If you even read the java specs and then write another java implementation, sun owns your code. Java is as proprietary a language as they get.
Ironically, it's probably more proprietary than most of what Microsoft does with.NET (other than bogus software patents). Somebody correct me if I'm wrong, but I don't think C# has any restrictions like this.
Exactly. Fact is, an Open Source Java would gain more platform support and would become faster and more stable on the platforms that already exist.
Free Software hackers would like to root for Java on this one, but they can't because Java isn't Free. So instead they are spending their time creating Mono or working with Python, Perl, or Ruby.
Right again. I would love to use Java instead of Python, but if I do, my software won't be picked up by Linux distros. Java is, hands down, far superior to any of the interpreted languages available to us. Python has it's benefits in other areas, but it can't compete on speed. Perl is.. well.. I won't go there. OK, it's good for simple scripting tasks like it was originally designed for: extraction and reporting.
Here's the bigger issue: Enterprise business software. This is by far the largest weak spot in the whole world of Open Source. We have absolutely nothing to compete here other than fledgling projects like GNU Enterprise (Python), which aren't ambitious enough to meet the needs of serious business software. Except for.. JBoss, which is Open Source and.. You guessed it: It requires Java!! While I'm not saying that J2EE is the ultimate platform, it does do a lot of things right and it's available now. If Sun opened its JVM/JDK, Linux distros could include JBoss and Free Software hackers would actually start using it! Sure, we should have more than one solution. I still see projects like GNU Enterprise as worthwhile for developing "middle of the road" solutions that aren't as complex as J2EE and aim for the KISS principle. However, Java is a powerful tool that is direly needed in the Open Source toolbox.
You are allowed to develop compilers on your own, for free. You are also allowed to develop your own JVM for free. You are just not allowed to use Sun's code to do it.
Clarification: you are not allowed to use Sun's documentation / specifications either. So trying to implement a free Java compiler / JVM is like flying blind with half your instruments out. That's why the various OSS Java projects like Kaffe are so behind and so incompatible. Even if, for some strange reason, Sun decides to keep their JVM code secret, they need to remove all the weird documentation licensing (NDA-style stuff).
And switching to Open Source is not free, indeed it is often not even cheap. The costs are real, but so too are the advantages.
This is certainly the correct perspective to hold. Open Source is not about a free lunch. It's about meeting your needs most efficiently. One thing that you should consider in evaluating Open Source solutions is the possibility of spending some money to improve them if they almost, but not quite, meet your needs. Maybe, for example, OpenOffice is all that 40% of your company's employees need except for a handful of features that could be added in a month's time either by your IT department or by paying someone involved with the project. While this cost would be included with the Open Source TCO, the final outcome may likely still save money now and definitely save money in the long term (no more costly upgrades).
There's also the snowball effect of OSS. If you support the OpenOffice project today, it will become more useful to other companies, encouraging them also to support it. Eventually, maybe 5 years from now, it will be good enough that you won't need any MS Office licenses. So part of today's Open Source TCO can be seen as a long term investment.
NOTE: The reason I use OpenOffice as an example is that this is software that every business needs. It's a shame there is no standardized way for businesses to contribute in exchange for specific needed enhancements.
This and the fact that most geeks do not need the software is why it will not be written anytime soon. No one wants to do the boring drudge work involved to make it usable. Just look that the ever-increasing number of half-finished OSS projects for proof. Onces the sexy code is written, development slows, documentation is neglected, and developers move on to the Next Big Thing.
The fundamental problem here is that many (most?) geeks view OSS as just free software / free lunch / hobby. Sure, nobody wants to do the drudge work, but there's big money in that drudge work! Consulting / development of OSS can be a very lucrative business, but most geeks are afraid to try! As a result, all the OSS business software sucks terribly. Let me be even more to the point: this is holding back the entire OSS revolution.
Remember: Open Source Software is not about a free lunch. It's about meeting your needs as efficiently as possible!
What we need is professional, Apache-style collaboration on a single complete and modular OSS business software solution. It must be designed with modern software architecture and not cut any corners whatsoever. That means professional n-tier design and not crappy, monolithic, web-based PHP/*SQL nonsense. Think J2EE but using entirely FOSS. Right now, the only project that is anywhere close to moving in the right direction is GNU Enterprise (www.gnuenterprise.org) but it needs a lot of development help before it can become usable. The other option is JBoss, an OSS J2EE solution, but it still requires the Sun JDK as the OSS JDK's are all way behind. J2EE is also massively complicated with a steep learning curve and kinda fails the KISS principle in my opinion.
I hear many people on slashdot harp on the benefits of OSS in one breath, and then complain about their jobs in the next. Wouldn't it make sense for these people to get a job writing software and getting paid for it, instead of writing high quality software for free and then earning money doing something they hate?
Absolutely! Once geeks realize that OSS is not just a toy but a means to make a good living, we will see the latent OSS revolution take place. For this to happen, however, we must truly collaborate and strike out on our own.
If you look around, most of the best OSS is that written by people who are being paid for their efforts one way or the other. Always remember this: OSS is not about a free lunch. It's about meeting needs in the most efficient way possible.
For core OSS business software, we need an organization like Apache that serves as a highly professional center of collaboration, funding, and outside contribution.
I used to be a CE major until I realized it was preparing me to become a Dilbert clone -- that is if I could get a job at all. My peers in the major were largely a bunch of dolts who were in it only because US News & World Report said that was where the good jobs were -- and not because of any passion for technology. I was the opposite. So what do people with a lot of passion who don't want to become corporate slaves in an unethical industry do? They start their own businesses. Sure, it's a risk, but the other option is a boring life if you're the type that likes to make decisions, do your own thing, change the rules, etc. My current passion is Open Source software and my profession is consulting. (Which is a great combination with virtually no overhead, btw. Yes, you too can be paid to write free software.) If I decide to do something else later on, there's nothing stopping me. I'm quite skilled in many fields and have enough business sense to not do anything stupid while trying. If I was ever to go back to school, it'd probably be in some cheesy business major just to get the degree. Yes, I say cheesy because compared to the rigorous hell that is an engineering / medical / science major, anything else is a piece of cake.
Intel doesn't want to risk being associated with these kinds of things (and you know if they released an open source driver, someone would).
