Get it through your fucking heads. Propriety software is just better than open source software. Just look at MATLAB vs Octave if you don't believe me. MATLAB destroys Octave in performance. Always. It has all sorts of libraries that Octave will probably never have. It has all sorts of features like a compiler that Octave probably will never have. The IDE is actually useful. There is no comparing the two. But Octave is made by people who aren't being paid to do it. They have fewer people working on it and the people they do have are less motivated and organized.
Free software is only free in the sense that the time spent working on it is worthless.
Name:
Anonymous2013-01-12 2:39
>>11
If I were to pull two programs off the top of my head (GCC and Visual Studio) then compare them based on a criterion I obtained in the same manner (how well they implement the C standard) obviously you'd know which one would be 'better' in that regard. Since GCC is free and Visual Studio is proprietary, then based on your logic, it would follow that all free software is better than all proprietary software.
Unless you can actually say something, directly, about the ethical issues involved with free/proprietary software, instead of whining about how you prefer the features of the programs (which is clearly unrelated to how the programs are licensed) then you're just speaking shit.
>>17
I use bc too, but only as a basic calculator. If I need to solve a partial differential equation or simply to multiply huge matrices, it doesn't even cross my mind. Those are the job of maxima and octave.
>>20
gnu bc sucks. I use Freebsd bc: http://www.freebsd.org/cgi/man.cgi?query=bc&apropos=0&sektion=0&manpath=FreeBSD+9.0-RELEASE&arch=default&format=html
And it runs a tad faster than awk when I make determined functions and differentials. So on writing a quick hdf5 formatter should not be that hard. I still use BerkelyDB because we have lots of people using the same arrays values at any given time to apply for other application. I love doing the programming, but hate doing the reports. Any other running functions, I use ATS to stop at a point and publish results when by boss asks me for some, but I still let it run.
>>20
If you mean graph plotting, as fast as your CPU can make FLOPS + 100-200~ (to run the dc parser). Plotting is just for and while loops. Most of the time I do for loops nested on while loops. Hehehe. And publishing on my comp I use grn and grap (from troff). We have other daemons that publish the graphs in stupid .pdf, .png, and javascript & <canvas> versions from the BerkelyDB keys/tables. The shitty svg is still in the works.
Name:
Anonymous2013-01-12 7:44
>>1
Octave is great. It is easily extended -it has a good FFI- and the language seems to be more sane than matlab. I think they almost have the same functionality. At least I haven't found any significant differences.
If you need a more symbolic computation engine, you can try maxima.
What it lacks is good support for visualisations. That sucks a bit. You cannot simulate a 3dimensional primon gas for example in a visual way. At least not very fast. It also doesn't have simulink.
>>19
That's awesome. I was actually looking for something like logic.bc.
Name:
Anonymous2013-01-12 22:56
>>12
gcc is bloated. You pick software from one of the largest open source projects on the planet and compare it to the pile of shit that is TurboC. And even then, gcc isn't the best.
You intentionally misinterpreted my argument. My argument is that propriety software, in general, have a more motivated and organized team working on it which leads to better software.
Name:
Anonymous2013-01-13 6:15
>>26 >>12-san showed a contradiction to your claim. Your criticism of gcc in comparison to turbo c is subjective at best, and anti fan boyism at worst. You then supported your original argument by restating it. But the counter example provided in >>12 still applies. Furthermore, the new statements you have provided on your own is also false in general. Not every propriety project has an adequate team, organization, or allocated time. The amount of effort varies with every programming project, regardless if it is open or closed source. I could compare the linux kernel to some shitty business app written by 2 people in 1 week. Or I could compare a terrain data collection system created by the military to some crappy open source game made by a high school student. The comparisons are meaningless.
Name:
Anonymous2013-01-13 9:19
>>26
GCC is the best. LLVM doesn't support the range of computers that GCC does. Clang isn't as feature complete as G++. The wide scope of features means that GCC won't produce optimal code compared to a focused compiler to a limited platform like Microsoft's compilers.
Proprietary software writers are not more motivated than open source writers. They may be better organized in general, but they don't have greater motivation or perform better (write maintainable and readable code that achieves the specs within a given time frame).
Name:
Anonymous2013-01-13 14:02
>>26 gcc is bloated
If you're interested in a lightweight C implementation, look at TinyCC.
You pick software from one of the largest open source projects on the planet and compare it to the pile of shit that is TurboC.
I compared GCC with Visual Studio, which is a commonly used, modern implementation of C (unlike Turbo C, which is designed for older platforms). You seem to be missing the point of what I wrote, though. The fact that you think I suggested Turbo C makes it seem like you didn't really read my post, and hence, didn't realise that I was just using that comparison to demonstrate a larger point.
My argument is that propriety software, in general, have a more motivated and organized team working on it which leads to better software.
Yet you don't provide anything to back that up. Why would ethical issues that determine how a program is licensed have any significant effect on the quality of the code that is written? The skill of the people writing the software, and their dedication to it, I would think, are much larger factors.
Name:
Anonymous2013-01-13 14:14
The wide scope of features means that GCC won't produce optimal code compared to a focused compiler to a limited platform like Microsoft's compilers.
