Companies that try to do more than one thing rarely do all of them well.
There are exceptions to this, but in general, most companies are highly specialized. They do something, they do it (hopefully) well, and that's how they make money. Everything they do is in furtherance of that goal: what's called in businesspeak a "core competency."
EBay's core competency isn't the same as Skype's. It doesn't really make sense for them to be in the same organization: with such differing strengths, it would be difficult for the corporate leadership to aggressively advance both of them. You'd always have one unit vying for resources and attention with the other. Rather than a cohesive organization, all working together, you'd have some weird joined-at-the-head abomination.
In that situation it makes perfect sense for the two companies to go their separate ways. By divesting itself of Skype, Ebay gets a lot of cash -- cash that it can invest into doing better what it already does well -- and Skype gets its independence.
It's not shortsighted. The acquisition in the first place was (IMO) not very well thought out, but they have a chance to fix their mistake -- albeit at some non-trivial cost -- and cut their losses before they run Skype into the ground and it stops being competitive.
I don't know - mix this with presence information in ebay accounts, and when looking at an auction, you get a link to the seller's skype for text and/or voice chat. That's a good argument for partnering with a VoIP/telephony company, but not really for owning one outright. In fact, the last few decades of accepted business-management wisdom advises exactly against this sort of acquisition. Just because you have a need for something, doesn't mean you're necessarily the best person to supply it, even to yourself.
Telephony was nowhere near eBay's core competency, and the match never looked that great. And even if they were hell-bent on getting into VoIP, there were probably other, better companies that they could have acquired, ones that would have provided more value -- companies that specialized in linking IP services to existing phones for instance. (So an eBay merchant could plug in their existing phone no. and have buyers be able to click on a button on an auction and call them, or something like that.) Not to mention, pretty much any of them would have been a lot cheaper than Skype.
The Skype acquisition always looked like something that had been decided on first, based on some sort of 'gut feeling' driven by hype or wishful thinking or a sense that they were 'missing out', and then the reasoning for it was backfilled in later. (As I think a fair number of Web 2.0 acquisitions were.) Turns out, that's not a really good way to run a business.
> Otherwise, all courses would be graded on the basis of a 100% final exam.
That's sort of how law school works. Sure, you get course grades, but the whole point is to prepare you for the Bar exam, and for a future career as a lawyer. The proof is in the result.
If Comcast wants to limit the service provided to most consumers, then they can just write that down in the service agreement. Um, they do. From the Comcast Terms of Service: (under "Additional Limitations On Comcast's Liability For Hsi")
Facilities Allocation. Comcast reserves the right to determine, in its discretion, and on an ongoing basis, the nature and extent of its facilities allocated to support HSI, including, but not limited to, the amount of bandwidth to be utilized and delivered in conjunction with HSI. The problem isn't that they're doing anything that's not allowed in their agreement, the problem is that their 'agreement' allows them to do just about whatever the fuck they want, and if they're the only HSI provider in your area, you're screwed if you don't want to let them have you by the balls.
Because consumers don't like that very much, Comcast is -- very slowly -- beginning to feel the heat from government. Reforming their Terms of Service to read less like an indenture agreement would be a good first step towards regaining some lost trust. But until they do that (and until they remove the sections that let them change the agreement at will), nothing they say or do has any meaning or value.
Amen. The thing I miss most about my Thinkpad versus the Dell I have now is the TrackPoint.
To me it seems like a no-brainer on a subnotebook, since it doesn't take up any additional space outside of the keyboard. A trackpad just seems like a terrible choice when you're space-constrained. I've never liked them (I liked mid-size trackballs, like Apple used to use on the 1xx series, better), and they only get worse the smaller the tracking surface. Why the Cloudbook designers went with what they did, instead of putting a TrackPoint/nipple in the center of the keyboard, boggles me.
The cheapest subnotes I've seen with a TrackPoint are all over $1k. There's the Asus S200N, but that's selling for something like $1400, and that's assuming you can get your hands on one in the U.S. to begin with; the Dell Latitude D430 also has one, but I think it lists at like ~$1100.
(The Palm Folio was supposed to use a TP with a scroll wheel, and would have been in the $500-600 range, but unfortunately it's DOA.)
Hopefully the UMPC people will see the light eventually.
While I agree that they're a bunch of lying fucks who shouldn't be trusted further than they can collectively be thrown, I'm not sure that
Anything they propose will not be binding and will not have the force of law. is necessarily the case. If they included their "Bill of Rights" as part of the contract of service, then it would be enforceable through contract law, just like any other part of their agreement is.
(I'm a bit rusty on the details, but I've been advised at various times by lawyers that there are situations where a company can be held via contract law to statements made outside the contract itself, if they basically define the relationship between the company and the customer. I doubt Comcast's lawyers are stupid enough to walk into this trap unknowingly, but you never know.)
Although I very much doubt that Comcast is acting in anything approaching good faith here, it's not impossible for them to make the Bill of Rights binding, if they were sufficiently motivated.
What needs to happen is that we, as users, need to make sure that Congress and various state legislatures aren't distracted by any sort of non-binding agreement on Comcast's part. If they want to avoid burdensome regulation, they can come up with a 'Bill of Rights' and then hold themselves to it contractually. But if they don't do that, or if they put it in their contract but then leave in a way of unilaterally amending the contract, it's not worth two squirts of piss.
I've been thinking for a while about purchasing one of the subnotes. Not as a primary computer or even a primary notebook, but as an alternative to the giant and totally impractical company-issue 19" notebook.
The EeePC 900 seems like it may be the leader right now; I was intrigued by the original EeePC but having that tiny screen surrounded by a giant bezel smacked of wasted space. I don't mind using a 7 or 8" screen, but if I am, I want the whole device to be that size. If it's big enough for a 9" screen, I want the screen to actually take up the available space.
The Cloudbook (aka PacBell EasyNote) sounded neat at first, and apparently it's got more metal in it than the Eee and feels more rugged, but the reviews of it have been pretty terrible. The one thing that really turned me off from it, oddly enough, was that its SD Card slot doesn't let the card fit flush. That just smacks of poor design; when a product has something that obviously dumb on it, it makes me wonder what more fundamental flaws it has under the hood. I hope they'll do another revision and try harder. (Ditching their thumbstick for a IBM-style keyboard stick would probably be enough to send me reaching for my wallet.)
