Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×

Comment Re:...but if you want free software to improve... (Score 1) 1098

Your point about commercial software developers benefitting from BSD by building closed source software based on open source software is not convincing to me. If there were no GPL at all, and all open sourced software would be BSD licensed, how much open source software would there be for a commercial developer to benefit from in the first place? Not much I fear, if we were all commercial developers like that.

Here's a real example: libpng. OK, it's actually the "libpng license" rather than BSD, but it's close enough. It's certainly not a restrictive licence like GPL.

So, I might be writing a commercial image manipulation program, a photo sticher, a game, whatever. Now say I want to load an image. With libpng, any developer can leverage that code. As a developer focussing on my area of speciality that requires my specific domain experience, I'm able to get on with the problem at hand and just "load the image" and get on with doing whatever it was I was doing. If the libpng library didn't exist, I'd have to write something from scratch. In some cases, I might just restrict my program to just BMP or TIFF files because they're easier to load. Or, I could take the time to write a PNG loader myself. Something which would probably not implement everything in the spec because it's not my area of expertise. Now, I can fully understand that some people might hold the view "You're making a commercial product, why should you benefit from someone's free library? Write your own or buy one." That's a fair viewpoint, however the libpng guys did the nice thing and said "do with this what you will as long as you mention our work in the credits". Now, for some companies, even that isn't compatible with their business. They'll just write their own PNG loader. But for many others, adding an extra credit in the copyright notice isn't a big deal. libpng people get recognition. The commercial software has less bugs. The library is getting more use and bugs are more likely to be discovered. Commercial users will often report bugs and fixes. The user of the software is happy as it loads their PNGs perfectly. Everybody benefits.

Now consider the GPL world. There will almost certainly be a GPL solution. Probably other GPL software will use it. But it's fundamentally incompatble with any commercial project. The author has said "do what you want with this, except make money". That too is a fair enough position to hold if you feel you put lots of effort into something and don't want anyone freeloading on it. I get that. However, at the end of the day, your project being GPL isn't going to convince someone that the project they were writing commercially should be GPL so they can use your library to load an image. No, rather you're effective just saying "Haha! I've done all the work that'd be perfect for your current problem, and while I'll give it away, you can have it because I don't like other people making money on it." Again, if that's your feeling, go for it. But who really benefits from this? The author of the image library doesn't. The potential user of their library doesn't. Other GPL users of the library don't - the commercial users might have found bugs and contributed back fixes. The end user certainly doesn't. So again, who benefits? Only someone who thinks "Ha! Nobody can make money from MY work but me!"

Now sure, this is an extreme example, but scale it up to the next level. So, this time it's not just a image loader, but a whole GUI toolkit. But the same things apply, more usage of the library, more potential bug fixes, more exposure of the library to users. Scale it up way further. take sendmail. There are commercial versions of sendmail, llvm, or whatever. The free version is never going away though, so nobody has "lost" anything. The commercial one either has to employ people to merge the open source changes into their tree, or they take the easier route out and contribute patches. As soon as their changes are integrated to mainline, they have a lot less work AND everyone else benefits. If they don't keep up-to-date with the free one, people will stop buying it. If it has features that are worth money to people they will pay for them. If they have features that people want, but aren't particularly worth paying for, someone will eventually add it to the open source version if it's useful enough. Everybody still benefits. Nobody has actually lost anything. Even people who don't want to pay for a better version but don't want to take the source and contribute a fix themselves haven't actually lost out - although they can't use the better version for free, there's no guarantee if the software was GPLd that someone would add the feature. In fact arguably, the features in the commercial version can drive development of the free version.

The crux of the issue is that while developers CAN make closed versions of things, if they were just simple clones nobody would pay for them. They must be adding value somewhere that people suddenly want to pay for it. Whether that's tivoisation or a boot loader on a phone, it's not the open source software that people are buying the thing for. It's for the value added - ability to record TV or make calls or whatever else. The original is still there. Everyone can still use it. But if you want the value added stuff, you pay the market rate for the extra value unless they decide to give back to the community.

At any point, it's the original developer's choice whether they want to allow others to benefit from their work in any way they like or whether they want to restrict others who might be able to get their good idea out in the open a bit quicker by using your stuff. I've see plenty of licenses back in the day that say "no military uses" or whatever, and if that's something a developer believes strongly in, then that's up to them. What's more, as someone who actually believes in freedom, I don't really care what developers do with their own projects.

However, it is disingenuous to say that GPL gives more freedoms over BSD and it really does annoy me when people claim that GPL is more free than BSD, because it's whole reason for existance is to restrict what other developers can do with it. What's worse, there are a lot of people who'd happily give their things away without restriction, except that GPL gets so much press that people associate it with open source that a lot of people don't realise what an insipid viral licence it is. Which is a shame, because it means that there are so many things that are implemented over and over and over and over that it must amount to an awful lot of wasted time for the computing industry worldwide.