Wrong. Their other non-Centrino wireless hardware is supported. As is that of just about every other manufacturer. Any manufacturer that makes claims similar to what you suggest is only throwing out BS to cover up their laziness / stupidity. But this is not why Intel is not documenting Centrino anyhow.
The volunteers who write and collect free software are not doing it for your sake. You are fortunate that you can benefit at all from their efforts; you have no standing to complain that they "waste" their time on what they see as interesting or worthwhile to do.
Wow, you are entirely missing my point. Of course people are free to do as they please and nobody is here to command them one way or other. On the other hand, in any community, there is wisdom in listening to others ideas before engaging in ones own actions. It's not a community if everyone does their own thing without any regard for a mutually beneficial outcome. There are good arguments why the endless creation of new Linux distros is deterimental to the community at large and wasteful of time for those who create them. Those people would be wise to listen to others thoughts on the matter before continuing. This has nothing to do with arrogance or whining on the part of those who make recommendations against actions that they personally see as unwise for the community as a whole.
The Open Source community has too many leaders and not enough followers.
Much of this regulation is acheived through the close-source drivers. Using modified drivers, it would be possible to make the card emit different frequencies or more power, thereby violating the usage licence.
Wrong. Any driver can be modified, whether closed source or not, to tweak the radio output. Closed source drivers are not at all how FCC regulation is achieved. This should be inherently obvious to the casual observer. In similar fashion, I could make my garage door opener fail to meet FCC regulation with some parts from radio shack. But if I do, I'm the one in violation, not the original manufacturer.
In reality what happens is the manufacturers document their wireless hardware and then the Linux/BSD wlan projects implement drivers. Yes, it just so happens that some hardware can be told through software to perform outside of FCC regulations. The default settings used by the open source drivers, of course, do not cause output violations.
Now, of course, the really interesting thing is when you get into designing your own antennas and tweaking the hardware to be just barely within legal limits. Similar to overclocking, a lot of wireless hardware has a significant "safety margin". Don't try this at home unless you know what you doing. (-:
It isn't a question of "we need", and it never has been. People create new Linux distributions for the same reason a lot of open-source software gets created -- because they want to. This is an obvious result of freedom: people can do what they want to, regardless of whether it is what anyone says "we" need.
While creating new Linux distributions is an exercise of freedom, it is unwise and wasteful. There is a huge difference between 'competing' software implementations (ex. Sendmail vs. Exim vs. Postfix vs. Qmail) and 'competing' Linux distributions. Writing new software ads diversity and brings about new ideas. In contrast, writing new distributions only duplicates the work of others. There is very little diversity and creativity involved. All Linux distributions are fundamentally identical. Sure, there are minor tweaks and different packaging and configuration tools, but there is no reason why these options cannot be rolled into the most popular distributions instead. The simple fact is that modern distros like Debian have plenty of flexibility to allow alternative components and experimentation.
However, I suspect that several factors may enter into the decision to make new software to accomplish the same goals as existing software: aesthetics, ethics, confidence, control of development, quality.
You are largely correct about these factors as applied to new software itself, but your point is invalid regarding the plethora of nearly-identical Linux distributions. I suspect that there is often something else at play: elitism. It's the "I'm too cool to use / contribute to a popular distribution, so I'm going to do my own thing instead and maybe get my name known too" This attitude is highly counterproductive in the open source community, where the tasks at hand are great and the available talent is currently limited. I say this as one who has done the whole "Linux from scratch" thing and realized that it was a total waste of my time beyond the first couple months as a learning tool.
In the commercial realm, this 'elitism' of sorts becomes the failed dot-com business model where you hope that branding alone will make you money. Plain and simple, there is little or no money in Linux distributions themselves. This is why RedHat discontinued its standard distribution to free up more resources for business solutions and services. Other aspiring Linux companies should take note and get behind existing popular distributions instead of doing their own thing. It's really better for everybody.
If the consumer really weren't benefitting, then they'd seek out alternatives: they do exist, as I'm sure you know. I've been using one for 9-1/2 years: Linux and associated alternative applications.
You make a large and faulty assumption: that consumer behavior is a good barometer of whether the free market is working. The only reason why consumers are not more actively seeking alternatives to MS crap is that MS has used its monopoly power to marginalize or destroy all of its competitors. Open Source is the first real 'competitor' that MS has ever had. If OpenOffice had started 10 years ago, MS Office would be holding on by a thread today (if existing at all) and we wouldn't be having this discussion.
Of course, standards would get us to Nirvana just as market dominance by MS is and allow the tiny minority of us who use something other than Windows to make full use of our computing power under our chosen environments; but there's hardly an economic or public benefit argument to be made for having the guv'mint do our dirty work and go after MS just because they don't make the lives of (generously) 5% of computer users easier.
Open standards are a large part of the solution and they should have been enforced far earlier in the development of the industry. There is a very strong public benefit argument for going after MS, but only in the right way. The one good thing the government could have done in the anti-trust case was to force MS to turn over all proprietary data format documentation without royalties and nullify all of their related patents. But they didn't. Instead, they slapped them on the wrist with a modest (in M$ terms) fine.
the government doesn't need to pursue litigation, but only needs to state that they will purchase only software that stores and transmits data in royalty-free formats so alternative vendors can be used effectively in price negotiations.
This would be a step in the right direction, but it doesn't go far enough. MS would find a way around this requirement. They'd do something like make a government-only version of the software capable of exporting to a royalty-free format. And really, given the political clout that proprietary vendors have in the US, how likely is it that such a requirement will ever be made given the current circumstances.
In the end, there is one solution to the problem: elimination of the threat. Anyone who thinks that the proprietary software industry is capable of reforming and playing nice is deluding themselves. We will not see the end of this software patent / DRM / proprietary formats / etc. nonsense until Open Source software fully dominates the market. This is both a worthy and realistic goal. What have YOU done lately to help this happen? Or are you just complaining on Slashdot and doing nothing about it?
Here's somewhere to start: http://www.openoffice.org/contributing.htm l Can't code? Send them some of the money you would have spent on Office licences. Or make an investment looking towards the future so that you can soon ditch Office and save a lot of money in the long term.
There's no such thing as a free lunch, but some lunches are much more efficient than others.
Re:Linux apps too hard to configure?
on
Build Your Own PVR
·
· Score: 4, Insightful
The thing that concerns me is that for some reason there's a mode of thought throughout most slashdot articles as of late (2-3 years) that linux should be as easy to use as windows. Do you really want this to be the case?
