admin. Then again, for such people, it's usually faster to just use SSH, so the existance of Webmin is hardly more than a "checklist" item. (ie. "yeah, of course we've got a web admin console..")
Astaro isn't free. By the time you build a machine and get an Astaro license, it costs just as much as any of the pre-built "security appliance" devices. Also, last time I did a comparison, it's not even very competitive with regards to features.
Almost all of the commercial firewall solutions are already based on Linux or BSD. The point is to create a free (and superior) replacement for the likes of Astaro. (and, if you will.. the "distros" that come with security appliances) I am 100% against proprietary Linux/BSD distributions. There's no need for them. The only reason they exist is that the Open Source community has dropped the ball just short of the goal. The advantage of free distros is you have full control and there's no reliance on any company for updates.
One particular scenario was a firewall. I suggested a Linux firewall due to the lower upfront cost. Now, there were a Microsoft shop, but a firewall is not something that has to be administered everyday (when it is working properly). Instead they decided to go with a Checkpoint firewall that cost them a hell of a lot more.
Here's the problem: A firewall today is not just about "Linux kernel + iptables." Those Checkpoint boxes (and others) are full featured "security appliances" as they call them. They have features such as:
- application proxies to filter / virus-scan / monitor content: HTTP, FTP, SMTP, etc. (so you can say.. stop employees in the sales dept. from playing games on Pogo during 9-5.. to give an example of the granularity of control available)
- network monitors and various intrusion detection / prevention methods
- complete mobile VPN services, including dynamic firewalling rules
- user authentication services (used for VPN, proxies, replication to other network services, etc.)
- very complete GUI admin tool / management console. (and multiple security devices can be linked together throughout the company..)
- daily automated security updates (virus updates, IDS signatures, firewall software updates, etc.)
Can you do that all with free Linux distros and available OSS tools.. Mostly. Will you be saving any money by the time you've got all the raw materials kludged together into a working solution? Nope.
The Open Source community has failed miserably at producing real-world solutions. It has produced an enourmous amount of quality raw material. (And if you examine the commercial firewall solutions, you'll find much OSS being used internally!)
I think there is a good solution to this: The major free Linux/BSD distros need to have subprojects focused on specific needs. For example, there should be a "Debian/Firewall" sub-distro. (note: not a fork) It should provide a more or less ready-out-of-the-box firewall solution using pre-integrated "best of breed" components from the base Debian distro. If there are shortcomings discovered, the improvements can be fed back into the base distro using standard processes. If there are flaws found in the raw materials, this is a perfect way to make sure that OSS meets real world needs through user feedback.
Now apply this principle to all major areas of network services.. mail servers, file servers, web servers, etc. As long as there is a decent web based admin interface, there will be no problem getting organizations full of Windows-only IT staff to use more OSS. (And meanwhile all the old-school Unix folks are squirming in their seats.. Sorry folks, I don't like it either, but sometimes pragmatism is required. There simply aren't enough smart Unix people to go around. So we either compromise or we let proprietary software continue to dominate the industry.)
Yeah.. try that one on your boss.. "Sorry, you didn't get that extremely important email, but it's more important that I continue my ineffective vigiliante attack on spam by the overzealous mail server configuration I've chosen."
RBL's and DUL's are evil.
http://www.gnu.org/licenses/gpl-faq.html
The issue is that you're only obligated to give the source to the people to whom you give/sell binaries. If you give the binaries to five customers, you have to release the source to those five customers.
Technically, if you do not disclose source code, you must provide, in writing, an offer valid for 3 years to disclose source code to those customers upon demand. The cost may not exceed the physical distribution cost of the source code.
As others have pointed out, the customers receiving the binaries and source are free to redistribute them, and probably cannot be constrained from doing so by any non-disclosure agreement..
Correct. NDA's cannot legally add limitations to the distribution of GPL-covered software. You may enter into an NDA with customers such that you agree to not distribute the modified software to any other party but your customers. But you cannot enter into any NDA that limits what your customers can do with the software once they've received it. And an NDA cannot trump your customers' rights to demand a copy of the source. Furthermore, if a customer was to redistribute a binary, the party receiving it would have the right to demand source from your customer. To meet that demand, the customer would be obligated to demand the source from you.
So when you get down to it, the GPL guards pretty well against this sort of thing by ensuring that any party involved can legally spoil this unethical business model. What would be really interesting is if an employee who was not under NDA with his employer disclosed the modified GPL software. In this case it would come down to whether the modified software is covered by trade secret law.
Is Mozilla actually more secure? Or is it just as bad as any other piece of software?
In terms of design decisions, you might easily say that Mozilla is more secure than IE. (not being integrated with OS and all..) In terms of coding bugs, Mozilla is no different than any other super complex piece of software. But there's another way to look at it. Because the Mozilla code is open, we might expect an ugly rash of bugs to be found near the beginning of its rise to popularity. But we might also expect this to rapidly taper off as all the major bugs are found and squashed. So you might say that now is a relatively dangerous time to use Mozilla (instead of say.. Konqueror or Safari). But, on the other hand, it's still not quite popular enough to attract the volume of real-world attacks that IE has received. Honestly, if you're some jerk running a malicious website, are you going to target this quirky bug in Mozilla or the myriad of IE exploits that are sure to pay off?
