1 |
I just found out that Firefox recently _removed_ support for IPv6 |
2 |
link-local addresses. It was a very useful feature -- at least to me |
3 |
-- but it wasn't required by law, so they removed it. Yes, that's |
4 |
_actually_ what the devs said in the thread I found. |
5 |
|
6 |
AFAICT, chrome has never supported it. Links doesn't. w3m doesn't. |
7 |
|
8 |
Internet Explorer does. |
9 |
|
10 |
Oh, the shame... |
11 |
|
12 |
You'd think with a nice geeky feature like that, it would be the other |
13 |
way around: supported by firefox/chrome/links/w3m but not by IE. |
14 |
|
15 |
Of course the _way_ that Microsoft supports it using some meaningless |
16 |
numerical index as the zone identifier is rather half-arsed compared |
17 |
to the interface names you use on Linux, but at least it _works_ in |
18 |
IE. |
19 |
|
20 |
[For those of you keeping score, curl does support IPv6 link-local |
21 |
addresses, so it's not a shutout.] |
22 |
|
23 |
Now that RFC6874 is standards-track, I assume Firefox devs will be |
24 |
forced (against their will, apparently) to put that feature back in. |
25 |
Hopefully Chrome, w3m, links, et alia will follow suite. |
26 |
|
27 |
IPv6 link-local addresses are _way_ cool for dealing with embedded |
28 |
devices that have network interfaces. You can actually set them up |
29 |
and use them without having to faff about with dualing DHCP servers, |
30 |
temporarily adding an IP address/route to your laptop/desktop, using |
31 |
proprietary Windows-only widget-management utilities, configuring the |
32 |
thing via serial console, USB port, hardware switches/jumpers, etc. |
33 |
|
34 |
-- |
35 |
Grant Edwards grant.b.edwards Yow! |
36 |
at |
37 |
gmail.com |