On Mon, Aug 9, 2021 at 2:46 PM Tom Herbert <tom@xxxxxxxxxxxxxxx> wrote:
On Mon, Aug 9, 2021 at 11:42 AM Vasilenko Eduard
<vasilenko.eduard@xxxxxxxxxx> wrote:
>
> Hi Tom,
> The problem may be overlooked for years
> Because not many people use Linux as a desktop with a browser.
> And even after this RTO should happen that is not mandatory for the majority of connections - the network should be overloaded for this.
> It is more important for this matter what is implemented on Windows and Android.
Eduard,
Android runs Linux. It's likely these patches are deployed on over a
billion devices at this time.
Tom
Possibly. Does Android run the vanilla Linux IPv6 stack or something rather smaller of their own devising.
Thing is that Linux tends to be latched on to in IETF terms as being the vanguard of new protocol deployment because patches get pushed out quickly. Meanwhile macOS and Windows frequently lag in feature deployment by up to a decade.
While the assumption is commonly made here that this behavior is laziness and it is the cost of implementation that causes deployment to lag, my conversations with engineers and very senior managers tells me that in some cases the lag is because their evaluation of the feature leads them to consider it should not be deployed at all.
So no, I would not assume that features from Linux automatically make their way into anything. Not without actually looking at the running code.
