No, it doesn't apply identically, really, and that's what brought us here.
Cogent has to pay L3 because they aren't as significant a player as L3. Their "portion of the internet", as they call it, just isn't that relevant to a lot of people.
Cable modem users and DSL users are unaffected (unless Cogent is their provider's only upstream, and I haven't heard of any that are in that situation). They can use the vast majority of interesting services on the 'net just fine. The vast majority of high-traffic web sites are on multi-homed connectivity. There *are* popular sites on cruddy connectivity, but they'll learn quick with this, as they should. Never keep all your eggs in one basket. That lesson is as old as the commercialized Internet.
Those who are truly affected are businesses who have web sites and are unwilling to pay for enterprise-class connectivity. Not L3's problem man, ya get what you pay for. If your customers suffer, then I suggest you buy some real bandwidth. And if you're complaining because the free sites can't afford it, well, that's just tough. There's no SLA from ANY provider that says "you can reach 100% of the internet 100% of the time," let alone from a $19.95/month web host or a $50/mo cable provider. Again, you get what you pay for. Good connectivity costs good money.
As for the home users who are whining about this, y'all need to get a life and just go outside when your favorite site is offline for a day or two. It ain't the end of the world.
I design and build high-traffic internet-connected web sites on a hardware and software level for a living. I've never managed to work in this arena for someone who would build a completely single-homed system. Most serious Internet businesses recognize the need for multi-homing before they begin the install. Any failure to implement redundancy in those environments is what we call "acceptable risk." The folks that don't realize the need quickly change their willingness level after their first major outage.
Most, though, just lease space at a colo that has decent connectivity already installed.
On another note, I *am* on RR. Cogent doesn't exist from my net.perspective. I read quite a few sites regularly, and I have yet to run accross one that's affected. The impact of this event is quite small unless you happen to have the misfortune of being a Cogent customer.
BTW, as far as "it's not your job", yes, it most certainly is if you're in Internet-related support. Internet support techs (be it for an ISP or a web site or what have you) have had to explain outage causes to normal people since day 1. And once those folks realize that Cogent sucks, they'll move on. Cogent screwed themselves by not playing ball with L3 from what I've heard, and ultimately they're the ones that will pay for that business decision when customers leave them in droves.
It certainly would have been nice if Level 3 had notified their customers, no doubt about that.
On the other hand, from my position at home and my position at a data center belonging to one of the top 20 web sites on the 'net, it looks to me like Level 3 didn't lose connectivity to anything except Cogent customers.
That would be Cogent's problem for not having a well-connected network -- which is a very good business reason for terminating a peering arrangement. If they had real connectivity to other networks, I imagine they'd be reachable by other means.
Speaking as someone who specifies massive amounts of bandwidth, I can say with reasonable certainty that it's rather unlikely. Most companies that do large-scale internet work are already multi-homed. Those that aren't probably can't afford the extra hardware anyway.
Those that are on cogent, however, are probably now considering other providers. That cogent can just "disappear" from a large fraction of the Internet says to me that their infrastructure and peering arrangements suck way too badly for a major provider.
Instead of bitching, why doesn't some enterprising slashdot reader write up an article and submit it for publication? Fox would probably love to have a good war going on in their pages
(1) You need to learn to make your point without cursing (even in subject lines). (2) The post you were responding to was made like two weeks ago. Get a life. (3) Get a sense of humor while you're at it. (4) If you really feel a need to argue about Slashdot UIDs for longer than ten seconds, then you really need to get out more. Seriously. They don't say anything at all about the person behind them, and there are more important things in life to worry about anyway.
I would hope it would be the restaurant that they would be suing for not taking their card...
Last time I checked, merchants can and will still type the numbers in manually if the card doesn't work. They want the sale, after all, so a not-working card isn't good for them, either...
I remember when computers came with user manuals. On paper.
WTF is up with that, anyway? Has the world gotten so stupid that they can only read the big pictures on the huge poster sitting on top of the computer?
