The lower transfer cost was not all that spectacular. 64k vs 56k or 28k6. 1.5 dB at best and 3 dB worst case. Against considerable cost for the equipment and an almost doubled montly feee.
GK> using your voice phone while data transmission is running,
Having two "lines" was indeed a notable "plus". I would have loved to have two "lines" in the mid nineties. My then wife could be very long winded on the telephone. Two hour were no exception. A second line would have been welcome. Unfortunately ISDN was not available in my area until after 1996. And then she died and I did not need that second line any more. So I skipped ISDN. why spend time and money on ISDN when POTS just worked fine?
When my present wife came to live with me we had DSL and mobile phones, so no need for a second fixed line either.
GK> just to name a few. IPv6 offers nothing for most sysops apart
GK> from time to spend on understanding it, making it work, keeping it
GK> running.
IPv6 offers advantages for sysops. For starters there is the obvious "IPv4 will not just keep working well" forever as you yourself have just found out with your DS-light glasfiber connection. You are not the first to run into this problem and you will not be the last. There just isn't enough IPv4 to given everyone on this planet his/her own unique gloabally routable IPv4 address.
IPv6 solves that problem. And more than that. You do not just get ONE IPv6 address, you get enough to let every device in your network have its own unique globally routable IPv6 address. You can have an unlimited amount of serves running without having to mess with non standard port numbers. And more...
GK>>> I think the question is not so much about the technical support
GK>>> (for Linux or *BSD, IPv6 is available since around 2000). I know
GK>>> a couple of admins who -even if their provider supports IPv6-
GK>>> simply disable it on all machines due to the unnecessary
GK>>> complexity it comes with.
MvdV>> Than these admins have not understood or they are just lazy.
GK> "lazy" is the wrong word here. Not everyone is a retired person in a
GK> single household. Having a job, running a family, maintaining a house,
GK> vehicles and other things the familiy needs consumes lots of time
GK> already. When there is not much time left in the first place, why
GK> bother with IPv6?
Because IPv4 will not "just keep working" The transition to IPv6 is anavoidable if only to solve the problem of IPv4 exhaustion. Admins that do so for a job have a duty to convey that message to those who pay them. If they neglect that they are not doing their job.
MvdV>> Yes, a dual stack system is more complex than a single stack IPv4
MvdV>> system. But IPv6 by itself is not more complex than IPv4. On
MvdV>> the contrary I would say. IPv6 is less complex.
GK> I beg to differ: I spent the better part of a weekend understanding
GK> just the parts I need to get everything running with my DSlite
GK> connection (mainly DHCP, DNS, routing). It *is* more complex than IPv4
GK> in many places,
You have been dealing with IPv4 for decades. You were not prepaired for the situation that it would not "just keep working" Of course IPv6 looks more complex to you. Had you started a decade ago with familarizing it slowly one bit at the time you would percive it what it is: less complex.
MvdV>> But it is different, there is a learning cuve and one has to
MvdV>> "unlearn" some of the "IPv4 think". But once past that, it is
MvdV>> relatively easy. And unavoidable in the long run anyway...
GK> I do not see any device or OS being sold as "IPv6 only" in the
GK> medium-term future. There are way to many installations that require
GK> IPv4 compatibility. There is no market for an "IPv6 only" device or
GK> software.
And yet Microsoft, Google and some other Big tech ones are running their server parks IPv6 only internally. They only do IPv4 to IPv6 translation ad the edge of their networks. Mobile networks that offer IPv6 heve their network do IPv6 only. They offer IPv4 "as a service". All in preparation of an IPv6 only internet.
