Mailing Lists: Apple Mailing Lists
Image of Mac OS face in stamp
Re: Apple's iChat AV + iSight
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Apple's iChat AV + iSight



On zondag, jun 29, 2003, at 16:01 Europe/Amsterdam, John C. Welch wrote:

Why do you object to me stating my desire that Apple would augment
their tool so it is suitable for my needs?

Because right now, iChat is not the app for IPv6.

I agree that other than the basic idea that all apps really should support IPv6, iChat isn't the most natural candidate for pushing IPv6. My main gripe with iChat is that it doesn't interoperate with "regular" VoIP and it only supports the very proprietary AOL instant messaging protocol. Based on rumors, I was expecting jabber support. This would have made it possible to interoperate with other IM services as jabber supports all kinds of gateways, and, very important to business users, you can run your own jabber server rather than have to rely on AOL's servers.

What would be a better direction is to push one of the
high - end conferencing vendors, because multipoint conferencing would
benefit FAR more from IPv6 than iChat. Or, start pushing the QuickTime
Streaming Server team.

How would IPv6 benefit streaming? I had more than enough problems to get it to work over IPv4 in the first place... I'm happy to see that the QT streamer supports multicast. In theory it would be easier to do multicast over IPv6 for average users (they use a tunnel anyway, so tunneling the v6 multicast shouldn't be much additional effort) but I've seen very little being done in this area.

I think Safari is the most important app that needs to become IPv6-aware. Not so much because the web really needs IPv6 (www is high on the list of best NATable applications) but because the deployment of IPv6 is the highest for the web, it provides a good test case and it allows people to experiment for themselves. Obviously compiling all the standard tools such as ssh with IPv6 support and including apache2 rather than 1.3 would help as well.

I don't know anyone who uses AIM.

Then you don't know a lot of people. ;-)

True enough, there are a lot of people I don't know. (-:

Actually, if you know anyone who uses iChat, you know people who use AIM.

Let me put it this way: I don't know anyone who uses AIM other than the people I met over AIM...

:-) Apple is WAY behind on Microsoft when it comes to IPv6... I'm not
even mentioning the Unix family.

Who cares.

People on this list maybe?

But they are NOT the target iChat market.

Are we still talking about iChat then?

This inability of the tech
community to pick battles better is a real killer in getting technology to
move forward. Streaming media likt QuickTime Streaming Server, Real, Windows
Video are FAR better targets than iChat, they would get more of a benefit on
the server AND the client end.

But here multicast is what is called for, not so much IPv6.

IPv6 is a research toy to the rest of the world, and will be for a
number of years to come.

To most people the internet is a toy anyway.

Not at all. Even at the home level, the internet is a critical way to
communicate both professionally and personally,

Not for most people. For them the internet is a way to exchange lame puns and to get their hands on illegal copies of stuff.

and there are very few
businesses in areas with the infrastructure to support it that are not
relying on the Internet as a mission - critical infrastructure.

I hope you're wrong, because most businesses connect to the net in very unreliable ways. Or should I say: a very unreliable way. They don't bother to put in a second connection most of the time.

In the modern world, the Internet is as critical as power and phone service.

I partially agree with the phone service (obviously power is more important as without it no internet or phone and then there's all the other stuff that needs power). However, that's not because the phone service as a rule works so well. Sure, when you drive your car in a ditch it's great. But try doing some business over the phone: it works less than 50% of the time because of voicemail or idiotic "voice response" (a false advertising claim waiting to happen) systems.

IPv6 was never a research project. If not for the advent of ethernet
switches many enterprises would have been running production IPv6 by
now.

You also forgot NAT and CIDR.

Not forgetting them: CIDR and VLSM predate IPv6 and NAT is just an easier way to do what people have done for a very long time: use proxies. I think the switches are the real "disruptive technology" here.

The addressing reasons are the worst reasons to push IPv6.

The only reason that "cuts any wood" to use a Dutch expression. (= the only valid one)

That was handled a while ago. Quite honestly, no one in their
right mind is remembering an IPv6 address anyway.

??? What does that have to do with anything?

But it's the other
properties of IPv6 that need to get pushed. All I ever see are
propeller-heads talking about address space. That's a chimera.

The address size is the only substantial difference between v4 and v6.

What SHOULD be the point is built - in security, QoS, etc.

IPv6 security == IPsec == !NAT == larger addresses

The QoS byte is the same in IPv6 and current IPv4 implementations. In theory there is the flow label but in practice that doesn't amount to anything as nobody has figured out how to take advantage of it yet.

Sorry to say, there are very few real world problems with NAT that are
insurmountable with relative ease. It's not the most elegant solution,
but it works for most of us.

That's because most of us don't use voice over IP and similar
applications.

Um, I've been using it for years. You can make it work in a NAT situation.

Sure. Just means more work for both the vendor and the user and it's less reliable (more constraints on the network architecture) to boot. You can also eat soup with a fork but a spoon makes the process much more enjoyable.

Their reasoning is that IPv4 only has a few usable
years left, so building something new that only works with IPv4 would
be a mistake, and they can tunnel IPv6 whereever there is IPv4 present
anyway,

Then they better start flogging switch and router makers, and work some
deals with ISPs.

Switches don't need IPv6 support. There are enough products that will happily terminate tunnels. That's all you need.

And I still say there's at LEAST 4-5 years of IPv4 dominance left,

No disagreement there. But it's not a question of dominance. At some point for some users, IPv6 will become a hard requirement. Whether they do 0.1% or 99.9% of their traffic over IPv6 by that time is completely irrelevant as the protocol must be supported either way.

maybe more at current adoption rates, because I have yet to
see anyone come up with an app that is compelling at the infrastructure
level.

That's not going to happen. Suppose there would be an application that absolutely, positively needs IPv6 to work at all. As IPv4+NAT can easily support 8000 billion different end systems, this seems higly unlikely, but let's suppose... Then next to nobody gets to use this application. How is it supposed to catch on?

The DoD move shows that the expectation that IPv6 will be adopted is
creating momentum for the actual adoption.

Um...you haven't dealt with the DoD much have you. When I see it happening,
I'll believe it. It's a nice idea, but until money has been approved AND
appropriated, and the project has survived at least one Administration
change, it's a fantasy. If you bought other DoD moves, we'd all be using
OSI, not TCP/IP, and programming in ADA.

That's not the point. The point is that they acted on their perception of future IPv6 deployment. Their statements in turn add to the momentum (whether they add to deployment remains to be seen, as you say) which in turn prompts others to percieve IPv6 deployment to happen so they start making their own plans... This mechanism makes sure IPv6 won't just linger: either it goes forward or backward. That's a good thing.

Iljitsch
_______________________________________________
ipv6 mailing list | email@hidden
Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/ipv6
Do not post admin requests to the list. They will be ignored.

References: 
 >Re: Apple's iChat AV + iSight (From: "John C. Welch" <email@hidden>)



Visit the Apple Store online or at retail locations.
1-800-MY-APPLE

Contact Apple | Terms of Use | Privacy Policy

Copyright © 2011 Apple Inc. All rights reserved.