Firstly, nothing is "written in COM". COM is a technology that can be used from just about any language, and is very similar in both design and function to CORBA. Interestingly, COM is yet another apparent example of Microsoft's NIH syndrome: CORBA was invented around the same time according to the available history, probably just after CORBA. Odd how that pattern recurs, no?
Mozilla may have tipped their hat to Microsoft's name, but Microsoft isn't the originator of the technology.
Secondly, the fact that IE uses COM is not a security issue at all. COM is perfectly safe when handled properly. The fact that IE can be coerced into automatically downloading an unknown COM component from J. Random Website and executing it -- THAT is the security problem that everyone talks about.
--S (if I'm wrong about the history of (D)COM and CORBA, somebody do please point it out with references - I'd like to know.)
Hell, I'm still waiting for widespread adoption of.NET. It doesn't seem to me to be the dominant programming environment on Windows just yet, and it was introduced quite a long time ago.
That's right folks, no more of that annoying consistency between GUI applications, now anything that the guy down in marketing can draw is a workable GUI - just think of the possibilities.
Fortunately, if this is *anything* like *any* GUI designer ever built, the Marketing and BizDev dweebs that do this sort of thing aren't going to be able to figure out how to use it. Can you honestly imagine them sitting around and placing (*gasp*) named fields on a form?
Unless, of course, you *LIKE* "field1", "field2", etc...
Nah, they'll stick to their paint programs. Because all those little functional details are oh so annoying...
I'm not holding my breath for the marketing folks on this one. If it manages to work for people who really do have UI design experience tho, it could be a good thing for Windows.
--S
Re:How can you vouche for the security of this?
on
Flash, Meet Sparkle
·
· Score: 2, Informative
Last time I checked, people were still backporting fixes into 2.2.x branches...
Besides, each new Linux kernel is free, which means upgrading is much more painless financially. In the Microsoft world, "it's too old/we don't support it" really means "you have to pay us another license fee to get your security patches! ".
And if you're running a super-old kernel, you have all the old source and new patches available and you can backport the fixes yourself. Try THAT with Windows...
The point: you're comparing apples and oranges, and saying that apples are worse because they aren't orange.
--S
Re:How can you vouche for the security of this?
on
Flash, Meet Sparkle
·
· Score: 1
It isn't in Microsoft's best interest to make IE7/8 truly standards compilant. They will pay lip service, they will be 'more' standards compliant, but they will also introduce yet more proprietary crap to keep people locked in.
As an aside, NT4.0 is indeed roughly ten years old -- which means that Microsoft had more or less the functionality available in Windows XP, ten years ago. Linux, on the other hand, was just getting its first GUI support back then and has now nearly caught up to Windows XP in both performance and functionality (there are still some issues that need to be resolved, but largely it's there). What has Microsoft really done in that time, other than sand rough corners and change their UI look and feel?
This makes me wonder what the next ten years will bring for Linux, and how that will compare to Microsoft's brand of innovation... The end of Linux's catch-up phase is nearing. Then it will be time to start innovating. THAT is when Microsoft will be in trouble.
Personally, I give it another two or three years before the time arrives. Maybe I should go sell my Microsoft stock...:-)
Before you mod someone as troll, expecially someone with a UID of 402, think a little bit about their argument and moderate the others as redundant if you're so eager to call someone a troll who's clearly not.
Yes yes, because a low-value slashdot UID is a sure indicator of intelligence!
No, it doesn't apply identically, really, and that's what brought us here.
Cogent has to pay L3 because they aren't as significant a player as L3. Their "portion of the internet", as they call it, just isn't that relevant to a lot of people.
--S
Actually, it's you who are missing the point.
Cable modem users and DSL users are unaffected (unless Cogent is their provider's only upstream, and I haven't heard of any that are in that situation). They can use the vast majority of interesting services on the 'net just fine. The vast majority of high-traffic web sites are on multi-homed connectivity. There *are* popular sites on cruddy connectivity, but they'll learn quick with this, as they should. Never keep all your eggs in one basket. That lesson is as old as the commercialized Internet.
