4 ms·
That is misrepresentation, it was never not held to high standards. There were bugs, one serious while much fever was made over less critical corner cases i.e.
by _zrlg 6y ago
That is misrepresentation, it was never not held to high standards. There were bugs, one serious while much fever was made over less critical corner cases i.e. the pfsense release that has been pilloried by Salter by way of Jason doesn't use jails. There are others like jumbo frames that are not common on internet outside of certain high end carriers.. less common but valid bug, good for new contributors. Most were personal preference: the pseudo driver framework, malloc style etc. There were fixes by multiple people in progress, including two well respected developers at Netgate. The communications were not handled in good faith by the rewrite party by their own admissions, so a rational discussion to fix or to disable the code ahead of 13.0 release in the stable branches was not able to be had with Netgate or by any of dozens of other mature FreeBSD developers. Instead drama was created. I am quite certain the right call would have been made without all of the bad blood spilled had this been conducted respectfully by all parties from the outset. So while new tooling and re-commitment to review process is nice, the processes FreeBSD had in place were already in action and it is sad to not see core assert this.
There is also a layer 8 and 9 vulnerability. One guy now tightly controls a protocol in several free *nix kernels and has interesting reactions whenever anything happens without his blessing. He was able to cause a disproportionate reaction by talking to a journalist. This probably doesn't matter if you are encrypting your home PC traffic but it does to people who work in the Internet industry. Say whatever you want about the particular technology and particular individuals, it boils down to whether you think the desire for control of implementation is a weird situation or not.