1. My mom saw a Verizon truck in the county fair parade today (I didn’t go; saw no need and was tired). The truck has “High Speed Internet coming soon!” or similar painted on te side. This is DSL, not FiOS (Verizon always refers to their fiber offering by its trademarked name). However from what I hear Verizon’s DSL release is just a ploy to drive up the sale cost of their system here; Fredericksburg isn’t a large metropolitan area, and there are too mayn bridged taps and other phone system anomaies here for DSL to reach beyond a small part of town. At least that’s what the scuttlebutt says. My prediction: Verizon offers some flavor of DSL to folks who can currently get Windstream DSL, then sells the system to whoever will take it, using the “we have DSL in our central office” argument to jack the price up slightly above what the system is worth.
2. Bee Creek Communications, the local our-hands-are-tied-and-if-anyone-competes-with-us-we-lose wireless ISP, was in the parade today as well. They also have a third-of-a-page ad in the paper advertising free installation to new customers until September 15th. A few years ago an installation with similar equipment would have run you around $220. The problem here is that their network is already over capacity in some areas, they’re spread too thinly in terms of manpower and their speeds are the stuff of yesteryear, barely competitive with satellite internet (though their monthly transfer caps are higher). The service also requires a two-year contract with a $125 early termination fee. If you want a one-year agreement service will run you another $10 per month. In short, don’t buy.
3. Clearwire is launching WiMAX service in San Antonio and Austin on September 1st. Maybe they’ll bring their service here. Would put Bee Creek right out of business, since Clear has an uncapped-download, one-megabit-upload (or is that uncapped too?) tier for a mere $45 per month. They even have cheaper tiers to compete with your garden-variety DSL options. They even have business tiers that will fill the gap between Bee Creek’s woefully slow packages and Time Warner Cable’s woefully expensive ones. In a nutshell, something for everyone as long as you’re within range of one of their towers. I’ll be keeping close track of what happens here. For what it’s worth, I told Bee Creek about Clear’s possible imminent launch nearby, and I wouldn’t be surprised if their larger-than-usual print ad was a direct result of this.
4. Hill Country Telephone Cooperative has run fiber around the city of Fredericksburg and plans to get CLEC (competitive local exchange carrier) operations started before the end of the year, or something close to that anyway. They may just focus on high-end business operations (like they’re doing in nearby Kerrville, where they recently connected the city’s independent school district) however they’re a residential provider in their home area, so they know how to deal with “normal” customers. Plus, they have a lot of options for deploying service; in addition to their new fiber-optic network they have an AWS (1700 MHz cellular) license for this area, a 3.65 GHz “light license” which they’re using for WiMAX service in the nearby town of Mason, and maybe (if the price is right) Verizon’s old copper network, though I’m not sure why they’d want it. I heard HCTC offered $50 million for the network a few years back, and Verizon wanted $80 million fot it. Knowing that the network wasn’t worth that much, HCTC walked.

In short, it will be interesting to see what happens around here in the relatively near future with regard to internet service. Right now Time Warner Cable is the only provider who will give you more than a 2 Mbps connection (okay, HughesNet will give you 5 Mbps but they don’t count due to obscenely low caps and ridiculously high monthly fees). They’re also the only provider around here with customer-available fiber; Verizon apparently doesn’t have any. Sounds like that may change soon though. I’ll tell all I know (or am allowed to tell, whichever is less) as I find out more.

So the big question in my mind is, how would I design the perfect set top box, from both a consumer and a provider point of view? Here are some specs I came up with:

