Research proves what open source folks have been saying all along: “An analysis of the 5.7 million lines of Linux source code shows that it contains fewer bugs per thousand lines of code than commercial counterparts.”
Research proves what open source folks have been saying all along: “An analysis of the 5.7 million lines of Linux source code shows that it contains fewer bugs per thousand lines of code than commercial counterparts.”
It’s hard to extend “Fewer bugs in Linux” to “Fewer bugs in open source”.
It’d be interesting to see if the study could expand to include information such how many developers are involved in the Linux project as opposed to MS Windows. And if the intent is to extend the research to open source in general, is there a tipping point? How many developers are required to achieve an optimum ratio of bugs per thousand lines of code. Does the number of bugs increase as the number of lines increase? Would increasing the number of lines of code in Linux by five fold result in a 10 fold increase in bugs?
I’m sure there is tons of horribly buggy open source software out there, hell I’ve probably written some of it, but it’s a validation of the model. Put this against the backdrop of people telling you (like they tell me) that open source is a dead-end model that can’t sustain itself and ultimately produces a lower-quality product than its commercial counterparts, then it has a different light.
was this against OSX?
Well, the argument that says open source produces lower-quality products has the same lack of evidence as the opposite argument. I just don’t think there’s evidence of a correlation between open source and less buggy code.
In the end, this article is just a validation of Linux. Which I think was already validated by the millions of people who use and develop for Linux.
I think the idea is that, as linux is the most developed Open Source product, it shows that the model can work.. and work well.
Mozilla shows another example of this I would argue.
Overall tho – its easier to get mostly, or partially, right fast with a ton of money, with Open Source you’ll probably have a better product in the long term IF you get that far. At least thats my perspective on it.
That’s true; I haven’t come across a study yet that proved all open source products are less buggy. I do think, though, that by their very nature they tend to be less buggy: code that is freely available to anyone and everyone is likely to be examined by many more people than code that is only available to those people who work at the right company. As a result, more people can point out as well as contribute fixes for bugs.
Sorry if this seems to be a little rambling, I just wrote whatever came to me. 🙂
Test.
I find it most interesting (or possibly annoying) that there must exist the argument of open-source vs. closed-source.
Is it not true that there is really good and innovative software products coming from both sectors?
I personally feel that both will ultimately be used together.
Also, to make the stereotype that if one company’s software is buggy or not, that this means all similar products are or aren’t, is ignorant to me.