Posts filed under ‘Internet’
C’mon In, The Water’s Fine
The blogosphere has been abuzz since last week about Comcast’s (CMCSA) new policy limiting the amount subscribers can download. Starting October 1st, Comcast will limit users to 250 GB of total downloads per month. Violators will first get a warning if they exceed the cap. A second “offense” within 6 months will risk loss of service for a year.
I continue to be amazed at the ISP business. The telecablecos are the only companies I know that limit the use of what they provide, instead of selling you more of it. As I wrote some months ago, the reason is largely due to the fiction of unlimited usage banging up against the reality of limited network design and oversubscription models.
I could rail at how unfair Comcast is being, or how out of touch they are with Internet users, or how ridiculous it is to punish people for exceeding usage limits they can’t measure. But I’ll leave that to other, better minds.
Instead, I’ll point out how Comcast isn’t even solving the right problem. The trouble with its network isn’t so much capacity in bytes. It’s peak speed.
[Let's ignore for the moment that the Internet is a two-way connection mechanism, and think like a telecableco, where the purpose of ISPs is to shove stuff downstream to you. We know different, but bear with me here.]
Ever take a shower at the same time as someone else in your house? What was the result? Yup, low water pressure, and a singularly annoying experience. Now imagine that on a neighborhood scale. Five people on your street decide to get clean at the same time and all you get is a dribble out of the shower head.
So what’s the solution? Well if you’re Comcast, you limit the size of the swimming pool your subscribers can have. Huh?
How many bytes you download is much less important than when you download them. If a thousand people try to stream a movie (shower) at the same time, they only use up 5 GB or so, but the experience sucks, because the speed (water pressure) is reduced for all. Conversely, download 250 GB (fill your pool) overnight when hardly anyone else is online, and you not only get a fast download but you don’t bother others.
Instead of limiting bytes–a poor proxy for usage–Comcast might be better served by limiting speed. Then they’d be in a position to charge different prices for different speed tiers. This would be relatively easy to do by capping modem speeds, would allow more accurate network capacity planning, and would solve the actual problem, namely congestion at busy times.
In other words, charge for water pressure (or size of water pipe), not the amount of water you use. If you want better pressure, pay extra. An alternative would be time-of-day charging, like traffic on interstates, bridge tolls, and electricity usage. (I suspect that would get too complicated for consumers, but you never know.)
Honestly, Comcast isn’t dumb. So why are they capping total bytes? Two explanations spring to mind, both only small contributors in my view:
- It’s easier to simply monitor total usage and kick people off. (Admittedly, most subs won’t run afoul of the new limits any time soon.)
- They’re clinging desperately to the fixed price, all you can eat model of bandwidth, and are loathe to change it unless their competitors do (that assumes they have competitors, of course).
But the real reason is that Comcast and their ilk want to be in the water business, not the pipe business.
Anybody think that the new usage caps won’t apply if the content you’re downloading comes from Comcast? Like, say with the new Network DVR service some of the telecablecos are itching to charge you for? You bet.
I have no doubt that if Comcast provided most of the video and other content you consume over its connections, their congestion problems would magically disappear. They’d probably even be advising you to build a bigger swimming pool.
And reminding you to fill ‘er up.
Disclosure: I hold no position in any of the stocks mentioned here.
Widgets And Idjits
Intel (INTC) and Yahoo (YHOO) recently announced a venture to develop a platform–dubbed “The Widget Channel”–that in effect turns your TV into an Internet thin client. Seth Gilbert over at Metue.com has a great summary.
Developers can write software widgets that can be uploaded to your TV and run in the background. You could check email, share photos with friends, bid on eBay, anything a widget on your Mac or PC can do. All while watching your favorite TV show or sports event.
Of course, you’ll also have to buy a new TV or Set-Top Box (STB) that is equipped with Intel’s CE3100 Media Processor. Good luck with that.
[I wonder sometimes whether anyone ever sees the obvious disconnect between relatively fast new media business development cycles--i.e. "Internet time"--and the much slower frequency with which people upgrade expensive items like TV sets. Many, many firms are vying to deliver a "convergence" solution. Which, if any, will be sufficiently compelling and have enough staying power to become embedded within a sizable share of TVs or STBs?]
Overall, I’m a bit skeptical of this and other similar initiatives.
What I do like is that it’s expected to be a relatively open standard (from the software point of view, at least–you still need Intel processors). Tapping the creativity of the wider software development community is a proven method for both good product and built-in viral marketing. iPhone apps, Google Maps mashups, and Firefox extensions are just a few examples. However, this alone won’t guarantee consumer adoption of the platform, just ensure functionality is available.