1. nVidia ION chipset. Everyone says the thing can decode 1080p video with no problem. Paired with a decent operating system and a halfway-decent processor (like the low-power Atom 330) you’ve got an STB that not only performs well, but also can be used for a variety of things and isn’t going to run up the power bill.
2. CableCard support, multiple streams. Granted, this is a given these days, but why not make things clear? RF video (via GPON) is what I’m assuming as the last mile medium; IPTV would work a bit differently of course.
3. Easy-access 12.5mm height 2.5″ hard disk bay to turn pa plain set top box into a DVR. Pop in a bare SATA drive and pay a one-time upgrade fee, then start recording shows. Or pay a slightly larger fee plus the price of a handpicked drive and a tech will do all of that for you. Bottom line: there should be no hardware difference between a regular set top box and an HD one, aside from the presence of a hard drive, or lack thereof. If you’re wondering why a 2.5″ drive, it’s because the drive is smaller and takes less power. If people want to store gobs of content they can add an external drive. There might come a time when all video content is cached at the edge of the network (networked DVRs) but until then, let people have as many bits in their box as they possibly can install.
4. Flash storage for non-video, namely the operating system and program guide. You want that stuff to load quickly, so why commit it to spinning metal when you can get enough flash to hold it for under $20?
5. HD on every STB so you can say that there’s no extra equipment required. Plus, having a single model of box makes things a whole lot easier to keep track of.
6. 2 USB ports. Aside from easy testing of devices, this would allow people to view media on their STB by plugging in a memory card reader or something equally quirky-but-cool. More importantly, people would be able to connect a whopping 2TB drive to their STB if they wanted and record shows onto it. The more shows they have recorded, the longer they’ll stick around to watch ‘em, and the longer they’ll pay for service. Also, you can hook…
7. …an external Blu-ray player to one of those USB ports. Make one that fits in with the decor of the typical media center and you’ve got yet another feature that makes that STB “sticky”. Why not? The onboard processor can handle it.
8. Whole-home functionality. No need having multiple DVRs recording the same thing when you can just pull between them. That way one DVR does all the work and the rest sit there with less stuff to break.
9. TiVo as the operating system, with no TiVo monthly fees. If you’re a service provider, you can probably work with TiVo on lowering the “price for life” subscription per box. Also, why use any other piece of software than the best? People say “I TiVoed that” just like they say “I googled that” so why not take advantage of the latent mindshare?
10. Farm out on-demand as much as possible, with an “you’re on our awesome box” commission of a few percent. Nobody wants to spend $8 on a cable-provided movie on demand. Though on the other hand some sports stuff has to be done that way, at which point just download the on-demand stuff over the internet rather than via RF.

The above list might make for an expensive” dumb” set top box, but then again it might now. I’m pretty sure that, when purchased in bulk, all the above components would total less than $200, so if you charge $8 per month per box with the first one free you’re making your money back before the boxes are obsolete. Plus the fact that you’re using a single box across your network will make things a lot easier to deal with for customer service. I’m pretty sure any provider who instituted all of the above would have people running from competitors, even if the service did cost a few bucks more per month. Then again, I’ve never had pay-TV in my life :p

What had really happened is the modem has been kicked off the network for a bit, been reassigned a new IP address, and I had forgotten to set up DynDNS again on my router (easy to overlook when you’ve had the same IP for months on end). So in reality the internet at my place was hunky dory…I just didn’t know where “my place” was on the ‘net.

So what was the cause of the modem reboot and subsequent switch from 76.xxx.xxx.xxx to 67.xxx.xxx.xxx for my IP address (people with 24.xxx.xxx.xxx addresses probably didn’t see a change; they’re still around here)? A DOCSIS 2.0 upgrade, as a matter of fact.

How do I know that the upgrade was to DOCSIS 2.0? I can’t access my modem configuration/status page; the modem in question (a Zoom 5241) has those funcctions walled off for whatever reason. However I *could* do a speed test, and Comcast’s own test revealed a near-shocking 7-11 Mbps on the upstream, a far cry from my typical 2.6-3 Mbps, limited by the DOCSIS 1.1 technology the system had been running on until recently. Another tip-off: Comcast e-mailed me a few days ago, tuting their new DOCSIS 3.0 services as “coming soon”…they’re still not here yet but DOCSIS 2.0 upgrades are one big step closer to that reality.

If you’re wondering why Comcast gave me the upgrade, it’s not because of competition; Qwest’s services are handily beat by DOCSIS 1.1 on Comcast. However if Comcast upgrades their systems to DOCSIS 3 (with DOCSIS 2 as an interim step) they’ll be able to release higher-speed tiers at higher prices, including a 50 mbps down, 10 Mbps up tier weighing in at $116 without TV (versus the $65 I’m paying right now for 8 Mbps down and 2 Mbps up, plus PowerBoost in both directions).

