3 ms·

That statement was what caught my interest in this piece. What is the "much evidence" which the author refers to? I'm genuinely interested; if anyone here know

That statement was what caught my interest in this piece. What is the "much evidence" which the author refers to? I'm genuinely interested; if anyone here knows of any such evidence, please, tell us! What is it? My personal experience is the exact opposite; I have not personally found closed-source libraries to offer better quality in general. Of course, there is a lot of junk out there which is open source. But most of the small number of rock-solid, tried-tested-and-true code libraries which I am aware of are open source.

imtringued 6y ago
I have seen lots of "evidence" that closed source companies will sweep all problems under the rug or even sue you if you find a vulnerability. Meanwhile maintainers of floss software will just fix the vulnerability, often within a few days if you already spotted the relevant part of the source code. My worst experience has been with a proprietary rack mount appliance. The closed source software was absolute garbage. It wasn't even a library, it's even more fundamental. It was network equipment, you know, a problem domain that didn't change significantly in the last 20 years. Crap closed source software exists, it just has a harder time propagating because it fails to bring in money. Meanwhile any random FOSS project can last a human lifetime, assuming it is archived properly.
mindslight 6y ago
I think low quality code is probably correlated with open source, but the causality runs in the other direction. It's easy to write garbage and throw it up on Github as your only deliverable, whereas the minimum deliverable for closed source is a functional binary. Which is why FUD like this is so dangerous. It makes sense in PHB land where they don't understand that a programmer doesn't just use the first hit for a given functionality, but investigates whether it's actually worthwhile to use.
- alexdowad 6y ago
  Hmm. This makes sense, but I wonder if it would be more accurate to say the quality of open source code is more variable than closed source. If you filter out the worst junk, would low quality code still correlate with open source? Just thinking out loud. Anyways, I think you may have identified what the author of the piece had in mind.

imtringued 6y ago

I have seen lots of "evidence" that closed source companies will sweep all problems under the rug or even sue you if you find a vulnerability. Meanwhile maintainers of floss software will just fix the vulnerability, often within a few days if you already spotted the relevant part of the source code. My worst experience has been with a proprietary rack mount appliance. The closed source software was absolute garbage. It wasn't even a library, it's even more fundamental. It was network equipment, you know, a problem domain that didn't change significantly in the last 20 years. Crap closed source software exists, it just has a harder time propagating because it fails to bring in money. Meanwhile any random FOSS project can last a human lifetime, assuming it is archived properly.

mindslight 6y ago

I think low quality code is probably correlated with open source, but the causality runs in the other direction. It's easy to write garbage and throw it up on Github as your only deliverable, whereas the minimum deliverable for closed source is a functional binary. Which is why FUD like this is so dangerous. It makes sense in PHB land where they don't understand that a programmer doesn't just use the first hit for a given functionality, but investigates whether it's actually worthwhile to use.

alexdowad 6y ago
Hmm. This makes sense, but I wonder if it would be more accurate to say the quality of open source code is more variable than closed source. If you filter out the worst junk, would low quality code still correlate with open source? Just thinking out loud. Anyways, I think you may have identified what the author of the piece had in mind.