So, as you know, I recently upgraded to DSL from Cable. The Actiontec DSL modem provided by Qwest is really cool and provides quite a few features, but the firmware upgrade provided at the company website makes the thing stop responding.

My self-assigned quest was to figure out exactly what was going wrong, and to do so in a non-destructive way. This involved attaching a serial cable to the modem’s circuit board and using it as a console to interactively diagnose the issue.

Some disassembly required. Some pictures provided.
http://www.tlarson.com/guides/dslhack

If you’ve ever had cable, you know that after your new-customer promotion is done, they about triple your monthly rate, that goes for both cable TV and Internet service. So once we had to start paying around $100/month for our basic service, we shopped around for other offers.

Thank heavens for competition. Our phone company was offering a 1-year promotion for $20/month. After the year is over, the price goes up to $25/month. I can live with that. Of course, you’re going to say, what about download speeds?

The cable company, with its massive bandwidth capacity, often advertises download rates that rival those business links that cost thousands of dollars per month. The catch, of course, is that business bandwidth is guaranteed, residental is not.

So in this rare opportunity where I have both DSL and Cable Internet access at the same time, on the same computer, I decided to conduct a scientific experiment. Using Speakeasy’s Speed Test, I did a few speed comparisons with each ISP. There was very little variation in the results; I’m not much one for statistics, but I’d guess a margin of error of about 2% or so.

A couple of points stand out here. First of all, DSL is hitting exactly 82% of its advertised speed both upload and download. That’s significant because general TCP/IP overhead counts for about 20% of your bandwith usage. That means that the DSL connection is actually delivering around 100% of its advertised bandwith.

Cable, on the otherhand, reaches a dismal 28% of its download capacity, and 56% of its upload. This test was conduced at 7pm local time, which is just after cable’s peak usage period. As you may know, total bandwidth is shared between everybody on your cable broadcast network, which is usually everybody in your immediate neighborhood. So while you may be able to get better download speeds in the middle of the night, what really counts is what you get when you’re actually using it.

Another interesting metric is price versus speed. For example, cable costs about 400% of the price of DSL (no promos), and gives you about a 136% of the speed. Conversely, DSL costs roughly 30% of what Cable costs, but gives you about 74% of the speed.

Also, the Cable provider is very protective of its bandwidth, since my usage cuts into other customers’ availablity. As such, certain methods of using your internet connection are prohibited; like for example, running any sort of server whatsoever from your home. As for DSL, it’s all the same to the phone company: whatever bandwidth you don’t use can’t be reallocated to someone else, so you might as well use it for whatever you want.

I’m working on a user-friendly content management system (Wiki-based, for those who care), and I wanted to add a WYSIWIG editor interface, like HTMLArea has, for example.

So, I dug through the javascript source code to find out how they do it. Obviously there’s some real deep magic involved, right? After all, some people pay a lot of money to use things like this on their web page.

So what did I find? Magic? No. It’s all built into the browser. Internet Explorer and and Mozilla-based browsers both have an “editable” mode that handles all the excitement. Microsoft calls their version The MSHTML Editing Platform, while Mozilla has a built-in module called Midas.

What you’re really paying for is that nifty toolbar that allows you to send commands to the browser that say “make this bold” or whatever. It seems almost dishonest for these vendors to brag about all the “features” they’ve included in their version (like “Fonts!”, “Sizes!”, and “Images!”), when it’s actually all provided as part of the browser… which the user already owns. Aw well, welcome to the world of Marketing.

Just for a heck of it, I created an editable version of my own homepage. It’s just the normal page, but with an extra line of JavaScript tacked on the end to turn on editing mode. Note that if you’re using a Mozilla-based browser (like Firefox), it seems to want to stay in editing mode even when you go to another page. You’ll have to close the current tab (or window) to get out of editing mode, since I don’t have a button on the page to turn it off.