When I'm at work, I have no choice. There is no option to release the source because in most situations it's the only way to recoup the costs of several thousand many years of development. That is what it is. But for all my personal source, if I'm giving something away, I'll just give it away no strings attached. BSD-style or true public domain is really the only option. It doesn't hurt me if someone uses my old projects to help them do something I'd never even thought of. Why should it? I've moved on to something else already!

Comment Re:...but if you want free software to improve... (Score 1) 1098

Analogy time: which society is more free, one where I can punch you in the face, or one where you're free from punches in the face? See how that works? It's, unfortunately, not possible to have your cake and eat it too. We can't have a society where you can punch people in the face while also being free from punches in the face.

The society where you can punch me in the face is more free. The other society places restrictions on what you can do. Of course, you are free to live in the society where you can punch me in the face but choose not to and be reasonably confident that I as a similarly nice person would choose not to punch you in the face. We're both totally free and we're both happy.

In a sense, this is a very similar issue. By allowing [near] total freedom on the first round of licensing (as with BSD), we're limiting (or at least allowing the limiting of) the freedom of subsequent licensees. By limiting freedom on the first round of licensing (as with GPL), we're maximizing (or seeking to maximize) the freedom of subsequent licensees. At this point, there's nothing left to do but choose which license is most compatible with your own subjective views on freedom.

But really, who cares about subsequent licensees? If the original product is so great, everyone is still free to use that. Their rights have not diminished in the slightest just because there is an additional option available to them.

If someone makes a trivial change and sells it as a closed-source package, it'll be trivial for someone to make the same trivial change and release it for the open source community. There's no reason for anyone to buy the closed-source variant other than convenience.

If someome makes a non-trivial change, why should they also have to make that non-trivial change available for free? Just because the original author didn't value their time or could afford to release the software without getting paid, it doesn't follow that other people doing non-trivial work are in the same position.

At the end of the day, it boils down to whether you want to give your work away, or give it away with strings attached.

Comment Re:...but if you want free software to improve... (Score 1) 1098

The issue is that "directly derived work" means "anything that includes any insignificant spec of GPL gets completely tarnished with it".

I'd never visited this site before, in fact only just now found it by clicking on the top google result. However, consider the latest "GPL violation": http://gpl-violations.org/news...

So, this is a result for user freedom? A big media player project was found guilty of a GPL violation for including iptables - and not even an intentional one because their suppliers had included it in some outsource work. Now, iptables isn't related to the core product, in fact there's no good reason for the media player to even need it in the first place. There's no point having a firewall on a device where there should only be a couple of ports open in the first place. Calling the entire media player a derivative work is totally disingenuous as it doesn't define the product. The inclusion of iptables has no tangible benefit to the user, the media player itself does.

Sure, they violated the GPL. And they (or rather their suppliers) really shouldn't have done that - they were making a commercial product and should have known the GPL wasn't compatible with this aim. But the lesson to be learned from this is "don't allow your company to touch GPL software with a bargepole". More than that, it underscores exactly how the GPL isn't free.

Comment Re:...but if you want free software to improve... (Score 1) 1098

To vastly simplify it... BSD emphasizes freedom for developers. GPL emphasizes freedom for end-users. Different goals, and impossible to ever say which is "more free", since it depends heavily on the context.

Or alternatively, BSD emphasizes freedom for everybody. GPL emphasizes freedom for end-users by attempting to ensure that any derivative works are also free. The real world effect of course is that people writing commercial software still write the same commercial software, but can't use anything involving GPL in those products. So, even if there's a commonly used tool that does most of the job, they have to reinvent the wheel which means that they waste more time doing that then adding value to customer (or have to charge the customer more to cover the increased development time).

You can just as easily point to instances where companies have taken BSD sources, closed them off, and sold them, saying that now the end-users have less freedom than they would with GPL sources.

The users have exactly the same freedoms as before PLUS the ability to buy a product that might better serve their needs. The end-users are still just as free to use the original software as they were before.

Comment Re:...but if you want free software to improve... (Score 3, Interesting) 1098

I know it's bad form to follow up on my own post, but I forgot to mention the Linux Action Show. If you really want to understand rms and his stance on GPL, you really must watch this: http://lunduke.com/2012/03/11/... which links to the video http://www.youtube.com/watch?v...

Essentially, rms believes that anybody who makes money from "non-free" software is "evil". That is, writing software is fundamentally incompatible with earning a living. Listen from 57:45 if you don't believe me! He gets even more radical at 59:00. "If it's not free software, I don't think you're making a positive contribution to society."

And remember the whole while, he's advocating a "less free" license in GPL. He wants to restrict the freedom of other developers because anything they do that might not be free software isn't a valuable contribution to society.