At the end of the day, do consumers even want this? Here’s what I think is true:
- People like to use the Internet for a growing variety of things, including watching video.
- People enjoy watching TV, preferably on a TV set. (I’d hazard a guess that most people watch TV with someone else in the room, but typically watch video on the PC alone.)
- Many do some form of Internet activity (surf, email, etc.) while they watch TV.
- Past attempts at interactive TV–at 15 years and counting–have been underwhelming, and that’s being charitable.
What isn’t at all clear, is whether those surfing are paying any attention to the program while doing so. What also isn’t clear is what the other people in the room are doing. Most likely, they’re actually watching the program.
So what happens when the surfer starts fiddling around with widgets, essentially “doing Internet stuff” while others are watching the show on the same screen. Even if the video portion of the screen is undisturbed, wouldn’t that be a bit distracting? Why does everyone seem to assume the surfer wants or needs to use the TV screen anyway? Aren’t they using a computer already?
Sometimes it seems like much of the Internet/TV/PC convergence is a supplier-driven attempt to create a market where there isn’t one. Perhaps it’s simply another self-reinforcing delusion, where media and equipment companies living in an echo chamber of trade shows, developer conferences, and press events convince themselves a market exists where it doesn’t. The 21st century’s equivalent of the videophone–a technology so compelling that consumers must want it. Except they didn’t.
I’ve seen this kind of thing countless times, especially in large, bureaucratic companies like Intel.
Someone somewhere (fairly high up in the management ranks, to be sure) has a brainchild for a new, compelling offering. A sure-fire way to help the company grow and break into new markets. So it’s funded, momentum builds, staff are assigned, and hilarity ensues.
Soon, lower level employees–who actually do the market research and understand what’s going on–figure out the idea is D.O.A. But nobody wants to tell the top brass they’re wrong, or especially that they’re “idjits” (idiots), in a shoot-the-messenger world. Particularly when their whole department was formed around the initiative. Job security will out, you know.
As the old joke goes, as you go up the management chain, crap becomes manure, then turns into fertilizer, which is recast as a way to grow the company. That’s when the flowery press releases begin. Companies rarely issue a release about how the initiative is abandoned some months later when the market fails to materialize.
Is The Widget Channel crap, or dynamic growth? It’s probably too early to tell. I suspect it’s got a decent chance to beat the competition, whatever that means. However, what’s more important is whether there is even a market to win.
Disclosure: I hold no position in any of the stocks mentioned here.
MicroHoo: “Just Resting”
I don’t think this parrot is dead–yet.
Despite Microsoft (MSFT) walking away from the purchase of Yahoo (YHOO), there’s probably still more to play out in this drama. Ballmer will wait a bit for the share price to settle back into the low 20s, and try again, perhaps for even less than the original $31 this time. If Jerry Yang and the rest of the Yahooligans can turn the ship around, perhaps they’ll be vindicated. But don’t hold your breath.
As I commented elsewhere this morning, all of this reminds me of the property currently for sale on my street. The house is in such need of repair that it’s clearly a tear-down. As such, the market values it at the cost of the land minus demolition costs. But the owners refuse to set the price properly, figuring the house has value as a living space (natural, since they reside there.)
Similarly, Wang seems to believe Yahoo has a greater value than anyone else sees. So far he’s been unsuccessful in his attempts at home renovation. But In this case, Microsoft also has a problem: there aren’t any other suitable vacant lots available, and Ballmer can’t afford to wait too long to jump on this one, even if he has to pay more than he wants.
Frankly, both situations will be interesting to watch play out.
Disclosure: I hold no position in any of the stocks mentioned here.
A One-Channel TV
Got to thinking the other day about the last 10 foot problem, i.e. getting media from either your computer or the Internet to the TV. No shortage of available or announced solutions: TiVo, AppleTV, XBox, etc. Even Nintendo is apparently trialing a service in the U.K. to get the BBC’s iPlayer to your living room via your Wii console.
I’m still a bit puzzled about the underlying strategy for this, however. There’s obviously an attempt to differentiate each box, but I just can’t see how it might drive sales by itself. XBox delivers movies, AppleTV allows YouTube access, Wii incorporates iPlayer, DLink’s DivX adapter had Stage6 (sigh). I guess for a select few this might make sense, that the service tail might wag the electronics dog.
Me, I want it all. I want YouTube, Hulu, iPlayer, Veoh, as well as any video currently on my PC. Not to mention whatever hot new thing comes out tomorrow. But since nobody offers a consumer electronics solution to provide all these, I guess I’ll just….wait.
And wait.
After all, who would want to buy a TV that only receives a few channels?