The reason for this is that it's easy to write a half-assed compiler that kind-of works on your platform in your spare time, but it takes a lot of skill and dedication to make a truly great compiler that works very well on a popular platform like Visual C++. You need someone to foot the bill in order to do great work. You either get your money from a University or from a commercial software company. When the money comes from a University, you may get some interesting ideas implemented, but there is no feedback from a market to indicate how useful the software is. So school funded free software ends up being a muddled mess that is of little use to most people. You need to get a commercial company involved to shape that into something people might want to use.
The only environment where there is any chance that you are forced to dedicate yourself to make something good that people want to use, and is not just a bunch of technological masturbation, is in a commercial software company. This isn't true for all software companies. Only that, in the few places they exist, they are always inside a software company.
They may be better organized in general, but they don't have greater motivation or perform better (write maintainable and readable code that achieves the specs within a given time frame).
So either the economic and social lessons demonstrated by the 20th century are wrong: wealth doesn't motivate people to better themselves; the USSR was a communist heaven that did not crumble into economic ruin. Or there is a tiny, but very obnoxiously loud, group of freetards on the internet hyping the stupid shit they work on in their spare time. A group of people trying to boost their over-inflated egos with the software they like to use.
Name:
Anonymous2013-01-13 14:36
>>30
You're talking about something unrelated to this discussion. There are free programs that are developed by commercial companies, and there are proprietary programs that aren't. Do you understand that 'free' has different meanings depending on the context in which it's used?
Name:
Anonymous2013-01-13 14:42
When the money comes from a University, you may get some interesting ideas implemented, but there is no feedback from a market to indicate how useful the software is.
There may not be a "market" but I've worked on University (grant) funded projects and there most certainly is feedback from the userbase.
Name:
Anonymous2013-01-13 15:47
>>29 Why would ethical issues that determine how a program is licensed have any significant effect on the quality of the code that is written?
These issues do not determine the quality of the code. I never claimed this.
The skill of the people writing the software, and their dedication to it, I would think, are much larger factors.
Indeed. But how does one go about getting skilled and dedicated programmers? In an open source project, your ability to bring in talent is crippled. You can only hope that the project you are working on is so admirable that it makes people want to work on it very hard for free.
These issues can be solved by selling your software. If you are working for a company that makes money by selling software, they can offer you incentive to work on their code. This company can also decide to only hire the best of programmers. Unlike open source projects, who take pretty much whoever they can, companies make sure to have some sort of standards.
And onto the dedication. While someone who is dedicated to an open source project may put in a couple hours a day at most into the project, at a software company, you are obligated to work for at least 8 hours on the code.
Propriety software is a better model for good software.
Name:
Anonymous2013-01-13 16:41
>>33
See >>31. Free software simply guarantees you the four freedoms that are defined by the FSF and proprietary software doesn't. The strategies used for development, funding, etc, aren't implied with either of those.
As for development models, I like the ones where I'm the only one working on the code. I dislike working with other people entirely, so neither the 'open' type where you have a bunch of illiterate critters submitting dirty, broken code to you, nor the 'closed' type where you're selecting a few humans and treating them like slaves seem appealing to me. Although, since both of those visualisations are obviously exaggerated, I probably wouldn't mind joining a project that used either of them, so long as a) I'm not using them for my own projects, and b) the code I write isn't violating any of the four freedoms defined by the FSF.
Oh, and to comment to some of the things you said:
There are times where I've put in around 12 hours or so on my own projects, so that's not really true, and I do it just because it's fun. When you enjoy what you're working on, you don't really care that you're not getting paid (but I guess it's still cool if you do get paid). It's only when you don't enjoy what you're working on that you're not going to work on the project — unless you're motivated through other means.
At least that's how I find my body works, and I write shitty software when I don't find what I'm doing fun, so it's not really advantageous in my opinion. You just need to find programmers who enjoy what they do.
Name:
Anonymous2013-01-13 17:07
>>30,33
Inconsistency detected! You called GCC bloated and Visual C++ truly great! What distinguishes bloatedness from greatness?
It's absolutely true that compensating programmers will attract more time and talent. There are software companies that develop open source software though, and they make money by receiving funding from various sources, because there are people that want their software to be developed. In this environment, employees can devote a full work day. But even in other circumstances, it isn't a weakness to only work for a couple hours a day on something for an extended amount of time. It is suboptimal for turning out a project within a short time frame, but it works fine for creating a product over a long period of time, and continuously making improvements to it as it is tested by the public.
Name:
Anonymous2013-01-13 17:09
Whoops, forgot the quote:
While someone who is dedicated to an open source project may put in a couple hours a day at most into the project
>>40
I think it's true, when compared being payed nothing. This may surprise you but there are people that are talented and also motivated by money to some degree.
I think it's true
It is true. Which is why propriety software > open source child's play
Name:
Anonymous2013-01-15 1:30
>>40
Nope. You're just too retarded to use the correct terminology, so you confused the entire argument. Learn what “proprietary software” and “free software” actually mean; they don't refer to development practices, but ethical issues. Until you manage to become less retarded, please redirect your output to the nearest toilet you can find.
>>42
While it's more rare, there are companies that pay employees to produce free software. That was the point I was making. See >>43.
Name:
★ RESU PIV ★ VIPPER2013-01-15 2:12
HAX MY ANUS
Name:
Anonymous2013-01-15 2:28
>>16
Not trying to be a dick here but MATLAB's syntax is insanely simple. Only thing that was annoying as fuck for me was the inability to define new functions from within your script