Yes, Linux absolutely should be available in a form that is as easy to use as Windows. What concerns me is that there is still such a holdout among elitists like yourself. The notion that allowing Linux to be newbie friendly will somehow make it worse for us hackers is downright asinine. I briefly held this belief myself before realizing how stupid it was. Every example you gave of the "horrors" of a newbie-friendly Linux is pure hyperbole and utter BS. (And frankly, they show that you have little clue what you're talking about.) The development of software and utilities that assist ordinary users will not in any way affect those of us who do most of our work in a bash shell and a vim session. To the contrary, it will help us immensely because more software will become available for our platform of choice and any kludgy configuration issues that exist today will be standardized by necessity as automation tools are developed.
Also, may I ask you a question? Do you ever plan to actually use the skills you infer to have to make a living? Well guess what? It's a no-go if ordinary people can't use the technology you are most adept at and if Linux doesn't drive a significant 'market-share'. So what is your solution? Let Microsoft and Apple have all the desktops and let Linux/BSD be niche players for the geeks only? Guess what? If that happens, we will have a world filled with DRM garbage and security nightmares. Get your head out of the sand.
...connectivity to a "mainframe" for financials, records, etc. which generally means some sort of TN5250/whatever emulation.
While this may be true of your shop, I've run across plenty that are using far more sophisticated and modern accounting / ERP / CRM / etc. software that is not so trivial to support. One of the key things that the Open Source movement needs is a solid foundation for creating this type of enterprise software. Sure, there are proprietary solutions already available for Linux. But the real revolution will occur when businesses see a reduced cost not only from what free OS's they can run, but what free *enterprise* software they can run on it. There are basically two options: J2EE or create our own solution. Unless Sun wakes up someday and realizes that there's no point in trying to keep the Java source and official documentation proprietary, we need to pursue the second option as a backup. (Incidentally, there are already JBoss and others on the open-source J2EE side, but there is virtually no OSS that actually uses J2EE and the JVM/JDK is still non-free as mentioned)
GNU Enterprise (www.gnue.org) is one probable option for an alternative, but it needs a lot more development help (and perhaps some funding) as the project seems largely stagnant at the moment.
Where's the money in all this? Consulting and services. Very few medium/large shops have their needs met by off-the-shelf business software. There is money in putting together custom solutions that work, whether you are using proprietary or Open Source software. The biggest difference is that you can better compete on cost with OSS because you don't have to bill your clients for exorbitant licensing fees. We need an Apache style model where consultants / IT departments collaboratively pool money into developing the base components of all major business software. After that, customization and support are the icing on the cake.
Among the predictions: no more rentals from video stores, no used games market, no lending games to friends, less upgradeable computers, pay-as-you-play software subscriptions, and other consumer-unfriendly changes.
If these predictions come true, then allow me to add one more: Open Source games take over the market.
In all seriousness, the community needs to write free, high-quality game engines upon which content and online services can be sold. And, of course, "the community" here largely represents companies who want to sell such content and services once the engines are collaboratively developed. Then we just need a non-proprietary console system.
Unfortunatly, BSD/Apache style licenses are looking like more a viable option.
BSD-style licenses are ok for some things but very wrong for others. As a consultant who develops and supports free software business applications, there is an enormous advantage for me if other people who use my work and improve it must contribute the changes back. I've put an enormous amount of work into the foundation of my software and I don't want people freeloading off my efforts by selling proprietary derivatives with just a little bit of extra polish. GPL ensures my work stays free and that others play fair in the open source "code economy."
If I were the mplayer developers, I too would be quite outraged. If KISS is in violation, they owe back to the community any improvements that they have made or else they must license mplayer under a commercial license and thereby support the project in doing so. Those are quite fair rules to play by. Open Source Software == efficient development model and != free lunch.
Tell me who did real-time time shifting of TV shows (including watching the beginning of a show while the end of that same show is still recording) prior to TiVo. You couldn't do that with a VCR, and nobody was using PC's to time-shift at that time
Anybody with a video capture card could do that easily and video capture cards were available long before TiVo. There just wasn't a dumbed up user interface and it was on a computer instead of a set-top box.
If you don't like patents in general, then you can argue against it on that position, though TiVo would likely be out of business without it.
Perhaps the contention involves what a patent really is. Very rarely does a patent represent a truly new idea that nobody has thought of before. Even rarer was any true R&D required to produce the patented idea. (And in such case I believe the patent term should be significantly shorter..) "Intellectual property" is an oxymoron. On the other hand, there is perhaps a certain value in allowing people / companies to try marketing something new without being immediately bombarded with larger, perhaps better-known competitors who will surely put them out of business by duplicating their designs and selling quantity before they can become established.
For this purpose, TiVo perhaps ought to have exclusive right to sell time-shifting set-top DVRs within the limited definition of their patent. On the other hand, their patents should not be allowed to apply in any way to software itself or to general purpose "multimedia PCs" that just happen to have software that does what their DVR's do. (Whether this is the case I am not sure, but that's how it *should* be.) In short, I believe patents ought to be limited to very specific *physical* devices / inventions. From that perspective, their patent does not seem to be excessive. However, I could be wrong since IANAL.
The author was just using SCOs stock as an indicator that the company is not going to lose the court case like everyone else assumes will happen.
My opinion: If that's the case, the author is an idiot. This whole situation shows the signs of a textbook pump-and-dump manuever. The fact that Linux was chosen as a bogus legal target is fairly irrelevant other than the fact that MS may have been involved with encouraging that decision. IBM is far larger and more powerful than both SCO and Microsoft put together. If they thought they stood a chance of losing, they would have just bought SCO outright.
My prediction: In the end, this whole thing will backfire on the evil men who started this mess. Linux will be championed not only as a victor, but as an unstoppable force. SCO will wither and die for lack of a workable business model. MS will continue to lose the PR war against OSS.
Sidenote: It is not ethical to invest in an unethical company just because their sleazy tactics are causing a temporary stock rise.
So then the next suggestion above is to hire someone. With what money? And how can I justify spending ten times or more the cost of some proprietary software package hiring programmers to improve (or create) a free software competitor? Especially when my hypothetical freelance business probably isn't exactly rolling in the dough.
