We’ve been hearing about all those undersea cables that have been cut recently (and yes, that’s five) and all the conspiracy theories surrounding them. So here’s a map of all the Internet undersea communication cables to get some perspective of the situation:

And I thought that since I’ve already posted maps on late night bandwidth, root servers, Internet address space, and global bandwidth, one more shouldn’t hurt…

I created a new blog last week. That brings my count up to… well, a lot now. Anyway, the target audience is middle school to high school kids. (Don’t get any strange ideas here, this is for the youth group at my church, ok?) Anyway, so this is for teenagers which basically means it’s for people that have grown up never not knowing what a computer or the Internet was. So I put a post up on the blog which gave them links if they needed help with this or that and then at the end of it I put my email address up there for them as a last resort. At first I thought this was a pretty simple case of using a “mailto:” link in order to create an email when the user clicked on the link. But after a little thought, I quickly started questioning my decision.

If you have an email application installed on your computer, then mailto works perfectly well: it pops up a new mail with the “To:” field prepopulated, so that the user is all set to type in their life story and click “Send”. However, I don’t think that’s the common case anymore (and especially not for my user base).

Last year, I bought a couple copies of Office for the home computers. I got the version of Office that has Outlook just so I could have all the cool functionality that Outlook has nowadays. The only problem is I never use it. I always use the webmail interface provided by Yahoo and Gmail. What’s sad is that I actually love Outlook. All the nice folder capabilities, calendar views, task functionality… it’s just a great app. But I want the same experience on all computers I use. I don’t want my read email downloaded and available only on one of the three computers I use everyday. With POP email you definitely have that problem. With IMAP it’s a little better, but you can still have issues of folders created on the desktop not being available in the webmail and so on. I don’t want to be tied to always going to one computer to manage my email, so I just stopped using it.

I decided to strictly use webmail to manage my personal email and am pretty sure that that’s becoming the norm.

So what does that mean for good ol’ mailto? Well, there won’t be an application defined to handle “mailto”, so when the user clicks on a mailto link nothing happens… and that’s never good.

There are programs that will fix this for some webmail apps. For instance, G-Mailto and GMail Notifier make mailto work for GMail, but of course these are just one off fixes that you would have to find and install for your particular webmail on every computer that you use to check your email.

But either way, mailto doesn’t automatically work for webmail users. Worse yet, the user may right click on the link and select “Copy Shortcut” in IE or “Copy Link Location” in Firefox and paste that into their webmail app. That will include the “mailto:” part of it… and that won’t work either:

is from Yahoo and this one is from GMail:

And of course you have to be careful when putting email addresses in plain text on a website. You can obfuscate it using “[at]” instead the “@” sign, but that’s pretty low tech and bots probably check for that now anyway. Better yet you can encode it, so it won’t look right until it’s actually rendered, but that’s a little beside the point here….

So what’s the better of the two methods? Provide a “mailto:” link to foster an aging paradigm or just make it simple text that the user has to cut and paste? Cater to the crowd with desktop email clients installed all the while making the experience for the majority of email users a little more error prone?

Sad as it sounds, it looks like plain ol’ text may win…

I think most people have at some point in time seen this picture of the Earth at night:

After seeing this again tonight, it reminded me of an earlier image I posted that showed global bandwidth:

I found it interesting (and I suppose expected) that the brighter areas coincided with the heavier bandwidth areas. If you look at the two images overlaid (after some resizing and adjustment) it’s much clearer:

So is this correlation due to man’s thrust to band together to form great city-states multiplied by the accelerated expansion of technology across the globe or maybe just all the late raids in World of Warcraft?