Speaking of rampant capitalism, apparently Comcast wants me to sign up for some of their other services. During my absence, the company sent me two letters. One offered cable for $30 per month (I think it was the mid-tier digital package; I didn’t get further than that since I don’t want cable and I certaily don’t want a one-year contract for the service). The other offered phone service for $15 per month (I already have a cell phone…and yes I know that the phone package is normally $40). Despite these offers, and the fact that the internet portion of m bill would go down if I added another service (TV in particular would drop my internet price by $12 per month), I’m not interested. Still, nice try guys.

Lastly, Comcast has been sprucing up their website portfolio lately. Comcast.net is now shiny and new, as are Comcast.co’s pricing pages. The company has even overhauled their account managment portal so that it now looks pretty cool. I’m pretty sure all these upgrades, combined with the relatively recent introduction of their SmartZone Webmail/web-voicemail system, means every portion of Comcast’s online consumer presence has been reworked in the past nine months or so.

Lastly, while messing around with my modem I decided to test IPv6 connectivity. Not that anyone cares about IPv6 yet, but I was able to ping and trace to a few hosts (he.net, softlayer.com, fdcservers.net). Though I must say, latency was rather absurd, on account of all IPv6 traffic being routed through AMS-IX (AMSterdam Internet eXchange) and SWIPNet before going to those fragments of the internet that are IPv6 enabled. Of course, the situation on that front will improve and it’s cool of Comcast to have the tech in place, but it’s largely a bragging-rights deal at the moment.

Personally, I’d rather have a minimalist online UI and 10 Mbps symmetric fiber with no PowerBoost and IPv4 only for $40 or $50 per month including one branded GMail address, but I suppose I’ll have to pioneer that service myself. For now though I suppose Comcast’s service is becoming more passable every day, and that’s rather Comcastic of them :p

So if you’re a techie and am wondering what the “Networking Green Belt” part of the post is, bear with me. First let me lay out the internet situation in Big Sprint: Alltel EvDO, Basin 2Way as a WISP, Suddenlink for cable, one or two numbers for dialup. Sprint (my typical cellular carrier) is 1xRTT-only here, as is Verizon for the moment (but they’ll be EvDO soon once they integrate Alltel’s network). Not sure yet abut Basin 2Way’s services and pricing, though I’ll bet it’s only economical for folks who can’t get cable. For cable, Suddenlink is the provider; I’ve detailed their pricing plans and traceroutes here, here and here. Please disregard the extraneous discussion about DOCSIS modulations. Dialup is one of two phone numbers to my knowledge, one for AOL and one for everyone else. This is one of the few areas where TOAST.net has no local numbers, depiste their aggregation of dozens of dialup access number providers. The numbers here are owned by SBC/AT&T, but not WorldNet. Fun stuff.

So back to my networking green belt.

The church here has Suddenlink internet, which is nice for me since I thus don’t have to use my phone on roaming to get decent internet speeds; since my dialup plan doesn’t have local numbers in the area my only other option would’ve been using my phone for 1xRTT, which is dog slow.

However the church was using a Linksys WRT54G2 to spread the internet around, and it wasn’t making it to the other side of the building. So I added a Linksys WRT320N to the team shopping list and swapped that out for the WRT54G2. The signal ended up better, but not by enough.

So I grabbed their old router and set about figuring out how to use it as a repeater. The router is rather capable, with 16MB of RAM, though as with other consumer routers the 2MB of flash memory meant I could only flash the micro-plus-ssh version of DD-WRT onto the device. Still, that’s a whole lot better than mere micro; that’s what I have on my home router back in Fredericksburg, a WRT54Gv5. To get DD-WRT onto the router I followed these instructions. Basically I downloaded Linksys’ TFTP utility, then sent the VxWorks prep file, then the killer file, then the DD-WRT firmware to the router. The whole process took maybe ten minutes.

I then went here to figure out how to set the router up as a repeater, something that I hadn’t done before. The setup image was very helpful in getting everything…um…set up, and shortly I had Gateway Baptist G throwing wireless signal into the far corner of the building. Score.

Granted, the WRT54G2 won’t be the speediest repeater, since all traffic has to be passed wirelessly from it to the main router, which is then connected to the Suddenlink Ambit cable modem, but that’s not a big deal when you’re working with a megabit internet connection to start with. The big deal is that everything works, and it works everywhere it needs to. Plus, I learned a bit more about DD-WRT, knowledge I’ll apply when I get back to Fredericksburg to expand my home network by a bit. I think that’s a Win-Win-Win or something…well, the WRT320N did cost too much, but oh well.