You are partially correct. OSS should not end up costing more than proprietary solutions. Personally hiring 3rd-party developers to improve the OSS you use rarely works.
This is why all the major OSS projects need to adopt means by which ordinary end users, such as the hypothetical graphics artist in this example, can donate reasonable amounts to respective projects. In return, they should get a say in prioritizing feature development. Considering how much graphics artists spend on proprietary packages, there is a lot of money out there. If a quarter of the people who use Photoshop skipped an upgrade and instead chipped in $100 to support the Gimp project, that would pay for a whole lot of developer brain-hours. Yes, it would be a longer-term investment, but certainly one that would pay off.
A more likely scenario is the graphics studio that finds Gimp to be, say, 99% of what they need and are willing to donate to the project what they would have spent on Photoshop in return for making Gimp 100% of what they need in the next 2 months. Then, 2 months later, another studio finds that the improved Gimp is now 99% of what they need.. and this cycle continues. In the real world, where not everyone is a developer, this is how collaboration in Open Source development must work.
It's as simple as this: If we want quality OSS, we need to financially support those who create it but are not financially rewarded elsewhere for their labors.
iTunes is a very simple piece of software and a successful marketing campaign. Just because it is (arguably) well done should not classify it as an innovative technology -- especially when it's not even a new idea.
I personally have no interest in using any of these flashy new online music stores. Until they are DRM-free, use an open protocol (ie. cross-platform), and offer lossless formats, just say no. In the meantime, support only independent artists.
Many companies lack the skills to maintain code--they simply don't have developers (or at least not the right sort of developer). To meaningfully contribute monetarily would erode at the cost savings. If the company is public, there may even be further complications.
Making meaningful monetary contributions makes sense if you get something in return. That's why OSS projects need to adopt various "feature adoption" schemes or other mechanisms to increase the incentive to donate to future development. Also, a "meaningful" amount is highly variable. If you have a popular piece of software receiving $100/year donations from thousands of companies, that is certainly meaningful.
There's also the "hire a consultant" option -- someone who can help maintain and improve the free software a business uses. Smart projects ought to start listing "registered consultants" who have proven themselves on the codebase. There is plenty of money in business software whether you are a consultant providing proprietary or OSS solutions. OSS just gives you an extra cost advantage over your proprietary consultant competitors who have to pass license fees on to their clients as part of the final software/service package.
Considering how many *useful* open source projects could use some funding, this is a really silly waste..
Now, if somebody would collect money to sponsor the completion of full SVG support in Mozilla, THAT is something I'd be more than willing to donate to. And it would help more than a tiny handful of people.
Open source is obviously faster/better/cheaper when 1000's of people donate their time to a single project. The only open source project I've been involved in was a collaboration among several corporations, all of which wanted to leverage each other's resources, but none of which could really contribute their own.... There's nothing like money to motivate people to work on a project for which people aren't willing to donate their time.
This is definitely true. If you look around, you'll notice that most of the best Open Source projects are those where people are getting paid to contribute in some way. That's not to say that those same people would not have contributed otherwise, but money allows you to do things like drop your day-job and go full-time doing what you really love. The Open Source community needs to take a good hard look at how more experienced developers can be brought 'on-board' full time. OSS is beyond a hobby at this point. It's quite time to put that into clear perspective.
Open Source, at it's core, is about collaboration to meet needs efficiently. Part of that collaboration needs to involve paying developers so they can work full-time. Corporations who pool resources and collaborate on OSS projects to meet mutual needs are a perfect example. The same idea can work for individual users and smaller projects, however.
Take, as example, a typical desktop application like personal finance managers. We have GNUcash, which is a pretty good start, but it's missing a lot of the useful features found in the far more popular Quicken and Money. I personally have little interest in helping to developing GNUcash, though I wish it was a better fit for my needs. I'm not familiar with its codebase and I already spend most of my free time working on my own OSS project. (which I eventually plan to provide professional consulting services around..) However, I am willing to pay somebody $40 to develop a couple features I need in GNUcash. $40 is about how much I'd have to spend on Quicken or Money, which already meet my needs. But alas, $40 is not fair enough compensation for the developer. That's where collaboration comes in. There are millions of people who use personal finance software. If even 100 people contributed $40, that's $4000 compensation to add maybe one or two features -- easily doable in a month's time by an experienced developer. Realistically, there are far more than 100 GNUcash users able to contribute and far more than 1 or 2 features that need added. Once users start contributing financially to Open Source projects, allowing their developers to work full-time, we will see the true OSS revolution take place. The key is how to organize this process.
While some of his points have some validity, such as cases of improper itch scratching, he misses some much bigger issues. There will always be immature people in any community or company. And there will always be people re-inventing wheels (often just for the learning experience, mind you) The challenges we face are more related to communication and better collaboration.
1.) Standards for simple everyday stuff While diversity in implementation is good, certain things simply need to be agreed upon community-wide:
Font handling: not only at the X server level, but for applications. There should be a single config used by all apps, whether KDE/Gnome-2/Gnome-1/Mozilla/etc.
Drag-and-drop. Fortunately, this one is being worked on for next-gen X11 versions, such as the Xdnd extension. Once complete, everybody needs to actually use this.
Browser bookmarks. As with fonts, there should be a single, common source used by any browsers installed. XBEL anyone?
Basic system configuration. There's no reason why all the distros can't decide on a standard configuration system for hardware detection / modules / network settings / X11 config / etc. This would allow KDE/Gnome interfaces to be developed and used across all distros.
Security related configuration. The reason why there are so many successful hacks of Linux machines is that it is so difficult for the average person (or even a fairly knowledgeable person) to properly lock down a machine. There needs to be a standard configuration interface for managing network services, building netfilter configs, etc. ACL-enhanced kernels and easy crypto filesystems also need to become the norm. Again, once a standard config interface is in place, standard tools can be written for Gnome and KDE and used across all distros.
2.) Enterprise database software This is the "killer-app" for nearly all businesses and should be the prime focus for those who want to see F/OSS on the corporate desktop. Trouble is, we don't have a good F/OSS enterprise-class application server! Forget cheesy PHP/MySQL apps. Forget Zope and other web application frameworks. That software has it's place, but it's not up to the task of hard-core database apps. We need something as powerful as J2EE, but without the steep learning curve, alphabet soup of acronyms, sketchy documentation, closed standards, and dependance on proprietary Sun code. Yes, I am aware of JBoss (open-source, non-official J2EE implementation), but it still has the issues listed above. Either a completely F/OSS JRE/JVM/JDK software stack needs to be written or else a completely new application server standard needs to be invented, perhaps using ObjectiveC or Python. GNU Enterprise (www.gnue.org) has a start, but that project seems to be stagnant and uninspired.