Those who are truly affected are businesses who have web sites and are unwilling to pay for enterprise-class connectivity. Not L3's problem man, ya get what you pay for. If your customers suffer, then I suggest you buy some real bandwidth. And if you're complaining because the free sites can't afford it, well, that's just tough. There's no SLA from ANY provider that says "you can reach 100% of the internet 100% of the time," let alone from a $19.95/month web host or a $50/mo cable provider. Again, you get what you pay for. Good connectivity costs good money.
As for the home users who are whining about this, y'all need to get a life and just go outside when your favorite site is offline for a day or two. It ain't the end of the world.
I design and build high-traffic internet-connected web sites on a hardware and software level for a living. I've never managed to work in this arena for someone who would build a completely single-homed system. Most serious Internet businesses recognize the need for multi-homing before they begin the install. Any failure to implement redundancy in those environments is what we call "acceptable risk." The folks that don't realize the need quickly change their willingness level after their first major outage.
Most, though, just lease space at a colo that has decent connectivity already installed.
On another note, I *am* on RR. Cogent doesn't exist from my net.perspective. I read quite a few sites regularly, and I have yet to run accross one that's affected. The impact of this event is quite small unless you happen to have the misfortune of being a Cogent customer.
BTW, as far as "it's not your job", yes, it most certainly is if you're in Internet-related support. Internet support techs (be it for an ISP or a web site or what have you) have had to explain outage causes to normal people since day 1. And once those folks realize that Cogent sucks, they'll move on. Cogent screwed themselves by not playing ball with L3 from what I've heard, and ultimately they're the ones that will pay for that business decision when customers leave them in droves.
--S
Hm, kinda defies the term 'peer' with respect to L3, now, doesn't it... :-)
--S
There was a disruption?
Wow, I didn't notice until I saw this story. Literally...
--S
Don't forget: THE SKY IS FALLING, TOO!
Seriously, in this business, you get what you pay for. So suck it up.
--S
It certainly would have been nice if Level 3 had notified their customers, no doubt about that.
On the other hand, from my position at home and my position at a data center belonging to one of the top 20 web sites on the 'net, it looks to me like Level 3 didn't lose connectivity to anything except Cogent customers.
That would be Cogent's problem for not having a well-connected network -- which is a very good business reason for terminating a peering arrangement. If they had real connectivity to other networks, I imagine they'd be reachable by other means.
--S
Speaking as someone who specifies massive amounts of bandwidth, I can say with reasonable certainty that it's rather unlikely. Most companies that do large-scale internet work are already multi-homed. Those that aren't probably can't afford the extra hardware anyway.
Those that are on cogent, however, are probably now considering other providers. That cogent can just "disappear" from a large fraction of the Internet says to me that their infrastructure and peering arrangements suck way too badly for a major provider.
--S
There's only one way to find out.
Instead of bitching, why doesn't some enterprising slashdot reader write up an article and submit it for publication? Fox would probably love to have a good war going on in their pages
--S
(1) You need to learn to make your point without cursing (even in subject lines).
(2) The post you were responding to was made like two weeks ago. Get a life.
(3) Get a sense of humor while you're at it.
(4) If you really feel a need to argue about Slashdot UIDs for longer than ten seconds, then you really need to get out more. Seriously. They don't say anything at all about the person behind them, and there are more important things in life to worry about anyway.
-S (</public_service_announcement>)
PS: Your momma wears combat boots! Nyah Nyah!
That doesn't work either, as the web was far, FAR less popular when IE was initially introduced.
-S
I would hope it would be the restaurant that they would be suing for not taking their card...
Last time I checked, merchants can and will still type the numbers in manually if the card doesn't work. They want the sale, after all, so a not-working card isn't good for them, either...
