Okay, so after an impromptu ten-day Dead Air Festival courtesy of the phone company, we're finally back and broadcasting from our normal digs. That is not to say, however, that everything is "back to normal," and so we thought we'd take this opportunity to fill you in on the gnarly sequence of events that's been plaguing us for the past couple of weeks. Why? Well, for one thing, you deserve an explanation for where we've been, as well as for why your viewing experience is possibly going to suffer from now on. For another, what happened to us might happen to some of you, and so we'd like you to know what you could be in for. And most importantly of all, we finally have a chance to bitch and moan about our woes to thousands of people-- who could resist?

First, the set-up. AtAT, as you may know if you've ever taken a look at our About page, is served from a single Power Mac sitting on a 768 Kbps SDSL line. That particular service is provided by the good folks at Network Access Solutions, who installed it in record time when our previous DSL provider, HarvardNet, bailed on all of its DSL customers and gave only thirty days' notice before service was cut off. The NAS connection was installed in mid-January, and it flawlessly provided 768K of AtAT-serving bandwidth-- until 11:54 EDT on April 4th. (We call it Black Wednesday.)

At that time, the line simply went dead and the WAN light on our DSL router kept blinking. On Wednesday night, NAS diagnosed trouble with the physical line itself; thus, Verizon entered the picture. See, here's the thing about DSL: apparently no matter what provider you sign up with to get your DSL service, the phone company still owns the physical lines carrying the signals. That's important to remember, because even though we aren't getting our DSL from Verizon (who, laughably, doesn't offer any DSL connection faster than 90K upstream to its "business-class" customers in this area), our DSL provider (just like all others) is completely at Verizon's mercy.

So Verizon agreed there was a problem and came out that Thursday to check things out from this end. On Friday they allegedly checked into possible problems at the local switching station. They came out to the AtAT studios again that Saturday and finally replaced a line pair; that essentially gave us a whole new physical line to replace our old one which had somehow "gone bad." Unfortunately, even after the line was replaced, no amount of configuration tweaking on NAS's end or router restarts on our end was able to get that WAN light to stop blinking and give us our connection back.

You can imagine what came next: more calls from us to NAS, more calls from NAS to Verizon, and more dorking around on Verizon's part. As of last Tuesday, Verizon swore up and down that the new line was finally testing clean and that it should carry a signal just fine. So on Thursday NAS sent out an engineer to replace our router and check the line from the phone box back to the AtAT server room. Guess what? Everything checked out fine, except for Verizon's new line (the "clean" one, remember?)-- which actually had two volts coming off it straight out of the connection box. And so, back to Verizon.

Verizon returned to the scene of the crime last Friday morning and checked the line again. Something else got tweaked, because the guy was out there for at least forty-five minutes-- possibly longer, but we don't know for sure because he left without telling us. And then another Verizon van showed up late in the afternoon, and a different guy worked on our box. For the umpteenth time, a Verizon employee told us that "everything should be fine now"-- and yet, of course, the WAN light kept blinking.

Finally, we come to this past Saturday. It was then that we had a long conversation with a very nice but very frustrated NAS tech who filled us in on the incredibly ugly situation. Basically, Verizon did something that messed with our original line when they replaced some equipment down the street. The problem was that the replacement line was functionally longer than the original one. DSL speeds are directly related to the length of the line from the customer location back to the phone company central office. Our original line tested out at 14,000 feet, thus allowing us our 768K connection. The new line was effectively 17,000 feet (due to lossier cable, or higher-resistance equipment, or whatever), which, apparently, won't carry a 768K connection no way no how.

Just to prove that this was what was happening, we spent a good part of the day Saturday working with NAS to reconfigure our router to drop the connection speed down to 384K. Once we did that, we had a solid WAN indicator and a real connection, but the packet loss was so bad (about 30%) that it was impossible for anyone to get any actual AtAT content. So we reconfigured again and dropped down to 256K. We finally had a solid, useable connection-- at 1/3 our original bandwidth.

Now, here's the rub: apparently, since the line is (finally) clean, Verizon is under no obligation to fix the distance issue. The company's only requirement, we're told, is to provide a line with a functional distance of 18,000 feet or less, which they have done. So AtAT's basically stuck at 256K, because even if we were to switch DSL providers, they'd all be limited by that new 17,000-foot Verizon line. The implication, of course, is that anyone with DSL faster than 256K might one day discover that their connection is dead due to phone company antics, and once the problem finally gets fixed, they are then only capable of much slower speeds. We doubt it happens often, but unfortunately, it can happen, and well in line with our usual luck, it happened to us.

So that's the scoop right now. We're going to limp along at 256K and see if that bandwidth can handle the traffic, and if it can't, we're going to have to investigate other non-DSL possibilities. (Any independently-wealthy AtAT fans out there want to donate $17,000 a month for a T3?) In the meantime, please be patient with AtAT's performance, and let us know if the speed is too slow to bear.

