<div dir="ltr"><div><div>esr: kill it<br><br></div>nigel: thank you very much<br><br></div>..m<br><br></div><br><div class="gmail_quote"><div dir="ltr">On Thu, Jun 16, 2016 at 12:28 PM Eric S. Raymond <<a href="mailto:esr@thyrsus.com">esr@thyrsus.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Putting this on the record on the devel list...<br>
<br>
Nigel Roles <<a href="mailto:nigel@9fs.org" target="_blank">nigel@9fs.org</a>>:<br>
> On 16/06/2016 12:10, Eric S. Raymond wrote:<br>
> >You're listed as the last person to touch the Arcron clock driver in NTP.<br>
> ><br>
> >Does the hardware for which this driver was made still exist? What,<br>
> >if anything, can you tell me about it?<br>
> ><br>
> >If this driver is a fossil I want to remove it.<br>
><br>
> Eric,<br>
><br>
> There may be hardware around, but you can't buy it, and mine long ago went<br>
> to land-fill.<br>
><br>
> I say remove it.<br>
><br>
> Nigel<br>
><br>
<br>
In a followup mail, Nigel adds:<br>
>Actually you did ask what I could tell you about it.... it was a serially<br>
>connected MSF/Colorado/DCF receiver. Mine was headless, but you could get ones<br>
>with displays. This is all that is left of the name:<br>
><a href="http://www.radiocontrolledclock.com/aratwa.html" rel="noreferrer" target="_blank">http://www.radiocontrolledclock.com/aratwa.html</a>. It worked very nicely in the<br>
>days when I only had dial up Internet.<br>
<br>
For values of "worked" equaling samples at 300bps and 63ms of jitter,<br>
heh. The doc page warns: "By default this clock reports itself to be at<br>
stratum 2 rather than the usual stratum 0 for a refclock, because it<br>
is not really suited to be used as other than a backup source." Indeed,<br>
a $30 GPS can do nearly two orders of magnitude better.<br>
<br>
Internet search does not turn up even the faintest trace of these<br>
things. (Other than lots of hits in old NTP documentation, that is.)<br>
I think this tells us they passed out of use before 2006.<br>
<br>
I'm going to say this is 1.5KLOC of dead weight and goes *PLONK*. Mark,<br>
this is just a hair more aggressive than our previous policy about<br>
legacy hardware that might still *conceivably* be running somewhere, but<br>
the last maintainer just told us to shoot it.<br>
<br>
Nigel: we're NTPsec, a security-focused fork of the Mills<br>
implementation. A main thrust of our technical strategy is to strip<br>
out as much legacy and dead code as possible to reduce attack surface.<br>
With what you just advised us to drop I think we'll get to 59% of the<br>
old code removed.<br>
--<br>
<a href="<a href="http://www.catb.org/~esr/" rel="noreferrer" target="_blank">http://www.catb.org/~esr/</a>">Eric S. Raymond</a><br>
_______________________________________________<br>
devel mailing list<br>
<a href="mailto:devel@ntpsec.org" target="_blank">devel@ntpsec.org</a><br>
<a href="http://lists.ntpsec.org/mailman/listinfo/devel" rel="noreferrer" target="_blank">http://lists.ntpsec.org/mailman/listinfo/devel</a><br>
</blockquote></div>