3.) More professional developers Simply put, the F/OSS of greatest quality is usually developed by people who are being paid to meet needs. There are plenty of viable business models and market opportunities. Hobby projects won't cut it. If you want F/OSS to succeed, you need to do your part in the marketplace.
Slashdot Mirror
If you even read the java specs and then write another java implementation, sun owns your code. Java is as proprietary a language as they get.
.NET (other than bogus software patents). Somebody correct me if I'm wrong, but I don't think C# has any restrictions like this.
Ironically, it's probably more proprietary than most of what Microsoft does with
Exactly. Fact is, an Open Source Java would gain more platform support and would become faster and more stable on the platforms that already exist.
Free Software hackers would like to root for Java on this one, but they can't because Java isn't Free. So instead they are spending their time creating Mono or working with Python, Perl, or Ruby.
Right again. I would love to use Java instead of Python, but if I do, my software won't be picked up by Linux distros. Java is, hands down, far superior to any of the interpreted languages available to us. Python has it's benefits in other areas, but it can't compete on speed. Perl is.. well.. I won't go there. OK, it's good for simple scripting tasks like it was originally designed for: extraction and reporting.
Here's the bigger issue: Enterprise business software. This is by far the largest weak spot in the whole world of Open Source. We have absolutely nothing to compete here other than fledgling projects like GNU Enterprise (Python), which aren't ambitious enough to meet the needs of serious business software. Except for.. JBoss, which is Open Source and.. You guessed it: It requires Java!! While I'm not saying that J2EE is the ultimate platform, it does do a lot of things right and it's available now. If Sun opened its JVM/JDK, Linux distros could include JBoss and Free Software hackers would actually start using it! Sure, we should have more than one solution. I still see projects like GNU Enterprise as worthwhile for developing "middle of the road" solutions that aren't as complex as J2EE and aim for the KISS principle. However, Java is a powerful tool that is direly needed in the Open Source toolbox.
You are allowed to develop compilers on your own, for free. You are also allowed to develop your own JVM for free. You are just not allowed to use Sun's code to do it.
Clarification: you are not allowed to use Sun's documentation / specifications either. So trying to implement a free Java compiler / JVM is like flying blind with half your instruments out. That's why the various OSS Java projects like Kaffe are so behind and so incompatible. Even if, for some strange reason, Sun decides to keep their JVM code secret, they need to remove all the weird documentation licensing (NDA-style stuff).
And switching to Open Source is not free, indeed it is often not even cheap. The costs are real, but so too are the advantages.
This is certainly the correct perspective to hold. Open Source is not about a free lunch. It's about meeting your needs most efficiently. One thing that you should consider in evaluating Open Source solutions is the possibility of spending some money to improve them if they almost, but not quite, meet your needs. Maybe, for example, OpenOffice is all that 40% of your company's employees need except for a handful of features that could be added in a month's time either by your IT department or by paying someone involved with the project. While this cost would be included with the Open Source TCO, the final outcome may likely still save money now and definitely save money in the long term (no more costly upgrades).
There's also the snowball effect of OSS. If you support the OpenOffice project today, it will become more useful to other companies, encouraging them also to support it. Eventually, maybe 5 years from now, it will be good enough that you won't need any MS Office licenses. So part of today's Open Source TCO can be seen as a long term investment.
NOTE: The reason I use OpenOffice as an example is that this is software that every business needs. It's a shame there is no standardized way for businesses to contribute in exchange for specific needed enhancements.
This and the fact that most geeks do not need the software is why it will not be written anytime soon. No one wants to do the boring drudge work involved to make it usable. Just look that the ever-increasing number of half-finished OSS projects for proof. Onces the sexy code is written, development slows, documentation is neglected, and developers move on to the Next Big Thing.
The fundamental problem here is that many (most?) geeks view OSS as just free software / free lunch / hobby. Sure, nobody wants to do the drudge work, but there's big money in that drudge work! Consulting / development of OSS can be a very lucrative business, but most geeks are afraid to try! As a result, all the OSS business software sucks terribly. Let me be even more to the point: this is holding back the entire OSS revolution.
Remember: Open Source Software is not about a free lunch. It's about meeting your needs as efficiently as possible!
What we need is professional, Apache-style collaboration on a single complete and modular OSS business software solution. It must be designed with modern software architecture and not cut any corners whatsoever. That means professional n-tier design and not crappy, monolithic, web-based PHP/*SQL nonsense. Think J2EE but using entirely FOSS. Right now, the only project that is anywhere close to moving in the right direction is GNU Enterprise (www.gnuenterprise.org) but it needs a lot of development help before it can become usable. The other option is JBoss, an OSS J2EE solution, but it still requires the Sun JDK as the OSS JDK's are all way behind. J2EE is also massively complicated with a steep learning curve and kinda fails the KISS principle in my opinion.
I hear many people on slashdot harp on the benefits of OSS in one breath, and then complain about their jobs in the next. Wouldn't it make sense for these people to get a job writing software and getting paid for it, instead of writing high quality software for free and then earning money doing something they hate?
Absolutely! Once geeks realize that OSS is not just a toy but a means to make a good living, we will see the latent OSS revolution take place. For this to happen, however, we must truly collaborate and strike out on our own.
If you look around, most of the best OSS is that written by people who are being paid for their efforts one way or the other. Always remember this: OSS is not about a free lunch. It's about meeting needs in the most efficient way possible.
For core OSS business software, we need an organization like Apache that serves as a highly professional center of collaboration, funding, and outside contribution.
I used to be a CE major until I realized it was preparing me to become a Dilbert clone -- that is if I could get a job at all. My peers in the major were largely a bunch of dolts who were in it only because US News & World Report said that was where the good jobs were -- and not because of any passion for technology. I was the opposite. So what do people with a lot of passion who don't want to become corporate slaves in an unethical industry do? They start their own businesses. Sure, it's a risk, but the other option is a boring life if you're the type that likes to make decisions, do your own thing, change the rules, etc. My current passion is Open Source software and my profession is consulting. (Which is a great combination with virtually no overhead, btw. Yes, you too can be paid to write free software.) If I decide to do something else later on, there's nothing stopping me. I'm quite skilled in many fields and have enough business sense to not do anything stupid while trying. If I was ever to go back to school, it'd probably be in some cheesy business major just to get the degree. Yes, I say cheesy because compared to the rigorous hell that is an engineering / medical / science major, anything else is a piece of cake.