The other feature that I haven't seen mentioned much so far, is support for connecting to wide-area data (cell data) networks, rather than just Wifi. Apparently Linux support for most WAN cards is pretty dismal, but Bluetooth (so you could go computer->cellphone->WAN) would be a nice plus. The smaller and lighter a computer gets, the more I think you want persistent connectivity.
But overall it's pretty exciting to be seeing machines for under $500 in a form factor that you recently had to pay upwards of $1k to even get into (even if all you wanted was the small size, and not any of the other stuff that comes on the very high-end subnotes).
The other nice bonus of computers getting cheaper is that it's becoming more and more apparent how much Windows actually costs. Sure, the Windows and Linux versions of the Eee 900 are the same, but if you want Windows, you only get 12GB of storage instead of 20. Even if I hadn't already decided on Linux, that would give me significant pause. (Although I suspect there will be a lot of Linux versions sold that just get wiped and replaced with pirated Windows installs; but hey, at least they're counted as Linux preinstalls.)
It's fundamentally not a fair comparison. For example, per kilogram, hydrogen has a lot more energy than gasoline. And fuel cells are more efficient than internal combustion engines. Of course, to make that hydrogen, a lot more energy was wasted than when making gasoline. And that hydrogen is bulky, hitting range. But that wouldn't hurt it here; by your rules, hydrogen vehicles would win easily. I don't know exactly what the rules were in this competition, but in some that I've read about, they compensate for those factors.
(I'm basing this off of my recollection of a Wired magazine article I read a while back.) I think what's generally done is if you run on something other than gasoline, they convert your energy source to gasoline-equivalent using predetermined conversion factors. Those conversion factors try to take into account generating the energy from a base source, and providing it in a form that can be used. It's sort of an average generation/transmission cost.
Obviously there's a lot of room for argument depending on how you fix the conversion factors, but I think even when you consider the "well to wheels" efficiency, they're still very competitive with gasoline engines.
Also, it's not as though gasoline is free from production costs. Just like hydrogen, it takes energy to extract and produce gasoline from raw materials, not to mention transport and distribute it. So you if you want to have a fair competition with efficiency as its goal, you have to apply the same conversions to all types of fuel.
All that taken into account, I don't think it's unfair to give something of an advantage to gasoline-fueled vehicles, if the goal is to develop realistic, easily-implementable efficiency-boosting technologies. This competition seems a bit removed from that (cars are too small, too light, etc., to be practical), but it's silly just to ignore the huge existing investment in liquid-petroleum-fuel distribution infrastructure, like it or not.
You're talking about a country which my President has told me has a nuclear weapons program. Something tells me if they're capable of getting nuclear engineers, they can figure out how to make a half-decent fighter jet if they really needed to. A basic nuclear device is cutting-edge 1940s technology. They haven't been really high-tech since television has been in color. That they're not widely available is a testament to the economies of scale necessary to produce the nuclear material, and (much, much less so) the efforts on the part of countries that already have them to discourage other countries from acquiring them. (And also the disinterest of many countries who would be capable of producing them, if they really wanted to; a list that includes almost every developed nation.)
The F-14 represents the pinnacle of aviation technology, circa 1972 or so. There's a ~30 year gap there, and that represents 30 years in which the United States made some pretty significant leaps in engineering technology.
So in short, just because a country can build a bomb, doesn't mean they could build an F-14.
And even if they could build an F-14, it doesn't mean it would be much of a match for more recent aircraft, which have a further 20+ years of development behind them. (Although I think you could argue that the pace of development slowed down dramatically after the end of the Cold War, at least in major weapons systems.)
It's not an unresolvable problem by any means, I just don't think it should have to be a problem at all.
I thought the behavior sucked the first time I encountered it, when it was called "Site Finder" and was being run by Verisign. It caused a fairly significant brouhaha, as many here may recall, and they were eventually slapped down by ICANN.
Although it's obviously more evil to do something like that when you're a default root server operator and not running an opt-in service like OpenDNS, I think the practice is crappy and should be considered broken regardless of who's running it. Bad practice is bad practice.
There are enough other public nameservers and alternate roots that I just can't see using OpenDNS, unless you really want the "added" services they offer (adult-site filtering, phishing protection).
The fact that they have large caches is nice, and they can definitely be an improvement over some ISP's default DNS (which really says more negative about the ISP's DNS service than it does good about OpenDNS), I just don't think it's worth the trouble. YMMV, of course.
I don't know much about them, but I think the "Open Root Server Network" might be a possible candidate. It's an alternate root system, independent from but currently mirroring ICANN's, located mostly in Europe. (The sole non-Europe rootserver seems to be run by Paul Vixie, actually.)
I gather from the Wikipedia page their major concern is monopolization of the DNS root by the U.S. Government.
Their site has instructions on switching to their roots, if you run your own DNS server, and a list of publicly-accessible DNS servers that use their roots, if you just want to re-point your workstation or router.
If they're not your style, WP has a list of other alt roots; most of them seem to revolve around the idea of having more or different TLDs than ICANN. The ones I'd probably consider first would be OpenNIC and the Open Root Server Confederation. The latter's website doesn't seem to indicate, at least to a quick reading, their root server addresses or any publicly-accessible DNS servers.
> orthodox MACROeconomics suggests that wages are sticky downwards -- i.e. they don't tend to fall based on an increase of supply...
This seems incredibly doubtful. I could believe it, if you're just talking about gross wages, but that's almost meaningless. Real wages, indexed against inflation, effectively decrease whenever gross wages aren't going up.
And negotiating a salary increase is very difficult in a slack labor market. This is pretty fundamental economic theory, not to mention something that can many people have actually experienced.
This is the situation we are increasingly finding ourselves in: gross wages haven't decreased, but in many sectors real wages have shrunk. If inflation increases, as it seems poised to do, this may get worse. And having a labor surplus will only exacerbate the situation.