Certainly, one could always hook up a PC to their TV directly. Or use a Media Center PC of some type that provides most/all of this. But that’s likely too much trouble (and technical savvy) for joe sixpack and sally soccer-mom.
Is it any wonder that there is no surefire convergence solution? It the holy grail here simply a browser on your TV?
Until that time, perhaps the PC still needs to be in the equation somehow. After all, everything on the internet is already accessible via the PC, including not-yet-dreamed up video sites and (this is important) easy billing solutions. Maybe trying too quickly to bypass the computer is a mistake. Maybe this is a two-phase process:
- First, something that allows easy, format-agnostic streaming of anything on or accessible by your PC.
- Evolution to a pure Internet TV (or a simple internet front end to the TV).
Why would I ever buy a box that didn’t do at least one of these things, unless it had another purpose (e.g. gaming, DVR)? The problem, of course, is that the industry is trying to manage the profit stream by linking boxes to services, cutting special content joint ventures, building new advertising paradigms, etc. That approach largely ignores the consumer.
Where would the television industry have been if RCA had cut a deal to deliver ABC content and local news, while Panasonic TVs showed only CBS and the weather? What if Sony VCRs had only played movies from Disney and Sony?
Disclosure: I hold no position in any of the stocks mentioned here.
No Waders Needed
I’ve commented before, here and also here, about P2P traffic, net neutrality, and broadband ISPs.
Nate Anderson at Ars Technica has a long, detailed, but very comprehensive post concerning the Internet and capacity concerns, including whether the Internet will crash due to a catastrophic “exaflood” of data.
He interviews Andrew Odlyzko, of the Digital Technology Center at the University of Minnesota. I met Andrew some years ago, and he is as authoritative as they come on the Internet.
For those lacking the time, here are the Cliff Notes:
- To paraphrase author William Gibson, Internet capacity is more than adequate, it’s just not evenly distributed.
- Upstream bandwidth from consumers is the key bottleneck, but gradually the ISPs will be forced to make their networks more symmetric.
- While it’s true a small minority consume the majority of local bandwidth (hello, ever hear of Pareto’s rule?) it turns out it’s not always the same minority.
- There’s as least as much politics in the Net Neutrality debate as there is business. Or science.
While the concept was not discussed in the article, I still maintain that in the absence of competition, U.S. providers will begin moving to a pay-per-byte model as capacity strains in local networks increase. But this could change if the telcos and cable providers butt heads in more markets, or some form of wireless gains traction at scale.
Ultimately, populist politics and public opinion will continue to vie with telecableco lobbyists and PR experts before meaningful progress in the last mile occurs.
Anyone who really wants to gain a better understanding of this issue would be well served by reading the whole 4-page article.
Will The Madness Never End?
Am I the only one out there who’s sick and tired of all the speculation about Yahoo and Microsoft?
Will Steve and Jerry tie the knot? Is Rupert going to swoop in and rescue the fair maiden? Or is Jerry destined for the arms of another, like say AOL, or Google?
Who cares, really? Yahoo shareholders ought to. Take the money and run, that is. No matter how much “synergy” there is, or how much sense this makes strategically (for either party), these things typically work out only one way–value gets destroyed, and some upstart comes in and disrupts the big guys.
Oh, they’ll hang around for awhile, sheer weight will see to that. But in a couple of years the market for search and advertising will look completely different than it does today–and neither Yahoo or Microsoft will be at the top. So why bother?
Let’s just get this over with one way or another, and go back to our regularly scheduled useless tug-of-war: Hillary and Barack.
Ride The Wave
Dan Rayburn has another nice article out this morning, this one detailing his recent visit to Akamai Technologies (NASDAQ: AKAM). It’s generally very positive about the company, and given Dan’s readership, I wouldn’t be surprised to see a nice bump in the stock today, all other things being equal.
In past discussions with Akamai, management has hinted to me that most of the analysts covering the stock don’t completely “get” its business model and in particular, its sustainable advantages. To do so requires a fair amount of technical acumen, and so I’m not surprised.
I do not see Akamai being disrupted in the short-term. Dan does a nice job in his piece of dispelling the myth of a CDN price war that has kept a lid on its stock price over the past couple of quarters. Certainly others such as Limelight Networks (NASDAQ: LLNW) and Level3 (NASDAQ: LVLT) can deliver content more cheaply, but not at the service levels many clients require. Really, it’s in the [delivery x service] product where Akamai shines.
One day, transit bandwidth may be an order of magnitude cheaper, and everyone will have 100 Mb/s or more of sustained bandwidth entering their homes (I wish). But that time is far off in the future. As long as Shannon’s Law holds, and people continue to ramp their media consumption via the internet, there will be a need for the kind of optimizing technology Akamai provides.