Intel doesn't want to risk being associated with these kinds of things (and you know if they released an open source driver, someone would).
Wrong. Their other non-Centrino wireless hardware is supported. As is that of just about every other manufacturer. Any manufacturer that makes claims similar to what you suggest is only throwing out BS to cover up their laziness / stupidity. But this is not why Intel is not documenting Centrino anyhow.
The volunteers who write and collect free software are not doing it for your sake. You are fortunate that you can benefit at all from their efforts; you have no standing to complain that they "waste" their time on what they see as interesting or worthwhile to do.
Wow, you are entirely missing my point. Of course people are free to do as they please and nobody is here to command them one way or other. On the other hand, in any community, there is wisdom in listening to others ideas before engaging in ones own actions. It's not a community if everyone does their own thing without any regard for a mutually beneficial outcome. There are good arguments why the endless creation of new Linux distros is deterimental to the community at large and wasteful of time for those who create them. Those people would be wise to listen to others thoughts on the matter before continuing. This has nothing to do with arrogance or whining on the part of those who make recommendations against actions that they personally see as unwise for the community as a whole.
The Open Source community has too many leaders and not enough followers.
Much of this regulation is acheived through the close-source drivers. Using modified drivers, it would be possible to make the card emit different frequencies or more power, thereby violating the usage licence.
Wrong. Any driver can be modified, whether closed source or not, to tweak the radio output. Closed source drivers are not at all how FCC regulation is achieved. This should be inherently obvious to the casual observer. In similar fashion, I could make my garage door opener fail to meet FCC regulation with some parts from radio shack. But if I do, I'm the one in violation, not the original manufacturer.
In reality what happens is the manufacturers document their wireless hardware and then the Linux/BSD wlan projects implement drivers. Yes, it just so happens that some hardware can be told through software to perform outside of FCC regulations. The default settings used by the open source drivers, of course, do not cause output violations.
Now, of course, the really interesting thing is when you get into designing your own antennas and tweaking the hardware to be just barely within legal limits. Similar to overclocking, a lot of wireless hardware has a significant "safety margin". Don't try this at home unless you know what you doing. (-:
It isn't a question of "we need", and it never has been. People create new Linux distributions for the same reason a lot of open-source software gets created -- because they want to. This is an obvious result of freedom: people can do what they want to, regardless of whether it is what anyone says "we" need.
While creating new Linux distributions is an exercise of freedom, it is unwise and wasteful. There is a huge difference between 'competing' software implementations (ex. Sendmail vs. Exim vs. Postfix vs. Qmail) and 'competing' Linux distributions. Writing new software ads diversity and brings about new ideas. In contrast, writing new distributions only duplicates the work of others. There is very little diversity and creativity involved. All Linux distributions are fundamentally identical. Sure, there are minor tweaks and different packaging and configuration tools, but there is no reason why these options cannot be rolled into the most popular distributions instead. The simple fact is that modern distros like Debian have plenty of flexibility to allow alternative components and experimentation.
However, I suspect that several factors may enter into the decision to make new software to accomplish the same goals as existing software: aesthetics, ethics, confidence, control of development, quality.
You are largely correct about these factors as applied to new software itself, but your point is invalid regarding the plethora of nearly-identical Linux distributions. I suspect that there is often something else at play: elitism. It's the "I'm too cool to use / contribute to a popular distribution, so I'm going to do my own thing instead and maybe get my name known too" This attitude is highly counterproductive in the open source community, where the tasks at hand are great and the available talent is currently limited. I say this as one who has done the whole "Linux from scratch" thing and realized that it was a total waste of my time beyond the first couple months as a learning tool.
In the commercial realm, this 'elitism' of sorts becomes the failed dot-com business model where you hope that branding alone will make you money. Plain and simple, there is little or no money in Linux distributions themselves. This is why RedHat discontinued its standard distribution to free up more resources for business solutions and services. Other aspiring Linux companies should take note and get behind existing popular distributions instead of doing their own thing. It's really better for everybody.
If the consumer really weren't benefitting, then they'd seek out alternatives: they do exist, as I'm sure you know. I've been using one for 9-1/2 years: Linux and associated alternative applications.
m l
You make a large and faulty assumption: that consumer behavior is a good barometer of whether the free market is working. The only reason why consumers are not more actively seeking alternatives to MS crap is that MS has used its monopoly power to marginalize or destroy all of its competitors. Open Source is the first real 'competitor' that MS has ever had. If OpenOffice had started 10 years ago, MS Office would be holding on by a thread today (if existing at all) and we wouldn't be having this discussion.
Of course, standards would get us to Nirvana just as market dominance by MS is and allow the tiny minority of us who use something other than Windows to make full use of our computing power under our chosen environments; but there's hardly an economic or public benefit argument to be made for having the guv'mint do our dirty work and go after MS just because they don't make the lives of (generously) 5% of computer users easier.
Open standards are a large part of the solution and they should have been enforced far earlier in the development of the industry. There is a very strong public benefit argument for going after MS, but only in the right way. The one good thing the government could have done in the anti-trust case was to force MS to turn over all proprietary data format documentation without royalties and nullify all of their related patents. But they didn't. Instead, they slapped them on the wrist with a modest (in M$ terms) fine.
the government doesn't need to pursue litigation, but only needs to state that they will purchase only software that stores and transmits data in royalty-free formats so alternative vendors can be used effectively in price negotiations.
This would be a step in the right direction, but it doesn't go far enough. MS would find a way around this requirement. They'd do something like make a government-only version of the software capable of exporting to a royalty-free format. And really, given the political clout that proprietary vendors have in the US, how likely is it that such a requirement will ever be made given the current circumstances.
In the end, there is one solution to the problem: elimination of the threat. Anyone who thinks that the proprietary software industry is capable of reforming and playing nice is deluding themselves. We will not see the end of this software patent / DRM / proprietary formats / etc. nonsense until Open Source software fully dominates the market. This is both a worthy and realistic goal. What have YOU done lately to help this happen? Or are you just complaining on Slashdot and doing nothing about it?