The only positive effect I can envision of such a surplus is that it might act as a brake on inflation, by preventing workers from negotiating salary increases as they might in a tight market, thus preventing a positive-feedback loop. However, I'm not sure whether having the surplus will be beneficial in the net or not, especially when you consider the huge costs associated with maintaining a large number of non-productive workers (unemployment, welfare, retraining, etc.).
Also, labor surpluses have a history of causing political and social volatility. It strikes me as odd that we're intentionally creating one in the U.S., since I doubt it's something that most voters would support if asked.
It's an alternate root, not a proxy server. I don't have the hate-on for OpenDNS that the GP does, but it does have several weaknesses as a service which caused me to stop using it.
The biggest problem, and one that the GP alluded to, is that OpenDNS resolves *everything* to a sort of 'parking' page. If you're using OpenDNS and you type in a bogus URL, rather than just not resolving, you'll get a redirect to an OpenDNS page. This is, IMO, misbehavior. However, there's no incentive for OpenDNS to stop, because it's on these pages that they place advertising and pay for themselves.
This behavior is particularly obnoxious when you combine it with an additional level of caching DNS. Let's say you have a DNS server on your LAN (like most home gateway/routers) and you point it to OpenDNS. If you're working with a site that may or may not exist -- say one that you're trying to configure -- OpenDNS will give you the parking page if it can't be found. But your local DNS server will cache the redirect, and it can take a while to purge. (I'm not sure what TTL they're set to, but it's evidently longer than it should be.) The upshot of this is that a site can look 'down' even though it ought to be up, because intermediate DNS servers cache the bogus OpenDNS result, rather than just failing to resolve.
I think it's great that there's an alternate root, and I really like that OpenDNS exists. It's a great concept. I just think their execution deviates from accepted practice and standards, and that's no way to run a DNS server. Too much rides on it.
Although I think it's a fairly big 'hop' to go from understanding interpersonal differences to deciding that one person is objectively better than another, even if you do get to that point, there's still a wide, gaping chasm before you get to the point of forcibly "fixing" people. Action is pretty distinct from having some knowledge or holding a certain opinion; it's a totally different category.
Certainly we need to be cautious and discuss these sorts of issues as we advance our knowledge of biology and the human body, but I think it's also pretty clear where we can draw the line. We don't have to wait for science to get there before deciding some things -- like forced gene therapy -- are just not okay.
However, we should be cautious not to overreact and throw the proverbial baby out with the bathwater. Lots of technologies create possibilities for abuse, and biotechnology isn't really any different.
Interesting. This is the first time I'd heard any hard, realistic-sounding numbers.
I suspect if Microsoft really tries, they will probably be able to rally a little bit with Vista's successor ('Longerhorn'?), but the writing is sort o on the wall for them as a company. They can't possibly go anywhere but down.
There's just no place to go! They owned an emerging market and rode it all the way. MBAs will be studying them until the Earth turns into a cold, dead cinder. But only someone suffering from a great deal of delusions would have ever said that it could go on forever in its current form.
What they do have is a whole lot of cash, so perhaps they can turn themselves into a high-tech version of GE Capital (only, with less suckage). They'll need to learn to stop sucking the life out of everything they buy, but I think that's probably manageable. Hell, if IBM could reinvent itself as a software-and-services firm, Microsoft can definitely pull off becoming a big incubator, if they wanted to go that route.
But that's beside the point: their OS marketshare is headed south and I really can't see a scenario that would recreate the environment that made their 1990s takeover of the market possible again, for them or anyone else.
Somebody once said something to me about the technology sector that I thought was pretty smart. It was "the technology always moves surprisingly fast, and the business always moves surprisingly slowly." Most predictions of the future by technologists or IT people vastly underestimate the amount of time that the rest of the world will take to catch up.
Right now we're seeing the technology move into place, with the desktop OS fading into the background of what makes a particular "device" as a whole, and thus becoming commodified, but I think it'll probably be a long wait for the market to realize that the Age of Windows has had its peak.
Can you please explain how that is so? Reading countless economics text books about the benefits of division of labor have confused me. Try at least a little harder than that.
Imagine a widget factory. The factory takes in raw materials, and produces finished widgets. The widgets are sold on the market for some price that exceeds costs, resulting in profit. The workers are paid a salary, which they can use to buy widgets. With the exception of possibly exhausting whatever raw materials are used to create the widgets, you can repeat this wealth-generating cycle forever. (I.e. it's not some sort of closed system, and it's not zero-sum; you're creating wealth by adding value via the raw-materials-to-finished-products process. There are other processes that create wealth, this is just the most obvious.)
Now, we outsource that factory to Somewhere Else, but continue to import the widgets to satisfy domestic demand, perhaps at a lower price. Now, consumers buy their widgets from Somewhere Else, meaning that wealth flows over there. At the same time, all the people who work at the widget factory are unemployed.
Do you start to see a problem here? If you can't find something else for your former widgetmakers to do, you end up just draining money out of your economy. If you have modern finance at your disposal, you can conveniently spend more wealth than you actually have, issuing debt and importing stuff; at least you can until people stop wanting to buy your debt. This isn't sustainable. Eventually you either literally run out of hard currency (the case if you use gold or something else that can't be created), or people decide to stop buying your debt. And then you have a bunch of angry, unemployed ex-widgetmakers who can't afford to buy widgets anymore. Problem.
Of course, there are cute responses to this. You could argue that this is just the way things are supposed to work -- if the widgetmakers couldn't compete, they deserved to go out of business. Fair enough, and that actually makes a certain amount of sense.
But suppose you have an entire nation of widgetmakers? An entire nation of people who have built themselves a nice lifestyle (oh, and by the way, a huge fucking quantity of nuclear weapons) for themselves, making widgets, and suddenly end up unemployed? What do you expect them to do, calmly and rationally reduce their standard of living so that they can compete better on price? I don't think so; not when they have the ability to go and take a lot of wealth via brute force.
Well, if you want to talk about 15 years down the road you might as well mention that in 15 years all the demand from our outsourcing will make the Chinese as well off as us, forcing them to charge as much, canceling out any benefit of outsourcing there.