What does bother me is that the Mozilla folks haven't taken automated updates seriously enough. I cringe to think of how many Firefox early adopters have no clue what that little red arrow at the top of their screen is. Or if they do, how many dial-up users will be patient enough to wait for the update to download.. which isn't really an update at all but a full copy of the latest version.
If you are good, you don't need the labels anymore
Herein lies the problem. There are plenty of unsigned bands willing to try new things to get noticed -- including internet-centric schemes. But the problem is that most of them aren't very good. Or if they are good, they're not different enough from existing mainstream hits. When you get down to it, instrumentalists are a dime a dozen. I have several musician friends who can perfectly imitate and they're not even career musicians. (note: It's even easier today with all the fancy DSP gear available.. turn the knob to "Hendrix" and your cheap solid state amp all of a sudden sounds like a custom-rigged tube rack.) Hence we have the "starving artist" syndrome -- the hundreds of thousands of talented musicians etc. who will never make it but keep waiting for their "big break" as they continue playing at local venues and working crappy day jobs to survive. But it's no surprise why they will never make it. They don't know how to develop their own styles. They all sound alike.. plain vanilla white-boy rock.. fomulaic, dull, uninspiring. Meanwhile, the musicians who decided to try something even slightly new were scooped up by labels before they could even realize what was happening to them. (And 90% were probably subsequently dropped and may have left music forever..)
All it will take is for ONE band with an original "breakthrough" sound to make it big online using the "free music is free advertising" model and the RIAA will soon be history. You can't stop money from talking. Very few concert venues will refuse a profitable opportunity. If the audience demands, they must provide or somebody else will. And as we all know, live performance is where the real money in music is.
After years of complaining that the RIAA and MPAA were missing the boat, and should have embraced things like Napster instead of supressing them, we got iTunes and the like.
..Except that this is not really what P2P technology was intended to accomplish. The original purpose of P2P was to enable free and efficient distribution of media content so that artists would not need to sign all their rights away to publishers in exchange for mass distribution and so that consumers could access a much larger and more diverse selection of media. Of course, before this could even be attempted, the whole idea was hi-jacked by a bunch of lamers who just wanted free crap-pop/rap major label music. (aka. original Napster, etc.) It was all massively popular because, well.. the music being shared was already popular -- thanks to radio airtime, MTV, etc. If there were any independent artists legitimately trying to make it on the old Napster, they were surely drowned out.
The fact is, iTunes and others have not really changed the music distribution landscape that significantly. They're just the new middlemen. Despite iTunes' relative popularity among other upstart online music ventures, it has made a fairly small dent in the overall market. More to the point, these new pay services do not yet offer the promise of P2P. Sure, the barrier to new artists has been lowered and that's a step in the right direction. But these services do not enable either the "sell tickets, give away music" or "give away some, sell the rest direct" business models. And why would they? That would cut them out of the game.. and there's big money in being a middleman! Of course the price of middlemen is paid by all other parties. Consumers get DRM, less-than-CD quality downloads, and higher prices. Artists lose a large percentage of their sales. The public gets an ever shrinking percentage of public domain or otherwise freely re-usable material.
If we really want to change the content industry for the better, we need to start thinking more strategically. Many good components are already out there.. Creative Commons licensing, FLAC and other lossless audio codecs, Bittorrent and copious other P2P technologies, companies like Magnatune and CD Baby, etc. But nothing combines them all into a single powerful tool!
What we really need is a true P2P service designed from the ground up to be a channel only for freely licensed content (of all types). It ought to have an indexing and searching service like the original Napster, but which only tracks legit free content and which uses hashes to validate original, unmodified content for those who want it. Then, add a user-feedback ratings and review system so that the inevitable crappy content does not waste everyone's time. Finally, make the whole thing commercially supported via artist services like online ticket and merchandise sales (linked through the index / artist database system.) All the software can be Open Source because there's no DRM involved, but don't allow contributions to the official project which add support for exterior searching. Trademark the name so that knock-off projects can't dilute the marketing. Make sure the original is so high-quality that nobody wants to fork it otherwise.
Incidentally, the whole indexing / artist services system could be run by a not-for-profit organization. Think of it as an "RIAA" for independent, freely-licensed content. Artist services could certainly be provided by outside businesses as well.
Another possibility is that I might convert them to Linux and KOffice (or whatever I end up using).
:-)
I gave KOffice a fair shot and there are definitely nice things about it (speed and memory usage for instance..) But KOffice simply doesn't have the file format support that OpenOffice does. The ugly reality is that people still need to read/write MSO document types. Save yourself some time by using OO.org for now and switching to something else later when it becomes available. (hopefully that'll be some sorta rich web app that makes word processing obsolete anyhow!