Here's somewhere to start:
http://www.openoffice.org/contributing.ht
Can't code? Send them some of the money you would have spent on Office licences. Or make an investment looking towards the future so that you can soon ditch Office and save a lot of money in the long term.
There's no such thing as a free lunch, but some lunches are much more efficient than others.
The thing that concerns me is that for some reason there's a mode of thought throughout most slashdot articles as of late (2-3 years) that linux should be as easy to use as windows. Do you really want this to be the case?
Yes, Linux absolutely should be available in a form that is as easy to use as Windows. What concerns me is that there is still such a holdout among elitists like yourself. The notion that allowing Linux to be newbie friendly will somehow make it worse for us hackers is downright asinine. I briefly held this belief myself before realizing how stupid it was. Every example you gave of the "horrors" of a newbie-friendly Linux is pure hyperbole and utter BS. (And frankly, they show that you have little clue what you're talking about.) The development of software and utilities that assist ordinary users will not in any way affect those of us who do most of our work in a bash shell and a vim session. To the contrary, it will help us immensely because more software will become available for our platform of choice and any kludgy configuration issues that exist today will be standardized by necessity as automation tools are developed.
Also, may I ask you a question? Do you ever plan to actually use the skills you infer to have to make a living? Well guess what? It's a no-go if ordinary people can't use the technology you are most adept at and if Linux doesn't drive a significant 'market-share'. So what is your solution? Let Microsoft and Apple have all the desktops and let Linux/BSD be niche players for the geeks only? Guess what? If that happens, we will have a world filled with DRM garbage and security nightmares. Get your head out of the sand.
...connectivity to a "mainframe" for financials, records, etc. which generally means some sort of TN5250/whatever emulation.
While this may be true of your shop, I've run across plenty that are using far more sophisticated and modern accounting / ERP / CRM / etc. software that is not so trivial to support. One of the key things that the Open Source movement needs is a solid foundation for creating this type of enterprise software. Sure, there are proprietary solutions already available for Linux. But the real revolution will occur when businesses see a reduced cost not only from what free OS's they can run, but what free *enterprise* software they can run on it. There are basically two options: J2EE or create our own solution. Unless Sun wakes up someday and realizes that there's no point in trying to keep the Java source and official documentation proprietary, we need to pursue the second option as a backup. (Incidentally, there are already JBoss and others on the open-source J2EE side, but there is virtually no OSS that actually uses J2EE and the JVM/JDK is still non-free as mentioned)
GNU Enterprise (www.gnue.org) is one probable option for an alternative, but it needs a lot more development help (and perhaps some funding) as the project seems largely stagnant at the moment.
Where's the money in all this? Consulting and services. Very few medium/large shops have their needs met by off-the-shelf business software. There is money in putting together custom solutions that work, whether you are using proprietary or Open Source software. The biggest difference is that you can better compete on cost with OSS because you don't have to bill your clients for exorbitant licensing fees. We need an Apache style model where consultants / IT departments collaboratively pool money into developing the base components of all major business software. After that, customization and support are the icing on the cake.
Among the predictions: no more rentals from video stores, no used games market, no lending games to friends, less upgradeable computers, pay-as-you-play software subscriptions, and other consumer-unfriendly changes.
If these predictions come true, then allow me to add one more: Open Source games take over the market.
In all seriousness, the community needs to write free, high-quality game engines upon which content and online services can be sold. And, of course, "the community" here largely represents companies who want to sell such content and services once the engines are collaboratively developed. Then we just need a non-proprietary console system.
"One Way Ticket to Mars" .. now that's a reality tv show that I might actually watch. (:
Unfortunatly, BSD/Apache style licenses are looking like more a viable option.
BSD-style licenses are ok for some things but very wrong for others. As a consultant who develops and supports free software business applications, there is an enormous advantage for me if other people who use my work and improve it must contribute the changes back. I've put an enormous amount of work into the foundation of my software and I don't want people freeloading off my efforts by selling proprietary derivatives with just a little bit of extra polish. GPL ensures my work stays free and that others play fair in the open source "code economy."
If I were the mplayer developers, I too would be quite outraged. If KISS is in violation, they owe back to the community any improvements that they have made or else they must license mplayer under a commercial license and thereby support the project in doing so. Those are quite fair rules to play by. Open Source Software == efficient development model and != free lunch.
Tell me who did real-time time shifting of TV shows (including watching the beginning of a show while the end of that same show is still recording) prior to TiVo. You couldn't do that with a VCR, and nobody was using PC's to time-shift at that time
Anybody with a video capture card could do that easily and video capture cards were available long before TiVo. There just wasn't a dumbed up user interface and it was on a computer instead of a set-top box.
If you don't like patents in general, then you can argue against it on that position, though TiVo would likely be out of business without it.
Perhaps the contention involves what a patent really is. Very rarely does a patent represent a truly new idea that nobody has thought of before. Even rarer was any true R&D required to produce the patented idea. (And in such case I believe the patent term should be significantly shorter..) "Intellectual property" is an oxymoron. On the other hand, there is perhaps a certain value in allowing people / companies to try marketing something new without being immediately bombarded with larger, perhaps better-known competitors who will surely put them out of business by duplicating their designs and selling quantity before they can become established.
For this purpose, TiVo perhaps ought to have exclusive right to sell time-shifting set-top DVRs within the limited definition of their patent. On the other hand, their patents should not be allowed to apply in any way to software itself or to general purpose "multimedia PCs" that just happen to have software that does what their DVR's do. (Whether this is the case I am not sure, but that's how it *should* be.) In short, I believe patents ought to be limited to very specific *physical* devices / inventions. From that perspective, their patent does not seem to be excessive. However, I could be wrong since IANAL.
The author was just using SCOs stock as an indicator that the company is not going to lose the court case like everyone else assumes will happen.
My opinion: If that's the case, the author is an idiot. This whole situation shows the signs of a textbook pump-and-dump manuever. The fact that Linux was chosen as a bogus legal target is fairly irrelevant other than the fact that MS may have been involved with encouraging that decision. IBM is far larger and more powerful than both SCO and Microsoft put together. If they thought they stood a chance of losing, they would have just bought SCO outright.