So, a plea to all the GPL advocates. Is that really your stance? Is that really what you believe? Seriously consider the implications of holding a worldview where you believe that anyone whose talent is writing code shouldn't be able to make money from that but instead have to find some other job and "maybe" fit in writing free software into their life as a hobby. Is that seriously beneficial to the field of computer science? Well, hope you enjoy your job in McDonalds!

Comment Re:...but if you want free software to improve... (Score 4, Insightful) 1098

Someone building on top of a BSD-licensed software project has the additional freedom to retain and not release changes they make to that software when distributing their own build. GPL advocates say that this is a freedom that people shouldn't have, in order for all players to be even.

Yes, this is exactly the issue. GPL isn't "more free" than BSD. Quite the opposite. GPL is far less free as it grants the users less freedoms.

The BSD approach is "Here is something nice I made - have it and do what you like, hope you have fun!"
The GPL approach is "Here is something nice I made - you can use it, but if you you have to let me play with you stuff. I don't care that your thing might be vastly better or more complicated than mine, if you're using my stuff you sure better make sure I can use everything you make."

Which is really more free?

Comment Re:Ignorant to their own research (Score 2) 444

A slightly more cynical person might think that by releasing these statistics, people might be less inclined to buy Seagate drives and thus the price they can negotiate becomes even lower when retailers are left with drives that don't shift as well. As the article says, "[they] buy drives the way you and I do: they get the cheapest consumer-grade drives that will work."

Comment Re:Verilog (Score 5, Informative) 365

The number of slices or logic cells or whatever else a particular synthesis program for a particular chip generates doesn't exactly correspond to a number of gates either. For instance, a single 4-in 1-out LUT on a Xilinx can be used for 1 gate or 6.

I wouldn't have much confidence in automatic C to HDL conversion either. Good HDL design is about understanding the problem in terms of gates and parallelism. FPGAs and ASICs in general aren't particularly good at things that CPUs are good for, and inversely CPUs aren't especially good for things that FPGAs and ASICs can do well.

The OP shows such a lack of understanding of hardware design that it's not funny! "Add = 3 gates, Divide = 6 gates" is quite comical to anyone who actually knows these things. A more ball park is that an n-bit add can be done with 2n LUTs, in terms of gates it's about 5n gates, but really that depends what gates you have available. A multiplier is massively more, dividing is even more complicated still. Fortunately, many FPGAs come with a few dedicator multipliers... Unless your algorithm requires only as many multipliers as you have available, you're probably best building a state machine and multiplexing a single multiplier unit, in much the same way as a CPU multiplexes the ALU at its core.

The whole thing is massively dependent on algorithm and experience of the person doing the porting. The best advice is to say "I don't know" or to hire someone who does or suggest them running the algorithm on an embedded CPU.

Comment Re:Startups should not be at CES, or CeBIT (Score 1) 89

That's a bit of a jump, there's no reason to suspect that startup companies turning up to CES aren't quite far along in development. Otherwise they'd not exactly have anything to show.

But yeah, it's pretty obvious to anybody that a massive trade show isn't the best way to get exposure. Getting key evangelists on board with your concept will get you far more exposure in the long run and for a lot less money than a booth at a trade show.

Every company I've ever worked at that's done the trade fair thing has done so simply because they're already high up in their game and they have to go to maintain their image, otherwise everyone just assumes they're not doing as well any more if they don't show or have a small booth. But it's not really about publicity - I know when I've been to the show myself, there's basically loads of people wandering aroudn just trying to get as much free stuff as they can. Possibly only 1% of people you turn up to your booth actually wants to know about your product and most of them won't actually generate any business.

Comment Re:or? (Score 2) 312

Yes, he's wrong. If you have a driving licence, you automatically "have" a provisional driving licence, which means you can drive a moped without supervision. For about 10 years, to drive a motorbike you still have to pass the CBT (cumpolsory basic training) test to be allowed to ride a low power motorbike. From passing the CBT, you have 2 years to take a proper bike test or you have to redo the CBT.

Comment Re:Does it matter (Score 5, Insightful) 207

Nonsense. People in general don't care about in privacy, right up to the point where it suddenly works against them. It's just the laziness and apatheticness of human nature. It'll take a lot more than these leaks before people are really enraged, because at the moment everyone is still happy that "it's catching terrorists".

It's just another example of "First they came..." - http://en.wikipedia.org/wiki/First_they_came_...

Comment Re:Nuber not that impressive (Score 1) 304

Actually, I live in the UK and what little I know about US wages comes from friends working in tech jobs in CA. But regardless of the actual figures, even if the average salary is double or half of what I said, the point was that 12 years in prison is an unfair punishment compared to the actual crime.

Slashdot Top Deals

Two can Live as Cheaply as One for Half as Long. -- Howard Kandel

Working...