Language:
switch to room list switch to menu My folders
Go to page: First ... 11 12 13 14 [15] 16 17 18 19 ... Last
[#] Thu Feb 11 2010 02:37:22 EST from fireball @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Yeah, but IPV6 is both relatively cheap and relatively painless when
migration happens, "eliminating spam" is a completely unobtainable
goal... if I went big against spam as an ISP all that will happen is the
spammers will go somewhere else. Anyway, spam is a fairly small number in the IP address subset, it's not like
there is as many spammers as there is end users by any means. And I
completely require the ability to obtain a public IP, I want to be able
to connect up and play games, or if I must host something like a BBS on
my home computer without having to pay a small fortune to do it. If I'm
denied the ability to have an afordable public IP address then really
the problem only gets bigger, the internet becomes controlled by the
rich and I'll tell you a secret, the spammers aren't really going to be
affected at all (I don't really even need a public IP address to spam
you,
I'll just be sending that over an open relay anyway). Unless
you're talking about domain squatting, that's a completely different
problem entirely.

[#] Thu Feb 11 2010 03:04:11 EST from fireball @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

And what is this spam that accounts for 60 percent of internet traffic?
I'm calling your bluff, where's that number come from? The number seems
somewhat funky.

[#] Thu Feb 11 2010 10:45:01 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

spammers will go somewhere else. Anyway, spam is a fairly small number
in the IP address subset, it's not like

I was referring to bandwidth not ips.

[#] Thu Feb 11 2010 10:46:16 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Put this in google:
percent of all internet traffic is

So it's easy to justify any claim, really.

[#] Thu Feb 11 2010 11:48:53 EST from fleeb @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

I love how 90% of email traffic gets translated to 90% of all web traffic.  Like 'web' means 'internet'.



[#] Thu Feb 11 2010 12:13:36 EST from IGnatius T Foobar @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]


Ford: if you can fix spam using the amount of resources that has been spent on IPv6 development so far, it would be a bargain and there are a lot of people who would fund you.

In any case, you're begging the question. You're saying that IPv4 depletion will not be a problem because IPv4 depletion will not be a problem. From my perspective it's a looming crisis, and it's not like global warming where you have to trust Al Gore and the IPCC when they make unprovable claims. You can look up the registrations and watch the supply of addresses being depleted. It's not like the global supply of oil either -- the exact number of addresses left is readily available, and the rate of depletion is measurable.

We're gonna have a shitstorm when IPv4 runs out. Everyone knows that we need to move to IPv6, but nobody wants to go first because there's no immediate benefit from it.

[#] Thu Feb 11 2010 12:27:39 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

I'm not saying it's completely not a problem, just that there are more pressing problems.

A lot can happen in a few years and it may turn out there are other cheaper solutions to the problem, that is all.


[#] Thu Feb 11 2010 13:19:37 EST from fireball @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

I'm not actually too worried about the transition (though I don't work
in it). As far as I'm concerned when there's any sign that IP addresses
aren't being given out anymore, people will convert their dns's, a small
ISP will announce that in order to properly serve their customers they
will give out IPV6 addresses (probably in addition to non public IPV4
addresses threw a NAT) as they can't afford to offer compeditive
service using IPV4. Within a year every small ISP will follow suit in
order to remain competitive pricing wise, within 2 years every big ISP
will follow... and that is assuming that the switch doesn't happen
before it becomes an issue.

As for 90 percent of spam emails that really justifies nothing,
especially when another study says over 50 percent of traffic is p2p
(more what I expected, but to be honest I think most of these studies are crap,
far more al gore
then the IANA allocations lists, of which the entire
organization system is built upon). Emails in general tend to be small,
and spam almost never actually has attachments, even then they're
probably tiny (hard to find a good open relay that will forward large
attachments I'm sure and no spammer want's to pay for the upfront
bandwidth cost that will happen. If it ever did become a major issue
there will just be a totalitarianism filtering scheme put into place,
and peoples lagitimate emails will be bounced and they'll be told to
file a false filtering report in order to ever get something through.
That or a "friend request only" style email system will gain some
steam...

Either way you seem to think bandwidth usage is a bigger problem and I
don't have a problem with that really. Overiously I rabbit trailed that
argument farther then I ever should've thinking you were claiming
it
related to the IP address issue. It is really quite hard to argue
priority, so I think I shant even try, but I will ask one thing exactly
how is using an article about spam in email to make a claim about
spam bandwidth usage valid when claims from companies like ARIN (the
regional internet registry for the North American region, as in who
actually deals with IP address allocations in that area) saying there is
a problem
(https://www.arin.net/knowledge/about_resources/ceo_letter.pdf) is
considered al gorish. The people making these statements aren't crazy,
they're people who have every little detail of the IP address pool
mapped out. I seriously think you under estimate the real scope of the
problem.


Obviously I'm ranting though now, so I probably should just stop
completely, I don't think I want to build a reputation as a lover of
arguments just yet.