My prediction: In the end, this whole thing will backfire on the evil men who started this mess. Linux will be championed not only as a victor, but as an unstoppable force. SCO will wither and die for lack of a workable business model. MS will continue to lose the PR war against OSS.
Sidenote: It is not ethical to invest in an unethical company just because their sleazy tactics are causing a temporary stock rise.
So then the next suggestion above is to hire someone. With what money? And how can I justify spending ten times or more the cost of some proprietary software package hiring programmers to improve (or create) a free software competitor? Especially when my hypothetical freelance business probably isn't exactly rolling in the dough.
You are partially correct. OSS should not end up costing more than proprietary solutions. Personally hiring 3rd-party developers to improve the OSS you use rarely works.
This is why all the major OSS projects need to adopt means by which ordinary end users, such as the hypothetical graphics artist in this example, can donate reasonable amounts to respective projects. In return, they should get a say in prioritizing feature development. Considering how much graphics artists spend on proprietary packages, there is a lot of money out there. If a quarter of the people who use Photoshop skipped an upgrade and instead chipped in $100 to support the Gimp project, that would pay for a whole lot of developer brain-hours. Yes, it would be a longer-term investment, but certainly one that would pay off.
A more likely scenario is the graphics studio that finds Gimp to be, say, 99% of what they need and are willing to donate to the project what they would have spent on Photoshop in return for making Gimp 100% of what they need in the next 2 months. Then, 2 months later, another studio finds that the improved Gimp is now 99% of what they need.. and this cycle continues. In the real world, where not everyone is a developer, this is how collaboration in Open Source development must work.
It's as simple as this: If we want quality OSS, we need to financially support those who create it but are not financially rewarded elsewhere for their labors.
iTunes is a very simple piece of software and a successful marketing campaign. Just because it is (arguably) well done should not classify it as an innovative technology -- especially when it's not even a new idea.
I personally have no interest in using any of these flashy new online music stores. Until they are DRM-free, use an open protocol (ie. cross-platform), and offer lossless formats, just say no. In the meantime, support only independent artists.
Many companies lack the skills to maintain code--they simply don't have developers (or at least not the right sort of developer). To meaningfully contribute monetarily would erode at the cost savings. If the company is public, there may even be further complications.
Making meaningful monetary contributions makes sense if you get something in return. That's why OSS projects need to adopt various "feature adoption" schemes or other mechanisms to increase the incentive to donate to future development. Also, a "meaningful" amount is highly variable. If you have a popular piece of software receiving $100/year donations from thousands of companies, that is certainly meaningful.
There's also the "hire a consultant" option -- someone who can help maintain and improve the free software a business uses. Smart projects ought to start listing "registered consultants" who have proven themselves on the codebase. There is plenty of money in business software whether you are a consultant providing proprietary or OSS solutions. OSS just gives you an extra cost advantage over your proprietary consultant competitors who have to pass license fees on to their clients as part of the final software/service package.
Considering how many *useful* open source projects could use some funding, this is a really silly waste..
Now, if somebody would collect money to sponsor the completion of full SVG support in Mozilla, THAT is something I'd be more than willing to donate to. And it would help more than a tiny handful of people.
Open source is obviously faster/better/cheaper when 1000's of people donate their time to a single project. The only open source project I've been involved in was a collaboration among several corporations, all of which wanted to leverage each other's resources, but none of which could really contribute their own. ... There's nothing like money to motivate people to work on a project for which people aren't willing to donate their time.
This is definitely true. If you look around, you'll notice that most of the best Open Source projects are those where people are getting paid to contribute in some way. That's not to say that those same people would not have contributed otherwise, but money allows you to do things like drop your day-job and go full-time doing what you really love. The Open Source community needs to take a good hard look at how more experienced developers can be brought 'on-board' full time. OSS is beyond a hobby at this point. It's quite time to put that into clear perspective.
Open Source, at it's core, is about collaboration to meet needs efficiently. Part of that collaboration needs to involve paying developers so they can work full-time. Corporations who pool resources and collaborate on OSS projects to meet mutual needs are a perfect example. The same idea can work for individual users and smaller projects, however.
Take, as example, a typical desktop application like personal finance managers. We have GNUcash, which is a pretty good start, but it's missing a lot of the useful features found in the far more popular Quicken and Money. I personally have little interest in helping to developing GNUcash, though I wish it was a better fit for my needs. I'm not familiar with its codebase and I already spend most of my free time working on my own OSS project. (which I eventually plan to provide professional consulting services around..) However, I am willing to pay somebody $40 to develop a couple features I need in GNUcash. $40 is about how much I'd have to spend on Quicken or Money, which already meet my needs. But alas, $40 is not fair enough compensation for the developer. That's where collaboration comes in. There are millions of people who use personal finance software. If even 100 people contributed $40, that's $4000 compensation to add maybe one or two features -- easily doable in a month's time by an experienced developer. Realistically, there are far more than 100 GNUcash users able to contribute and far more than 1 or 2 features that need added. Once users start contributing financially to Open Source projects, allowing their developers to work full-time, we will see the true OSS revolution take place. The key is how to organize this process.
1.) Standards for simple everyday stuff While diversity in implementation is good, certain things simply need to be agreed upon community-wide:
2.) Enterprise database software This is the "killer-app" for nearly all businesses and should be the prime focus for those who want to see F/OSS on the corporate desktop. Trouble is, we don't have a good F/OSS enterprise-class application server! Forget cheesy PHP/MySQL apps. Forget Zope and other web application frameworks. That software has it's place, but it's not up to the task of hard-core database apps. We need something as powerful as J2EE, but without the steep learning curve, alphabet soup of acronyms, sketchy documentation, closed standards, and dependance on proprietary Sun code. Yes, I am aware of JBoss (open-source, non-official J2EE implementation), but it still has the issues listed above. Either a completely F/OSS JRE/JVM/JDK software stack needs to be written or else a completely new application server standard needs to be invented, perhaps using ObjectiveC or Python. GNU Enterprise (www.gnue.org) has a start, but that project seems to be stagnant and uninspired.
3.) More professional developers Simply put, the F/OSS of greatest quality is usually developed by people who are being paid to meet needs. There are plenty of viable business models and market opportunities. Hobby projects won't cut it. If you want F/OSS to succeed, you need to do your part in the marketplace.