--S
Or they could be doing backups every other week...
--S
I remember when computers came with user manuals. On paper.
WTF is up with that, anyway? Has the world gotten so stupid that they can only read the big pictures on the huge poster sitting on top of the computer?
There oughta be a law...
--S
True. There's always Kazaa, Napster, Bittorrent...
(I suppose I should run away now)
--S
Judging by what's coming out of Hollywood lately, the other thing must be shit.
--S
Yaknow, I would mod you funny, but I think you're serious, and that scares me.
Hint: Relativity has nothing to do with why faster ethernet cards might use a faster clock. Think "Frequency" instead.
--S
Those jokes were terrible. You're all grounded.
--S
Firstly, nothing is "written in COM". COM is a technology that can be used from just about any language, and is very similar in both design and function to CORBA. Interestingly, COM is yet another apparent example of Microsoft's NIH syndrome: CORBA was invented around the same time according to the available history, probably just after CORBA. Odd how that pattern recurs, no?
Mozilla may have tipped their hat to Microsoft's name, but Microsoft isn't the originator of the technology.
Secondly, the fact that IE uses COM is not a security issue at all. COM is perfectly safe when handled properly. The fact that IE can be coerced into automatically downloading an unknown COM component from J. Random Website and executing it -- THAT is the security problem that everyone talks about.
--S (if I'm wrong about the history of (D)COM and CORBA, somebody do please point it out with references - I'd like to know.)
[TinFoilHat] :-)
Maybe Sun wants to buy Novell so they can force the dropping of Novell's counterclaims against SCO.
[/TinFoilHat]
(for the humor impaired, yes, this is a joke!)
--S
Hell, I'm still waiting for widespread adoption of .NET. It doesn't seem to me to be the dominant programming environment on Windows just yet, and it was introduced quite a long time ago.
--S
Fortunately, if this is *anything* like *any* GUI designer ever built, the Marketing and BizDev dweebs that do this sort of thing aren't going to be able to figure out how to use it. Can you honestly imagine them sitting around and placing (*gasp*) named fields on a form?
Unless, of course, you *LIKE* "field1", "field2", etc...
Nah, they'll stick to their paint programs. Because all those little functional details are oh so annoying...
I'm not holding my breath for the marketing folks on this one. If it manages to work for people who really do have UI design experience tho, it could be a good thing for Windows.
--S
Last time I checked, people were still backporting fixes into 2.2.x branches...
Besides, each new Linux kernel is free, which means upgrading is much more painless financially. In the Microsoft world, "it's too old/we don't support it" really means "you have to pay us another license fee to get your security patches! ".
And if you're running a super-old kernel, you have all the old source and new patches available and you can backport the fixes yourself. Try THAT with Windows...
The point: you're comparing apples and oranges, and saying that apples are worse because they aren't orange.
--S
It isn't in Microsoft's best interest to make IE7/8 truly standards compilant. They will pay lip service, they will be 'more' standards compliant, but they will also introduce yet more proprietary crap to keep people locked in.
:-)
As an aside, NT4.0 is indeed roughly ten years old -- which means that Microsoft had more or less the functionality available in Windows XP, ten years ago. Linux, on the other hand, was just getting its first GUI support back then and has now nearly caught up to Windows XP in both performance and functionality (there are still some issues that need to be resolved, but largely it's there). What has Microsoft really done in that time, other than sand rough corners and change their UI look and feel?
This makes me wonder what the next ten years will bring for Linux, and how that will compare to Microsoft's brand of innovation... The end of Linux's catch-up phase is nearing. Then it will be time to start innovating. THAT is when Microsoft will be in trouble.
Personally, I give it another two or three years before the time arrives. Maybe I should go sell my Microsoft stock...
--S
Yes yes, because a low-value slashdot UID is a sure indicator of intelligence!
--S (Not.)
Er, aren't the Sun drivers for Linux open-source? Can't the OpenBSD folk just port them?
--S