[#] Thu Feb 11 2010 13:29:24 EST from fireball @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Yeah 3 pages is too long :P I will say though, as far as the pace of
larger ISPs and bigger corperations go, 2 years is right here right
now... and you treditionally want to get out of dodge before thing start going down.

Nor is IPV6 really a costly thing to impliment, all the hardware that
everyone has should support it. It's kindof like the Y2K shift, it was
actually somewhat of an issue, it just goes over so easily no one really
even noticed. I certainly am not making claims that the transition will
be the end of the world and the internets itself.

Really I should probably drop out now before I get well more rantish. I
really just like talking loudly about technology... becoming a CS major
does bad things to you, I wouldn't suggust it on anyone. :P

[#] Thu Feb 11 2010 13:39:22 EST from fleeb @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

I suspect it's the software that's really at issue here.

Software has to work with IPv6, and I suspect quite a bit of it does not.



[#] Thu Feb 11 2010 14:20:39 EST from fireball @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Which is why when you do crazy things like make your software reach from
the application layer where most software belongs all the way down to
the network layer you get the privalage of heavy rewriting. It's the
same thing as Y2K a chance for some major rewriting and coders to learn
from past mistakes. Also a chance to finnaly free ourselfs of a layer
of completely crusty old software. That transition has been predicted
by most and slowly incorperated to software since the standard was
created in 1998. I still feel that unupgradable hardware is the biggest
issue. Then again the chances of me eating my hat and being
forced to migrate some old internet server software to IPv6 is prety
high (and like legacy hardware there is legacy software to boot, and
boot it I would like). Ah well, atleast I can rest assured that the
migration process to me as a end user with newer hardware (that I've made sure is IPV6
complient).

[#] Thu Feb 11 2010 14:29:24 EST from fleeb @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Well, it's a little more involved with software.

First, there are the objects and such that you'd use to work with the networking layer.  They are different between v4 and v6.  The difference is very minor, but so were the differences required for Y2K. For most of the networking-related code I use, it's really just the use of a different set of objects, but is otherwise identical to v4.  I have servers that already expect either v4 or v6, and will work with both (because I set up both sets of objects).  The clients aren't quite as smart, but hopefully I can figure something out for them soon.

Secondly, there's the bit where you resolve an IP address.  If you always expected a DNS entry, that'll go okay.  If you expected an IP, though, you're in for a heavy rewrite of the UI part of your code to make sure that it can handle the v6-style IP addresses (or rewrite everything to expect DNS entries).  I'm looking at some of our current GUIs where I work, and thinking that moving to v6, for the guy who has to deal with the GUIs, is going to suck hard.

Then, how do you present the option to the user?  Do you have a checkbox to indicate that you want to connect via v6?  Do you guess?  How do you guess, especially with DNS entries?

I've been trying to prepare for it, though, on my end.  My boss thinks I'm silly, but he'll thank me for the relatively smooth transition later.



[#] Thu Feb 11 2010 14:42:31 EST from fireball @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Yeah, I'd rather be responsible for rewriting a gui app to resolve
addresses via DNS (like I think most apps should) then being responsible
for hacking y2k support into something that you don't even have the
source for anymore (your database support software for example). Not
that that kind of software will transition properly, I just atleast have
the option of connecting to it via IPv4 still and then rewriting my
server to work. It'll be an issue, but it still seems smaller then Y2K
was. Either way it doesn't hurt me at all as a coder, just gives me
more oppertunities in the end, but I think too differently then big
businesses, and therefore it'll likely be a far bigger problem then I
see it, as in my eyes forced development or abandonment of stagnant
applications is a good thing (tm).

[#] Thu Feb 11 2010 15:52:14 EST from IGnatius T Foobar @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Bandwidth is *far* cheaper than it was even a few years ago. We buy bandwidth for a few dollars per megabit now.

[#] Fri Feb 12 2010 11:09:17 EST from dothebart @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

the spam problem will only go away if somebody gets some spammer killed by a drastic death.



[#] Fri Feb 12 2010 13:51:49 EST from fireball @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Feb 12 2010 11:09am from dothebart @uncnsrd
the spam problem will only go away if somebody gets some spammer
killed by a drastic death.

Sadly last time I checked "fighting spam" was a crime, or I'd consider
that an interesting proposition.

[#] Fri Feb 12 2010 16:14:36 EST from fleeb @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Hmm... maybe that should be the name of my next Champion's Online character.  Spam Eradicator!



[#] Fri Feb 12 2010 23:34:00 EST from fireball @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Vigilantee Justice at it's best.

[#] Thu Mar 04 2010 12:27:16 EST from IGnatius T Foobar @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

The end of an era!

http://tinyurl.com/byebyenetware

General support for NetWare ends this week. (Extended support ends in two years; self-service support resources end in five years.)

A moment of tribute is in order, for the Novell of old, the company that wasn't merely the Linux division of Microsoft.

[#] Thu Mar 04 2010 12:31:58 EST from the8088er @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

A few months ago I touched up a real estate company's NetWare server. It kept crashing because the CPU fan siezed. Pentium I or 486 it was.... running for 14 years.

Go to page: First ... 11 12 13 14 [15] 16 17 18 19 ... Last