Not a chance; not with the population they have. Maybe in a century, but fifteen years? That's ridiculous. There are millions upon millions of people in China (and India, and quite a few other places) who have grown up and are used to far cheaper standards of living than the average person in the U.S. That translates into dramatically lower labor costs for the foreseeable future, since they're going to be willing to work for less. Someone who remembers life in a mud-and-thatch hut on a rice paddy is probably going to have a markedly different bar for 'success' than someone who grew up in the U.S.'s heyday and expects to be able to do better than that.
You're a little capitalist, and you don't even realize it. Want all the jobs to stay in our country? That's greed; the same thing driving those shareholders to make more money. Unfortunately, whining doesn't get much done, so we'll all have to work really hard and offer some kind of advantage to keep the jobs. It's called "competing".
That's a great thought but it's a little lacking in substance. What do you propose the U.S. ought to specialize in? I'm quite honestly interested, and I've asked this question over and over to a lot of fairly intelligent people and have yet to get a satisfactory answer back. I'm not sure there is one. Do we try to go the Neal Stephenson route? Music, movies, microcode, and pizza? Other parts of the world are chipping into 'software' already, and there's no reason to think that we have some kind of automatic, natural, competitive advantage in any of those.
About the only thing we do have here in the U.S., at least at the moment, is a hell of a consumer market. Until we figure out exactly how we're going to keep ourselves going, I don't think it's necessarily illogical to want to carefully manage access to the one thing of value we have left. I'm not proposing or advocating for complete isolationism, just a careful analysis of exactly who we're allowing access, and to which markets, and what the effects are.
More bluntly, I don't see any reason why the U.S. ought to open any market to foreign competition unless there's a clear indication that opening it results in a net benefit to the United States. Now, it may be that fully-open markets are the best (or least-worst) policy for Americans in general, but I haven't seen any of the politicians pushing for open markets really going out of their way to demonstrate this. And from where I'm sitting, it looks a lot like we're just letting ourselves go bankrupt on imports without much of a thought towards the long-term sustainability of this situation.
Even if by restricting imports it increased the cost of non-essential goods to consumers, but in doing so bought us a few more years or decades of solvency in which to work on our comparative advantage (or for the Chinese and other developing markets to bring their labor force's standards of living, and thus costs, closer to par), I can't see why that would necessarily be bad.
National governments have a mandate to serve the best interests of the people they represent. If free trade and open borders are demonstrably the best path, I'd be more supportive, but right now they look suspiciously like a path that leads off a cliff.
> But I remember all the old Dell commercials - the main thing they had going for them was customization.
I think the point is that those were the old Dell commercials. If you look at ones today, they're all about price. Features and price, admittedly, but price is the biggest thing.
This is a reflection of the market for PCs. When they represented a substantial capital investment, you wanted to tailor them to your particular needs, and avoid paying for anything you didn't absolutely need. That made customization and U.S.-based assembly locations worthwhile. Now, people don't want that as much. The PC, as a unit, has become increasingly commoditized. I bet a lot of buyers today don't even look at specs; they just buy "a computer" and make a lot of assumptions about what they'll be able to do with it. (Assumptions that are actually pretty safe if you don't plan on doing much beyond typical consumerish tasks with it.)
As a result, the goal is no longer "build me a PC to my exact specifications," it's "build me as much PC as possible for $500". Or $300, or $250. I suspect before too long it'll be $99.
That doesn't favor having a lot of assembly points close to consumers; it favors doing all your assembly in a quasi-slave-labor camp somewhere, to better keep costs down, and then shipping tons and tons of identical boxes in bulk to wherever the consumers are. 'Who cares if it's not exactly what you want? It's $500 and it's more power/features/speed than you'll probably need, so just buy it,' is the message.
It's easy to blame Dell here, but it's buyers of technology that are driving it. Not enough people want essentially bespoke computers (or the ones that do aren't buying them from Dell), and Dell is going to eliminate the facilities that provide that service.
You're ignoring that software is already protected via copyright. So particular implementations of a new idea are automatically protected -- and not for a mere few decades, but for more than a century.
Personally I'd be up for one or the other: software patents would be acceptable if you couldn't copyright source code (and actually this would be pretty nice, since it would dump everything into the public domain in a relatively short period), but having both -- being able to copyright a particular implementation while also patenting the general principles -- is atrocious. It's really surprising that we can do as much today with computers as we can; I suspect that's really only because most large companies have such large "MAD" patent portfolios that they can violate others' patents with relative impunity.
I think the traditional way to seal a document using a postmark is to close the envelope, then place the stamp (and the address) on the BACK of the envelope, so that the stamp (and, when mailed, the postmark) are across the flap, effectively sealing it.
Obviously if you put the stamp on the front, you have no way of proving that you didn't just mail an unsealed envelope and stuff the papers in there after the fact. It's only of any use if you put the stamp on the back.
However, this is all a moot point because even if you do the envelope trick right, that's not enough to stop a patent filing later on. You need to PUBLISH the information, not conceal it. Putting it on a blog and letting Archive.org pick it up, or posting it to an appropriate Usenet group / public, high-traffic mailing list, etc. would be better.
The solution is even easier than encryption. Just don't broadcast a unique identifier!
In this case there's no reason for each tire pressure sensor to be broadcasting one. All they need to do is chirp back the pressure inside the tire. That's it. Give them enough power to hit a receiver located in the wheel (which might be 4-6" away in a very large tire, probably a lot closer than that, and it's all inside the steel-belted tire) and call it a day. Unless you are playing Ben Hur, you're not going to get close enough to another car's tires for it to become a problem -- use a high frequency and you're going to get a substantial bit of attenuation via the tire itself, and then you're decreasing as the square of the distance through free space. You're never going to have more than one valve-stem sensor per wheel-mounted receiver, so why bother with it?
If you really do need a weak form of identification, rather than hardcoding a UID, it would be pretty trivial to have each sensor randomly choose a number from a range such that the chance of collisions was low (deriving the randomness from resistor noise or by oversampling whatever analog sensor they use to determine pressure) and reset periodically or each time the car is started. That eliminates the problem of having to coordinate UIDs and prevent duplicates (cf. the cheap Bluetooth transceivers that caused problems because their MAC-ish addresses were all zeros). Every unit can be completely identical.
