Robert McLaws: Windows Edition

Blogging about Windows since before Vista became a bad word

Open Source = Bad Code? Say It Ain't So!

Open Source programer Harald Welte enforces GPL provisions for a living. So you'd think he'd walk the party line about code quality, right? Heh. CNET reports:

"If somebody asks me how much the actual free software source code benefits from the code that was released by the vendors, my honest reply would be simple and sad: None," Welte said in his blog Monday.

"The code quality is usually extremely bad. Looking at kernel patches from the various vendors, I'd say the code quality is, by far, off any scale that would ever even remotely be considered to be suitable for upstream inclusion" in the code base of mainstream projects such as Linux, he said. The commercial programmers spend no time making sure code will be portable to different varieties of processors, including 32-bit and 64-bit chips, will work on multiprocessor machines.

"This code is 'throw-away software,'" Welte said. "I would be the most embarrassed man if I ever was involved with any such software. Having your name associated with such poor quality would be like a stigma. Any technical person would laugh. And yet, the managers of those respective companies proudly announce the availability of their so-called 'GPL code releases.'"

But wait a second. I though Open Source was supposed to lead to better software. You mean the code quality is just as bad as closed-source companies? But I... what the... huh?!?

I need to go lie down. I can't keep this stuff straight anymore.



  • > I though Open Source was supposed to lead to better software

    Your point exactly? Do you know the broken-window principle? If there is nobody responsible for fixing the broken windows soon more windows will get broken. Open Source leads indeed to better software but it is not because it is open source that there shouldn't be code reviews.

    October 31, 2006 4:56 PM
  • You think code reviews would have changed his mind? The broken window principle can only go so far in explaining this. You think a code review would have solved the problem? If a company is so poor that it has to huse free code to get it's own product out the door, what makes you think they care to spend enough time to do code reviews? The people writing in the first place don't care enough about their own code to not write crap, so what incentive does the reviewer have to make it any better?

    The argument that "open source leads to better software" is an opinion that has zero basis in fact. Firefox is no more secure than IE, Linux is no more secure than Windows. Those platforms are not attacked as often because it's not worth the investment to create an attack if it's only going to affect a small number of people.

    I have my own theory: "If a problem is left up to anyone to solve, few will care to try". Why? Because fixing it would take effort, and most people are lazy. Software designed by leaderless commitee will always have problems, because there is no one to take responsibility for the problems, or rally the troops to fix it.

    October 31, 2006 5:59 PM
  • Dario Solera said:

    I don't think that open-source software is "automatically" better than close-source software, but for big projects, the community usually improve the code in some way, reporting bugs and writing patches (in my experience).

    Bugs are everywhere, and IF there are many programmers looking at the code, the probability to find an error is higher than usual.

    November 1, 2006 3:35 AM
  • James said:

    I'm not certain that the post is saying what you think it's saying: my reading is that free software has higher standards than vendors. Otherwise, vendor-released code would benefit free software, right?

    It's the vendor-released code that has poor quality. Code similar to that which we use as consumers of closed-source software. Anyone who has ever downloaded a Creative Labs driver knows what Welte is talking about.

    Anyone can release source code, and anyone claiming that open-source code is automatically better quality than closed source is obviously hepped-up on goofballs. The (open) question is whether popular projects that solicit external investment tend to have better code.

    Let's not fight FUD with FUD, okay?

    November 1, 2006 8:21 AM
  • List244 said:

    Well, open source applications definitely have something going for them. Like Dario said, it does increase the amount of viewers and lead to finding problems. Also, it is open to the hackers, which can increase the speed of attacks and also the speed of patches. Though, at the same time, it drops security in the sense that holes are able to be seen. Also, if code is written sloppily, finding these bugs can be even more difficult. Also, fixing a bug could require massive rewrites where cleaner code might have required much less change.

    It is true though, Linux, Max, Firefox, and others are attacked less, because there is less reason to attack. You don't bomb the little guy, unless the little guy has done something to hurt you. If you want to do harm, you attack the big guy. I find it funny, all the people saying Mac and Linux are so much greater, so much less vulnerable. The fact is, if someone wants to attack them, they can. If another OS becomes the mainstream OS, people will see, it isn't just Microsoft.

    November 1, 2006 9:13 AM
  • FatCat said:

    Totally agree with List244's point of view.  

    November 6, 2006 11:53 AM