Or, I'm a company with 5000+ developers and if I suddenly decided to put them all into Linux application development, I would go bankrupt.
With that many developers, you'd have quite a bit of negotiating power to lower the price. Of course you probably wouldn't be writing strictly desktop apps anyhow.
Suddenly jumping on to become completely dependent on the mercy of Trolltech is not good business sense, even if it is a fact on life in Windows world.
If you are writing proprietary apps, you are creating vendor lockin for your own customers. Pot calling the kettle black, eh?
What if Trolltech suddenly decided to go evil and put on a price tag of, say, $10000?
They'd go bankrupt. Somebody else would continue the GPL project and another would probably re-implement the Qt API's with a LGPL licensed project. (Which incidentally is always an option anyhow..)
On the other hand, it would be a complete disaster if Qt suddenly became the "default" toolkit for Linux.
Market demand would then allow Trolltech to lower their developer license prices. And the rest of us would have a superior and consistent GUI toolkit. (-:
I still maintain that the reason OpenOffice is moving forward so slowly is because it's so damn complex, and the number one priority of everybody involved with OpenOffice should be to make it simple enough that more hackers can help out.
:-)
I've seen the code and it's hideous (though slowly improving). It would probably be easier to bring the KOffice suite up to the feature level of OpenOffice.org than to completely re-write OO.org. But OO.org has momentum and corporate backing and KOffice sadly has virtually none. Incidentally, remember how long it took to turn Netscape Navigator into Mozilla Firefox?
Say what you want, but a large percentage of the folks that are gutsy enough to be rolling out OpenOffice.org are doing so at least in part because they are Free Software advocates. In short, they are pushing OpenOffice.org for ethical reasons, and not for practical reasons.
.NET.
.NET works quite well alongside Java and that MS is actively trying to woo Java developers to the Longhorn platform, with all it's proprietary but dazzling client-side APIs.
Reality check: Most people who are experimenting with rolling out OO.org are doing so on Windows desktops, whether or not it's for ethical or practical reasons. Most people who are doing really huge OO.org rollouts in large companies are doing it entirely for practical reasons.
Sun is just being stupid on this front. Java has already fragmented into several mostly compatible forks. IBM has their own JVM, as does Apple, Oracle, Borland, and there are a wide range of Free Software Java-alike systems.
I agree that Sun is being stupid by not releasing more of their JVM and class libraries as Free Software. On the other hand, it is inaccurate to say that the forks of Java are "mostly incompatible." The situation is generally better than C/C++ compilers, so put it in perspective.
Freeing Java would completely kill all of the non-Sun Java toolkits, and it would give Sun the Free Software allies it needs to compete against Microsoft's
Yes. And this is even more true considering that
And I might have gotten OpenOffice made an acceptable alternative to MSOffice. But I'm not going to try. I won't recommend products that I don't use.
So this hypothetical company is going to continue filling MS's coffers because of your silly purism? Do you quite see how ironic this is? That's like being outnumbered in battle but shooting one of your fellow soldiers because you didn't like the joke he made.
There is almost no question that Java will be fully Open Source within the next 5 years -- either by action of Sun or "Java-like" OSS projects. There will be plenty of time for purism after the MSO monopoly tumbles.
Free Software hackers want to like Sun, and they want to like Java, but the licensing issue is a big deal to them. Unfortunately for Sun, it absolutely needs the Free Software hackers to jump on board.
Free Software hackers with any common sense and/or pragmatism like Java for jobs where it is the best tool, but they don't necessarily like Sun. The two are not really tied together -- especially considering that much of the Java software developed by the Apache/Jakarta project (and even Eclipse) is starting to run on Free JVM's and GNU/ClassPath. Does Sun need the F/OSS community? You bet.. and they know it by now. If it wasn't for OSS alternatives to some of the awful marketecture specs coming out of the JCP, Java would not likely be in the position it is today.
Heck, I develop in Java for a living (web development), and yet I still don't use a single desktop Java application outside of Eclipse and it isn't pure Java.
Same here. I don't see Java as a language well suited for lightweight desktop app development and I think it's silly that the OO.org developers went with it due to memory requirements alone. On the other hand, Mono/.NET is currently way behind Java for serious business applications. (at least on the backend / middleware..)
Yeah? What if Microsoft gobbles up Sun at some point in the future?
Then we'd all switch to IBM's JVM, silly. And everyone would start working overtime to get the various Open Source JVM's and GNU/Classpath up to where Sun's reference implementation had been. You act as if "Java" is a program rather than a language. You do realize that there are literally dozens of JVM's and Java compilers right?
Obviously the optical challenges would be pretty huge.. which is why I meant the post primarily as a joke about silly design patenting.
On the other hand, some sort of extreme wide-angle lens (fresnel?) combined with perspective correction in software might be able to sorta make it work.. at least with more distant objects behind the screen. It'd be interesting eye candy regardless..