On further consideration, I can't really imagine why the designers of the TPMS would have given each sensor a UID (especially since it would probably cause confusion when you rotate tires, if the car's computer tracks them)... making me wonder if this is just an elaborate 4/1 hoax.
I think the $1.5Bn number may only be in reference to consulting services rendered to the Federal government, and not hardware/systems sales. It does seem small even for that, though.
The name had to be changed due to a trademark issue.
My understanding is that the lead developer started working on Ethereal while working at one company (as an F/OSS project), and then left for a competitor but continued working on it. Although the codebase was undisturbed, since it was GPL, the first company retained the rights to the 'Ethereal' name.
Slashdot Mirror
Companies that try to do more than one thing rarely do all of them well.
There are exceptions to this, but in general, most companies are highly specialized. They do something, they do it (hopefully) well, and that's how they make money. Everything they do is in furtherance of that goal: what's called in businesspeak a "core competency."
EBay's core competency isn't the same as Skype's. It doesn't really make sense for them to be in the same organization: with such differing strengths, it would be difficult for the corporate leadership to aggressively advance both of them. You'd always have one unit vying for resources and attention with the other. Rather than a cohesive organization, all working together, you'd have some weird joined-at-the-head abomination.
In that situation it makes perfect sense for the two companies to go their separate ways. By divesting itself of Skype, Ebay gets a lot of cash -- cash that it can invest into doing better what it already does well -- and Skype gets its independence.
It's not shortsighted. The acquisition in the first place was (IMO) not very well thought out, but they have a chance to fix their mistake -- albeit at some non-trivial cost -- and cut their losses before they run Skype into the ground and it stops being competitive.
Telephony was nowhere near eBay's core competency, and the match never looked that great. And even if they were hell-bent on getting into VoIP, there were probably other, better companies that they could have acquired, ones that would have provided more value -- companies that specialized in linking IP services to existing phones for instance. (So an eBay merchant could plug in their existing phone no. and have buyers be able to click on a button on an auction and call them, or something like that.) Not to mention, pretty much any of them would have been a lot cheaper than Skype.
The Skype acquisition always looked like something that had been decided on first, based on some sort of 'gut feeling' driven by hype or wishful thinking or a sense that they were 'missing out', and then the reasoning for it was backfilled in later. (As I think a fair number of Web 2.0 acquisitions were.) Turns out, that's not a really good way to run a business.
> Otherwise, all courses would be graded on the basis of a 100% final exam.
That's sort of how law school works. Sure, you get course grades, but the whole point is to prepare you for the Bar exam, and for a future career as a lawyer. The proof is in the result.
Because consumers don't like that very much, Comcast is -- very slowly -- beginning to feel the heat from government. Reforming their Terms of Service to read less like an indenture agreement would be a good first step towards regaining some lost trust. But until they do that (and until they remove the sections that let them change the agreement at will), nothing they say or do has any meaning or value.
Amen. The thing I miss most about my Thinkpad versus the Dell I have now is the TrackPoint.
To me it seems like a no-brainer on a subnotebook, since it doesn't take up any additional space outside of the keyboard. A trackpad just seems like a terrible choice when you're space-constrained. I've never liked them (I liked mid-size trackballs, like Apple used to use on the 1xx series, better), and they only get worse the smaller the tracking surface. Why the Cloudbook designers went with what they did, instead of putting a TrackPoint/nipple in the center of the keyboard, boggles me.
The cheapest subnotes I've seen with a TrackPoint are all over $1k. There's the Asus S200N, but that's selling for something like $1400, and that's assuming you can get your hands on one in the U.S. to begin with; the Dell Latitude D430 also has one, but I think it lists at like ~$1100.
(The Palm Folio was supposed to use a TP with a scroll wheel, and would have been in the $500-600 range, but unfortunately it's DOA.)
Hopefully the UMPC people will see the light eventually.
(I'm a bit rusty on the details, but I've been advised at various times by lawyers that there are situations where a company can be held via contract law to statements made outside the contract itself, if they basically define the relationship between the company and the customer. I doubt Comcast's lawyers are stupid enough to walk into this trap unknowingly, but you never know.)
Although I very much doubt that Comcast is acting in anything approaching good faith here, it's not impossible for them to make the Bill of Rights binding, if they were sufficiently motivated.
What needs to happen is that we, as users, need to make sure that Congress and various state legislatures aren't distracted by any sort of non-binding agreement on Comcast's part. If they want to avoid burdensome regulation, they can come up with a 'Bill of Rights' and then hold themselves to it contractually. But if they don't do that, or if they put it in their contract but then leave in a way of unilaterally amending the contract, it's not worth two squirts of piss.
I've been thinking for a while about purchasing one of the subnotes. Not as a primary computer or even a primary notebook, but as an alternative to the giant and totally impractical company-issue 19" notebook.
The EeePC 900 seems like it may be the leader right now; I was intrigued by the original EeePC but having that tiny screen surrounded by a giant bezel smacked of wasted space. I don't mind using a 7 or 8" screen, but if I am, I want the whole device to be that size. If it's big enough for a 9" screen, I want the screen to actually take up the available space.
The Cloudbook (aka PacBell EasyNote) sounded neat at first, and apparently it's got more metal in it than the Eee and feels more rugged, but the reviews of it have been pretty terrible. The one thing that really turned me off from it, oddly enough, was that its SD Card slot doesn't let the card fit flush. That just smacks of poor design; when a product has something that obviously dumb on it, it makes me wonder what more fundamental flaws it has under the hood. I hope they'll do another revision and try harder. (Ditching their thumbstick for a IBM-style keyboard stick would probably be enough to send me reaching for my wallet.)
The other feature that I haven't seen mentioned much so far, is support for connecting to wide-area data (cell data) networks, rather than just Wifi. Apparently Linux support for most WAN cards is pretty dismal, but Bluetooth (so you could go computer->cellphone->WAN) would be a nice plus. The smaller and lighter a computer gets, the more I think you want persistent connectivity.