In the intermediate term there’s nothing I see that would indicate severe headwinds for Akamai. They recently won an intellectual property suit against Limelight, and in general have a solid IP portfolio. They are the market leader, have very high profit margins, and continue to dominate the high end for CDN and application acceleration services, both of which are expected to enjoy double digit growth rates in the coming years.
My view is that they’re currently undervalued (though not drastically) and should enjoy steady appreciation for some time.
Pando Bears
Dan Rayburn has a note out this morning about the lack of traction that P2P vendors such as Pando Networks and BitTorrent are experiencing with Content Delivery Networks. Dan’s the most knowledgeable guy I’ve read about CDNs, and I agree with him here. P2P certainly has its commercial uses, but as Dan points out:
From what I can tell in the market, P2P is not as big of a story as it was at the end of last year. The topic has cooled off a bit except when its being discussed as it pertains to carriers blocking or filtering of P2P based traffic on their networks. Aside from that, customers are not asking me about P2P and 55.2% of those we surveyed about their content delivery needs said they did not plan to even look at P2P as a delivery solution for 2008.
Cost is usually touted as the primary reason to use P2P for content delivery, and as I’ve argued before, this won’t scale–ISPs will eventually demand their pound of flesh, one way or another. Plus, as Dan says over and over, cost isn’t even the most important factor for most content providers. My view is that P2P will eventually take its place as a valuable niche method for video delivery, and several of the larger players will gain traction (and/or be acquired). But it’s likely to remain a total delivery solution only for file traders and small content owners.
The Broadband Salad Bar
The Wall Street Journal reports this morning (article, paid subscription required) on talks between Comcast (CMCSA) and BitTorrent, Inc. to make nice over Comcast’s admitted blocking–excuse me, “delaying”–of bittorrent generated traffic.
At first glance it seems a rather ham-handed attempt to mollify Comcast critics. The WSJ and others are reporting it as a fairly straight up development, if something of a backpedal by Comcast. But these miss the point, I think, sidetracked by conspiratorial discussions of “net neutrality” and how the big bad telecablecos are angling to limit our choices, take over the world, and generally do evil.
Do I believe the big ISPs want to control content (or access to it) as well as the pipes? Absolutely. There’s gold in them thar hills! None of them want to be limited to connectivity or bit delivery services. But will they succeed? Is the above fictional ad to be our broadband future? I doubt it. Anyway, it’s beside the point.
There’s another motive for ISPs to manage bandwidth on their networks–it’s a finite resource. No, Virginia, there is no unlimited bandwidth. You can talk all you want about Moore’s Law, etc. Bandwidth ought to be cheap. In a perfect world–or Asia–it is cheap. But not here, not without real competition.
The telecablecos have promised this always-on, flat-rate, high-speed internet access. It was a fiction created by the need for market share, and by consumer demand for a broadband salad bar. All-you-can-eat, with a fixed price. Sounds wonderful. But it could never have lasted since the infrastructure is largely shared and was built on the expectation that demand would be low (or at least intermittent).
Now the growth of video is stealing the condiments, and file sharers are sneezing in the salad.
There’s no free lunch. So you do one of two things. Limit how many trips each patron can make for salad. Or charge them for each trip. Comcast has tried the former. My bet is on the latter.
You believe it’s a fantasy? That no one will go for tiered or bit-based pricing? Think again–Time Warner (TWX) is already trialing it. And the more successful we are at regulating the ISPs to ensure net neutrality (in the absence of competition), the more likely it is we’ll see pricing by the byte.
Joost Passin’ Through
Hey, it’s always nice to be right, even when it isn’t for exactly the reason I predicted. Lots of fodder recently, here and there, about how P2P video startup Joost is becoming increasingly irrelevant in the land of VeohHuluTube.
In addition to its P2P strategy not scaling, Joost’s software client ran slow on my machine, and apparently many folks resist having to download and install it. But going to a more mainstream web or browser-based presentation won’t save the company either. It makes them no more compelling than any of the other portals (remember that word?) angling to become video “on ramps”.
In fact, it seems to me that all of these sites–including the much lauded Hulu–are not where the money is. It’s the content, dude. Content owners now spray their wares onto any website it’ll stick to. Site owners have no leverage left, especially as the advertisements become more attached to the content and less to the site. How would ABC or CBS have ever differentiated themselves if you could have found the same shows on any channel you switched to?
Only the sites that provide something unique (social connectivity is the big one that springs to mind) have a chance. Joost seemed to have recognized that–with a built-in capability to chat with other viewers–but its technology and presentation prevented it from catching on. Now it’s probably too late.
by Michael Mauboussin