Since this is getting long enough to be an Oscar acceptance speech, we should probably wrap things up by thanking all of you for bearing with us through these "interesting" times. We also want to thank Nico and the rest of the freakin' awesome MacAddict gang for kindly posting AtAT's episodes for a week so that AtAT junkies could get their daily fix. Lastly, thanks to our families, our agents, Steven Spielberg, and all the little people who helped us along the way to make this moment possible. You like us! You really like us!

One of the downsides to a weekday-only broadcasting schedule is that when something exciting happens on Friday night, we're pretty much relegated to playing the less-dramatic "follow-up" role on Monday, because everyone's already heard the news. That said, we're still almost as excited now as we were on Friday evening when faithful viewer George W. first told us that Apple had finally released the long-awaited Mac OS X 10.0.1 updater. Why? Because the devil's in the details, and this seemingly minor update has made Mac OS X a lot more fun to use.

We aren't entirely up on what exactly 10.0.1 is supposed to fix-- it certainly doesn't add DVD playback or CD-burning capabilities, for example (not that you should have expected it to), and Apple's own terse description simply states that it offers "better support for third party USB devices, Classic compatibility and overall application stability as well as support for the popular Secure Shell service." Oh, and if you're a Japanese user, you can also get "an update to the Epson printer drivers." Now, personally, we at AtAT don't use Epson printers, we don't use Secure Shell, and the only third-party USB device we ever hook up to the PowerBook is a gamepad for use with Virtual Game Station-- and we have to boot into Mac OS 9 to play that, anyway.

But that improved "overall application stability" sure is nice. We've noticed a significant drop in app crashes since installing 10.0.1, most notably in OmniWeb; maybe it's a coincidence, but we haven't seen the 'Web crash even once since we upgraded. Classic applications seem faster, and when Classic is starting up, we can actually do other things-- it's almost like pre-emptive multitasking really works, or something. Everything just feels a bit zippier, like Apple poured a little caffeine in the Aqua. Perhaps most importantly from our perspective, for us, at least, launch times got a lot shorter-- some applications seem to start up in less than half the time that they chewed up in 10.0. And while the volume and mute keys on our PowerBook are still useless, at least the brightness controls work again.

Perhaps best of all, the updater doesn't appear to have introduced any new problems, so thank Steve for small blessings. As for actual bug fixes, well, with 10.0 we had exactly two OS crashes, both in exactly the same situation: we'd wake the PowerBook from sleep and even before the lid was all the way open we'd see the unmistakable lettering of the dreaded kernel panic. Since installing 10.0.1, however, the closest we've come is waking the PowerBook from sleep and staring at a grey screen with an eternally-spinning rainbow cursor. Depending on your optimism level, you might choose to consider that progress. Works for us!

Okay, so maybe you're not as charitable as we are, and Mac OS X 10.0.1 still doesn't float your boat. We admit it: Apple's still got a long way to go before Mac OS X approaches even a tenth of the polish and functionality of Mac OS 9. It may sound harsh, but it's true. Then again, the classic Mac OS has also had the benefit of seventeen years' worth of tweaking and repair, so it's no big surprise that Mac OS 9 currently has an edge over Mac OS X as far as overall productivity and consistency are concerned. That said, given how young it still is, we're extremely impressed with just how good Mac OS X is already. Sure, it's sluggish, yes, the apps are scarce, and no, it's nowhere near as fluid as Mac OS 9-- but it's the future, and it's here now.

Still, "normal people" are probably going to want to stick to Mac OS 9 until X is more mature-- at least mature enough for Apple to feel comfortable pre-loading it on all systems, including iMacs. When Apple thinks Mac OS X is ready to pass the Mom Test, then it'll be ready for prime time. According to Apple, that's supposed to happen this summer, and unless your synapses are misfiring badly, you're likely expecting the real Mac OS X to surface at this July's Macworld Expo. That will be the so-called "Cheetah" build that is destined to bring Aqua to the masses.

Well, MacEdition's CodeBitch took a brief time-out from her one-woman crusade for Web standards to inform us that the Naked Mole Rat is at it again, and this time his drug-addled musings are all about Cheetah. (Well, okay, all about Cheetah and some hallucinatory "major diplomatic incident" in Apple's parking lot that would likely sound strangely familiar to us if we ever watched the news instead of old "Get Smart" reruns.) Reportedly Apple's top priority with Cheetah is to make it faster; like you couldn't have guessed that from the code name alone. Apparently the Cheetah build will rely on "minimizing I/O times, calling faster APIs and waiting to load services and frameworks until they're needed" in order to fulfill this need for speed.

So, with luck, come this summer Mac OS X will start fulfilling its promise of super-speed in a big way-- just in time for what we assume to be a slew of new Apple hardware that can really make the operating system sizzle. We're looking forward to July in a big way, not only because our one big beef with Mac OS X is its lack of speed, but also because we've been holding off on replacing our desktop Mac until we could get a G4 preconfigured with Mac OS X. Hmmm... time to starting cashing in those penny jars...