But overall it's pretty exciting to be seeing machines for under $500 in a form factor that you recently had to pay upwards of $1k to even get into (even if all you wanted was the small size, and not any of the other stuff that comes on the very high-end subnotes).
The other nice bonus of computers getting cheaper is that it's becoming more and more apparent how much Windows actually costs. Sure, the Windows and Linux versions of the Eee 900 are the same, but if you want Windows, you only get 12GB of storage instead of 20. Even if I hadn't already decided on Linux, that would give me significant pause. (Although I suspect there will be a lot of Linux versions sold that just get wiped and replaced with pirated Windows installs; but hey, at least they're counted as Linux preinstalls.)
Here's the details I've been able to glean so far:
- They were using gasoline. TFA is wrong; the single "liquid petroleum gasoline" car they refer to was actually running on LPG, liquefied propane gas.
- They were using some sort of small gasoline internal-combustion engine; my WAG is that it's maybe a 70cc motorcycle engine? Not clear on this point.
- The competition required them to go 15MPH. Not sure how far they had to go, but it certainly wasn't a gallon of gas!
I'd be really interested in more information on what type of engine they were using, if they made any modifications to it, etc.
(I'm basing this off of my recollection of a Wired magazine article I read a while back.) I think what's generally done is if you run on something other than gasoline, they convert your energy source to gasoline-equivalent using predetermined conversion factors. Those conversion factors try to take into account generating the energy from a base source, and providing it in a form that can be used. It's sort of an average generation/transmission cost.
Obviously there's a lot of room for argument depending on how you fix the conversion factors, but I think even when you consider the "well to wheels" efficiency, they're still very competitive with gasoline engines.
Also, it's not as though gasoline is free from production costs. Just like hydrogen, it takes energy to extract and produce gasoline from raw materials, not to mention transport and distribute it. So you if you want to have a fair competition with efficiency as its goal, you have to apply the same conversions to all types of fuel.
All that taken into account, I don't think it's unfair to give something of an advantage to gasoline-fueled vehicles, if the goal is to develop realistic, easily-implementable efficiency-boosting technologies. This competition seems a bit removed from that (cars are too small, too light, etc., to be practical), but it's silly just to ignore the huge existing investment in liquid-petroleum-fuel distribution infrastructure, like it or not.
The F-14 represents the pinnacle of aviation technology, circa 1972 or so. There's a ~30 year gap there, and that represents 30 years in which the United States made some pretty significant leaps in engineering technology.
So in short, just because a country can build a bomb, doesn't mean they could build an F-14.
And even if they could build an F-14, it doesn't mean it would be much of a match for more recent aircraft, which have a further 20+ years of development behind them. (Although I think you could argue that the pace of development slowed down dramatically after the end of the Cold War, at least in major weapons systems.)
It's not an unresolvable problem by any means, I just don't think it should have to be a problem at all.
I thought the behavior sucked the first time I encountered it, when it was called "Site Finder" and was being run by Verisign. It caused a fairly significant brouhaha, as many here may recall, and they were eventually slapped down by ICANN.
Although it's obviously more evil to do something like that when you're a default root server operator and not running an opt-in service like OpenDNS, I think the practice is crappy and should be considered broken regardless of who's running it. Bad practice is bad practice.
There are enough other public nameservers and alternate roots that I just can't see using OpenDNS, unless you really want the "added" services they offer (adult-site filtering, phishing protection).
The fact that they have large caches is nice, and they can definitely be an improvement over some ISP's default DNS (which really says more negative about the ISP's DNS service than it does good about OpenDNS), I just don't think it's worth the trouble. YMMV, of course.
I don't know much about them, but I think the "Open Root Server Network" might be a possible candidate. It's an alternate root system, independent from but currently mirroring ICANN's, located mostly in Europe. (The sole non-Europe rootserver seems to be run by Paul Vixie, actually.)
I gather from the Wikipedia page their major concern is monopolization of the DNS root by the U.S. Government.
Their site has instructions on switching to their roots, if you run your own DNS server, and a list of publicly-accessible DNS servers that use their roots, if you just want to re-point your workstation or router.
If they're not your style, WP has a list of other alt roots; most of them seem to revolve around the idea of having more or different TLDs than ICANN. The ones I'd probably consider first would be OpenNIC and the Open Root Server Confederation. The latter's website doesn't seem to indicate, at least to a quick reading, their root server addresses or any publicly-accessible DNS servers.
> orthodox MACROeconomics suggests that wages are sticky downwards -- i.e. they don't tend to fall based on an increase of supply...
This seems incredibly doubtful. I could believe it, if you're just talking about gross wages, but that's almost meaningless. Real wages, indexed against inflation, effectively decrease whenever gross wages aren't going up.
And negotiating a salary increase is very difficult in a slack labor market. This is pretty fundamental economic theory, not to mention something that can many people have actually experienced.
This is the situation we are increasingly finding ourselves in: gross wages haven't decreased, but in many sectors real wages have shrunk. If inflation increases, as it seems poised to do, this may get worse. And having a labor surplus will only exacerbate the situation.
The only positive effect I can envision of such a surplus is that it might act as a brake on inflation, by preventing workers from negotiating salary increases as they might in a tight market, thus preventing a positive-feedback loop. However, I'm not sure whether having the surplus will be beneficial in the net or not, especially when you consider the huge costs associated with maintaining a large number of non-productive workers (unemployment, welfare, retraining, etc.).
Also, labor surpluses have a history of causing political and social volatility. It strikes me as odd that we're intentionally creating one in the U.S., since I doubt it's something that most voters would support if asked.
It's an alternate root, not a proxy server. I don't have the hate-on for OpenDNS that the GP does, but it does have several weaknesses as a service which caused me to stop using it.
The biggest problem, and one that the GP alluded to, is that OpenDNS resolves *everything* to a sort of 'parking' page. If you're using OpenDNS and you type in a bogus URL, rather than just not resolving, you'll get a redirect to an OpenDNS page. This is, IMO, misbehavior. However, there's no incentive for OpenDNS to stop, because it's on these pages that they place advertising and pay for themselves.
This behavior is particularly obnoxious when you combine it with an additional level of caching DNS. Let's say you have a DNS server on your LAN (like most home gateway/routers) and you point it to OpenDNS. If you're working with a site that may or may not exist -- say one that you're trying to configure -- OpenDNS will give you the parking page if it can't be found. But your local DNS server will cache the redirect, and it can take a while to purge. (I'm not sure what TTL they're set to, but it's evidently longer than it should be.) The upshot of this is that a site can look 'down' even though it ought to be up, because intermediate DNS servers cache the bogus OpenDNS result, rather than just failing to resolve.
I think it's great that there's an alternate root, and I really like that OpenDNS exists. It's a great concept. I just think their execution deviates from accepted practice and standards, and that's no way to run a DNS server. Too much rides on it.
Although I think it's a fairly big 'hop' to go from understanding interpersonal differences to deciding that one person is objectively better than another, even if you do get to that point, there's still a wide, gaping chasm before you get to the point of forcibly "fixing" people. Action is pretty distinct from having some knowledge or holding a certain opinion; it's a totally different category.
Certainly we need to be cautious and discuss these sorts of issues as we advance our knowledge of biology and the human body, but I think it's also pretty clear where we can draw the line. We don't have to wait for science to get there before deciding some things -- like forced gene therapy -- are just not okay.
However, we should be cautious not to overreact and throw the proverbial baby out with the bathwater. Lots of technologies create possibilities for abuse, and biotechnology isn't really any different.
Interesting. This is the first time I'd heard any hard, realistic-sounding numbers.
I suspect if Microsoft really tries, they will probably be able to rally a little bit with Vista's successor ('Longerhorn'?), but the writing is sort o on the wall for them as a company. They can't possibly go anywhere but down.
There's just no place to go! They owned an emerging market and rode it all the way. MBAs will be studying them until the Earth turns into a cold, dead cinder. But only someone suffering from a great deal of delusions would have ever said that it could go on forever in its current form.
What they do have is a whole lot of cash, so perhaps they can turn themselves into a high-tech version of GE Capital (only, with less suckage). They'll need to learn to stop sucking the life out of everything they buy, but I think that's probably manageable. Hell, if IBM could reinvent itself as a software-and-services firm, Microsoft can definitely pull off becoming a big incubator, if they wanted to go that route.
But that's beside the point: their OS marketshare is headed south and I really can't see a scenario that would recreate the environment that made their 1990s takeover of the market possible again, for them or anyone else.
Somebody once said something to me about the technology sector that I thought was pretty smart. It was "the technology always moves surprisingly fast, and the business always moves surprisingly slowly." Most predictions of the future by technologists or IT people vastly underestimate the amount of time that the rest of the world will take to catch up.
Right now we're seeing the technology move into place, with the desktop OS fading into the background of what makes a particular "device" as a whole, and thus becoming commodified, but I think it'll probably be a long wait for the market to realize that the Age of Windows has had its peak.
Can you please explain how that is so? Reading countless economics text books about the benefits of division of labor have confused me. Try at least a little harder than that.
Imagine a widget factory. The factory takes in raw materials, and produces finished widgets. The widgets are sold on the market for some price that exceeds costs, resulting in profit. The workers are paid a salary, which they can use to buy widgets. With the exception of possibly exhausting whatever raw materials are used to create the widgets, you can repeat this wealth-generating cycle forever. (I.e. it's not some sort of closed system, and it's not zero-sum; you're creating wealth by adding value via the raw-materials-to-finished-products process. There are other processes that create wealth, this is just the most obvious.)
Now, we outsource that factory to Somewhere Else, but continue to import the widgets to satisfy domestic demand, perhaps at a lower price. Now, consumers buy their widgets from Somewhere Else, meaning that wealth flows over there. At the same time, all the people who work at the widget factory are unemployed.
Do you start to see a problem here? If you can't find something else for your former widgetmakers to do, you end up just draining money out of your economy. If you have modern finance at your disposal, you can conveniently spend more wealth than you actually have, issuing debt and importing stuff; at least you can until people stop wanting to buy your debt. This isn't sustainable. Eventually you either literally run out of hard currency (the case if you use gold or something else that can't be created), or people decide to stop buying your debt. And then you have a bunch of angry, unemployed ex-widgetmakers who can't afford to buy widgets anymore. Problem.
Of course, there are cute responses to this. You could argue that this is just the way things are supposed to work -- if the widgetmakers couldn't compete, they deserved to go out of business. Fair enough, and that actually makes a certain amount of sense.
But suppose you have an entire nation of widgetmakers? An entire nation of people who have built themselves a nice lifestyle (oh, and by the way, a huge fucking quantity of nuclear weapons) for themselves, making widgets, and suddenly end up unemployed? What do you expect them to do, calmly and rationally reduce their standard of living so that they can compete better on price? I don't think so; not when they have the ability to go and take a lot of wealth via brute force.
Not a chance; not with the population they have. Maybe in a century, but fifteen years? That's ridiculous. There are millions upon millions of people in China (and India, and quite a few other places) who have grown up and are used to far cheaper standards of living than the average person in the U.S. That translates into dramatically lower labor costs for the foreseeable future, since they're going to be willing to work for less. Someone who remembers life in a mud-and-thatch hut on a rice paddy is probably going to have a markedly different bar for 'success' than someone who grew up in the U.S.'s heyday and expects to be able to do better than that.Well, if you want to talk about 15 years down the road you might as well mention that in 15 years all the demand from our outsourcing will make the Chinese as well off as us, forcing them to charge as much, canceling out any benefit of outsourcing there.
That's a great thought but it's a little lacking in substance. What do you propose the U.S. ought to specialize in? I'm quite honestly interested, and I've asked this question over and over to a lot of fairly intelligent people and have yet to get a satisfactory answer back. I'm not sure there is one. Do we try to go the Neal Stephenson route? Music, movies, microcode, and pizza? Other parts of the world are chipping into 'software' already, and there's no reason to think that we have some kind of automatic, natural, competitive advantage in any of those.You're a little capitalist, and you don't even realize it. Want all the jobs to stay in our country? That's greed; the same thing driving those shareholders to make more money. Unfortunately, whining doesn't get much done, so we'll all have to work really hard and offer some kind of advantage to keep the jobs. It's called "competing".
About the only thing we do have here in the U.S., at least at the moment, is a hell of a consumer market. Until we figure out exactly how we're going to keep ourselves going, I don't think it's necessarily illogical to want to carefully manage access to the one thing of value we have left. I'm not proposing or advocating for complete isolationism, just a careful analysis of exactly who we're allowing access, and to which markets, and what the effects are.
More bluntly, I don't see any reason why the U.S. ought to open any market to foreign competition unless there's a clear indication that opening it results in a net benefit to the United States. Now, it may be that fully-open markets are the best (or least-worst) policy for Americans in general, but I haven't seen any of the politicians pushing for open markets really going out of their way to demonstrate this. And from where I'm sitting, it looks a lot like we're just letting ourselves go bankrupt on imports without much of a thought towards the long-term sustainability of this situation.
Even if by restricting imports it increased the cost of non-essential goods to consumers, but in doing so bought us a few more years or decades of solvency in which to work on our comparative advantage (or for the Chinese and other developing markets to bring their labor force's standards of living, and thus costs, closer to par), I can't see why that would necessarily be bad.
National governments have a mandate to serve the best interests of the people they represent. If free trade and open borders are demonstrably the best path, I'd be more supportive, but right now they look suspiciously like a path that leads off a cliff.
> But I remember all the old Dell commercials - the main thing they had going for them was customization.
I think the point is that those were the old Dell commercials. If you look at ones today, they're all about price. Features and price, admittedly, but price is the biggest thing.
This is a reflection of the market for PCs. When they represented a substantial capital investment, you wanted to tailor them to your particular needs, and avoid paying for anything you didn't absolutely need. That made customization and U.S.-based assembly locations worthwhile. Now, people don't want that as much. The PC, as a unit, has become increasingly commoditized. I bet a lot of buyers today don't even look at specs; they just buy "a computer" and make a lot of assumptions about what they'll be able to do with it. (Assumptions that are actually pretty safe if you don't plan on doing much beyond typical consumerish tasks with it.)
As a result, the goal is no longer "build me a PC to my exact specifications," it's "build me as much PC as possible for $500". Or $300, or $250. I suspect before too long it'll be $99.
That doesn't favor having a lot of assembly points close to consumers; it favors doing all your assembly in a quasi-slave-labor camp somewhere, to better keep costs down, and then shipping tons and tons of identical boxes in bulk to wherever the consumers are. 'Who cares if it's not exactly what you want? It's $500 and it's more power/features/speed than you'll probably need, so just buy it,' is the message.
It's easy to blame Dell here, but it's buyers of technology that are driving it. Not enough people want essentially bespoke computers (or the ones that do aren't buying them from Dell), and Dell is going to eliminate the facilities that provide that service.
Now if it was $10/phone unlimited, then I could see how you may say it is competitive. It's competitive with other carriers.
Basically the marginal price is more connected to what people will pay for a service, and less what it actually costs to provide that service.
There isn't enough competition in the wireless market to drive the price down to the marginal cost of delivery, unfortunately.
You're ignoring that software is already protected via copyright. So particular implementations of a new idea are automatically protected -- and not for a mere few decades, but for more than a century.
Personally I'd be up for one or the other: software patents would be acceptable if you couldn't copyright source code (and actually this would be pretty nice, since it would dump everything into the public domain in a relatively short period), but having both -- being able to copyright a particular implementation while also patenting the general principles -- is atrocious. It's really surprising that we can do as much today with computers as we can; I suspect that's really only because most large companies have such large "MAD" patent portfolios that they can violate others' patents with relative impunity.
I think the traditional way to seal a document using a postmark is to close the envelope, then place the stamp (and the address) on the BACK of the envelope, so that the stamp (and, when mailed, the postmark) are across the flap, effectively sealing it.
Obviously if you put the stamp on the front, you have no way of proving that you didn't just mail an unsealed envelope and stuff the papers in there after the fact. It's only of any use if you put the stamp on the back.
However, this is all a moot point because even if you do the envelope trick right, that's not enough to stop a patent filing later on. You need to PUBLISH the information, not conceal it. Putting it on a blog and letting Archive.org pick it up, or posting it to an appropriate Usenet group / public, high-traffic mailing list, etc. would be better.
The solution is even easier than encryption. Just don't broadcast a unique identifier!
... making me wonder if this is just an elaborate 4/1 hoax.
In this case there's no reason for each tire pressure sensor to be broadcasting one. All they need to do is chirp back the pressure inside the tire. That's it. Give them enough power to hit a receiver located in the wheel (which might be 4-6" away in a very large tire, probably a lot closer than that, and it's all inside the steel-belted tire) and call it a day. Unless you are playing Ben Hur, you're not going to get close enough to another car's tires for it to become a problem -- use a high frequency and you're going to get a substantial bit of attenuation via the tire itself, and then you're decreasing as the square of the distance through free space. You're never going to have more than one valve-stem sensor per wheel-mounted receiver, so why bother with it?
If you really do need a weak form of identification, rather than hardcoding a UID, it would be pretty trivial to have each sensor randomly choose a number from a range such that the chance of collisions was low (deriving the randomness from resistor noise or by oversampling whatever analog sensor they use to determine pressure) and reset periodically or each time the car is started. That eliminates the problem of having to coordinate UIDs and prevent duplicates (cf. the cheap Bluetooth transceivers that caused problems because their MAC-ish addresses were all zeros). Every unit can be completely identical.
On further consideration, I can't really imagine why the designers of the TPMS would have given each sensor a UID (especially since it would probably cause confusion when you rotate tires, if the car's computer tracks them)
I think the $1.5Bn number may only be in reference to consulting services rendered to the Federal government, and not hardware/systems sales. It does seem small even for that, though.
The name had to be changed due to a trademark issue.
My understanding is that the lead developer started working on Ethereal while working at one company (as an F/OSS project), and then left for a competitor but continued working on it. Although the codebase was undisturbed, since it was GPL, the first company retained the rights to the 'Ethereal' name.
There was a Slashdot FPP on it not that long ago.