We’ve seen Internet traffic maps and Internet address space maps, but there’s another interesting type of map applicable to the Internet. This is the root server map. Root servers are basically the gate keepers of the Internet (or would that be the key masters?) since they are the highest level authorities in the DNS namespace hierarchy. There are 13 root servers specified from A - M:

Although each lettered root server does not neccesarily represent a single physical server. Root servers C, F, and I-M each have physical servers located in different places:

There have been DOS attacks on the root servers to try and bring down the Internet, but none of them have succeeded. The distributed nature and replication of the servers has helped prevent a blackout of the Internet and based on the state of the root servers as of February 2007, this is increasingly less likely to ever happen:

Similar to the other maps of the Internet, here’s a nice visualization of the complete address space of the Internet (and in javascript, no less). You can zoom in and out and move around all Google Maps style. Click on any part of it to display who owns that address. The grey boxes are unrouted, the black boxes represent the registries and the red boxes represent known spammers.

From just a little scouting around, you can see how much of the Internet isn’t routable yet. I don’t know why IANA has all those reserved blocks, but it sure looks like we don’t need IPv6 anytime soon…

I’m not asking if we need UDP as a network protocol at all, but if we really need it as a transport protocol? Does it actually provide any real transport layer features? Let’s take a look at the UDP header:

We see a couple port numbers and a checksum. If we look at the TCP header we see a lot more:

It has port numbers and a checksum as well, but there’s also a sequence number, an ack number, control flags, window size, and other goodies. If you want to compare feature lists you can see how disparate they are. UDP provides:

• port numbers
• simple error-checking

So it’s useful, but nothing compared to what TCP offers:

• port numbers
• simple error-checking
• reliable delivery
• in-order delivery
• flow control
• congestion control
• segmentation

But maybe it’s not fair to compare UDP against TCP, the end-all-be-all-get-my-packets-there-or-else of the transport layer protocols. How about some of the others? Here’s a table I stole borrowed from Wikipedia:

 As you can see, UDP is really the runt of the family. So it seems as if our TCP/IP model is a little off:

TCP and the some of the  other transport protocols offer everything that UDP does and more. So why is this redundant functionality in UDP, a sibling protocol? Wouldn’t it make more sense to put it in a layer beneath these other protocols:

Creating a new “Session” layer (term borrowed from the OSI model) seems to fix the problem here. The protocols that don’t need UDP and don’t provide any real transport layer features were moved into this new layer too.

I like this new model a lot more. It better represents the real world and how these protocols are actually used: If you need simple port-to-port connectionless communication you choose UDP. If you need all that, but also a connection-oriented reliable communication stream, you choose TCP. So why hasn’t it been this way all along…

I was planning on writing about TOR sometime in the near future, but Bruce Schneier’s post spurred me on. Anyway, the onion router or TOR works by taking your Internet traffic, encrypting it, and forwarding it from one TOR router to another to another to another until it reaches the final TOR router, or exit node, where it decrypts the data and finally dumps it onto the Internet.

This means that posting to some odd website (or even the fact you visited that website) will be anonymous in the sense that the IP that the website sees will not be your true Internet IP address. And for that matter, anyone who sniffs your traffic will not see it either. However, this does not imply that you have privacy. Anything you send in cleartext like your gmail username and password will still be in cleartext when it gets dumped on the Internet. TOR doesn’t encrypt your traffic from the exit node to its destination.

You also have no guarantee that the TOR routers themselves will play fair. The exit node, since it sees everything right before it goes onto the Internet, can log every packet it sends. And apparently some do that very thing. How embarrasking…

Found a couple cool maps showing Internet bandwidth on a global scale. Here is one from 2005:

The striking thing shown here is how much the States are a bottleneck for the rest of the world. We have a nice pipeline to Europe and another to Asia, but there’s no pipeline between Europe and Asia… at least nothing comparable to the one created by the US. And thus we have created a star topology. Probably the worst that could have happened to a global structure, but that’s what we have. Here’s another map (although I don’t believe it’s quite as recent) that shows a similar statistic:

And as an added bonus, here’s a diagram showing peering over the Internet. The US is on the bottom, Europe on the upper right and Asia on the upper left: