This question is not nearly as theoretical as the question suggests: there are many countries where various forms of metered or tiered access are the norm. You just have to look at what these countries offer (and how consumers react) to get an idea of what works and what doesn't.
Here's an example: Videotron cable internet (Montreal, Canada)*. They have packages that run from $30/month to $80/month, depending what you want. They all have usage limits (2 GB/month to 100 GB/month), and charge a fee per additional GB beyond this basic usage.**
Does it "work"? Of course. Customers buy the package they want. If they are routinely going over their monthly limit, they either cut back on usage or upgrade their package. Yes, it is slightly more complicated for the customer than just having a single "unlimited!" package, but then again it's also more honest. In fact the unlimited packages have hidden terms and limits, which makes them more complicated... or at least more annoying.
I'm a heavy internet user (as most Slashdotters probably are). I don't mind paying a premium to get the speeds and usage limits I need: as long as that service level is actually delivered! This isn't rocket science: just provide a variety of packages and let the customers pick. Importantly, price the packages so that you won't go out of business if a sizeable percent of your customers actually use the service you sold them.
[*] Note that I was a Videotron customer when I lived in Montreal. I'm not endorsing their service; merely using them as an example.
[**] Note also that if you really want unlimited usage, you can upgrade to business class service. Again, you pay a premium if you want that level of service, which is fine.
One model for secure communication uses quantum cryptography to exchange a key that is actually pairs of entangled photons. In rough terms, you have a source that generates entangled photon pairs, and you keep one and send the other whoever you're trying to communicate with. You use this stream of photons to generate cryptographic keys, with the added quantum bonus that you can detect whether someone else has intercepted the key exchange (because, if so, the entanglement will be broken so the correlations between the two sets of photons will be "wrong").
For this to work, you need a way to reliably generate single photons or single photon pairs, and a way to transmit these photons without them losing their entanglement. This paper helps address the first part, by generating single photons on demand. Better yet, they generate 'indistinguishable' photons, which is necessary because the objective is to interfere two photons with each other to generate entangled pairs.
a pool of trained searchers that could be called upon for difficult queries (paid at maybe a fourth the rate of salaried employees). I know that I've had searches for work that took most of a day just to find that one formula I needed from 30 years worth of journal papers. (emphasis added) That's the problem right there. The really difficult searches require very specific domain knowledge. I've also spent days searching for some specific bit of information. But I doubt that anyone else would have recognized the bit of information when they came across it, unless they also had the training and domain knowledge that I have. Also, searching is itself a learning process:
1. While searching, you pick up tangential tid-bits of knowledge that are related to what you're trying to find (even if not a perfect match).
2. While searching, you begin to gain an appreciation for how widespread vs. novel (or established vs. fringe) the data you're trying to find is.
3. While searching, you learn to modify your search criteria (e.g. based on 1 and 2), which helps you iteratively discover what you're looking for.
So these trained searchers would need to be highly trained in the subject area, in which case they wouldn't really tolerate being low-paid. Put otherwise: it would be just as expensive to pay you to search for yourself, rather than to outsource the search.
Okay, maybe I'm being a bit extreme. I certainly agree that there is a wide middle-ground of searches that are sufficiently complex that they are not on the Google top-10, but sufficiently simple that they can be explained to someone else. A trained searcher could, at least, scour the literature for things that seem to match what you want (and reject obviously bad data, spam, etc.). This would still require the search requester to check through the top results, but would save time if the trained searcher knew of many search tricks that others do not (e.g. reference librarians are often able to find data much faster than anyone else).
I guess I'm partially convinced. Trained searchers could be useful in some cases, but in other cases it's more useful for the intended user of the data to be part of the discovery process, since they will learn more and find the best answer faster.
In the short term, Apple is a powerful force for eroding Microsoft's monopoly. But in the long term, Apple is just as prone to lock-in as Microsoft. In fact, Apple has been hugely successful at creating lock-in and selling incremental upgrades. Moreover, if trusted computing scares you, then the idea of a single vendor being in control of both the hardware and OS design is very scary indeed: Apple can make sure that they only ship devices where the trusted-computing enforced lock-in is already firmly in place (and not removable).
In the long term, the only salvation from such initiatives is software that is truly open and Free. In cases where the OS actually makes the capabilities of the trusted computing chip available to the end user, then everything changes: we can use the chip as a guard against viruses and malware by authorizing trusted keys (e.g. Debian's or Red Hat's). As long as the end user is in control, they can bypass the authentication when required (e.g. to run code they wrote themselves).
Obviously this all breaks down if the hardware manufacturers only ship computers with pre-loaded certifications for big-name vendors (Microsoft) and no way for the user to add new certs. In such a world, the end-user can't be said to "own" the hardware they buy in any meaningful way. Luckily I think that dystopia isn't realistic: running Linux servers is important (and profitable!) so there will always be a market for commodity gear that we can use to run Free operating systems.
Another 'danger' (from Microsoft's perspective) would be "the WINE effect", a.k.a. reverse engineering. If they separate their OS into well-defined modules, then others can create replacements for those modules. Even if the interfaces are secret and there is no public documentation (which is likely to be the case), the partitioning into modules will mean that at some level there is a well-defined API (even if it isn't publicly disclosed). So people can reverse-engineer that API and write their own drop-in replacement modules.
This would be great for lots of people: other companies could write competing modules to replace Windows functionality (why pay for Microsoft's system-wide search module when Google's is so much better?). Also, free and open-source modules will probably be created for many of those features.
Of course, it may be that Microsoft intends to create a complicated system of internal certificates and code signing so that only MS-approved modules can use these hidden APIs. It seems like that would add a considerable performance penalty, but then again I guess that's not too different from the decisions they made in designing Vista.
This is a perfect example of why Windows looks less appealing over time, and why I started shifting everything I do to be inherently platform agnostic (e.g. if I'm going to learn a new piece of software, I preferentially find programs that work on any OS).
Basically, they are redesigning Windows not so much for engineering reasons, but for marketing reasons. Having a modular OS could be a very good idea, if done properly. However if this is just artificial fragmentation so that they can nickel-and-dime their customers, then this means the software is going to get worse.
Why? Because it's hard enough to optimize software just to do its job properly. You can't optimize for every constraint simultaneously, so if you add requirements like "separated into marketable modules" or "resistant to user tampering" then the coders will necessarily have to compromise on other optimizations, like "speed of operation" or "flexibility" or "reliability" or whatever.
Software is becoming more and more of a commodity... which means that open-source software is rapidly catching up (in those fields where it wasn't already the leader), and also that companies like Microsoft who are still caught in the "must sell proprietary code in boxes" mindset have to add more and more user-hostile features.
Just to speculate a bit here... It seems to me that people naturally, inherently try to ascribe 'personality' and 'morality' to entities like corporations. I guess it's an extension of the natural human desire to assess people's character, and use this assessment to determine trust relationships. In a normal human-to-human interaction, you can determine a person's character (whether they will treat you well or not) and use this to decide whether to trust them. It works because other people tend to be relatively consistent and constant over time.
The problem is that people then unconsciously port this methodology into the domain of assessing a corporation. In this case it doesn't work: you can have a positive experience with one part of the company, but that actually says little about how other parts of the company will treat you (e.g. a nice salesman versus a rude phone support person a week later). This confusion is very much intentional on the part of the company: the marketing departments are very good at creating the image of friendlieness, or trustworthyness, or hipness, or whatever... but this bears no correlation to the actual engineering or sales departments.
It's been said before that if corporations are persons then they are surely insane persons. Indeed. The problem is that corporations 'behave' in inconsistent ways. It's like they have mental disorders (bipolar? multiple personalities?), and hence violate the normal rules we would like to use for consistency and trust.
All of that to say that we should be very careful about assigning personality to corporations. A statistical analysis of a company is meaningful (e.g. "I use this company because 80% of customers who call the support line get a satisfactory solution within 5 minutes"), but we should not fall into the (natural) trap of treating the company as a single personality (e.g. "I use this company because it's always been nice to me").
Just in case you're serious (or someone else suffers from the misconception embodied in your post):
The issue here is not with users voluntarily using Google services (search, gmail, etc.). Rather it is with companies who want to outsource their data needs to Google. In addition to the visible public products that Google has, it also offers corporate solutions: for instance if a company wants to outsource their email system, or have Google run search and collaborative software for use inside the company.
Google is trying hard to make these new kinds of products work. But unfortunately U.S. laws mean that any data that ends up on Google servers can be snooped by U.S. authorities. Many companies don't like the idea that the U.S. government will have such broad access to their data. In many countries where strong privacy laws exist (Canada, U.K., etc.), allowing the data to be managed by a U.S. company would then actually be illegal--since the company couldn't guarantee integrity or privacy of the data.
The end result of this is that Google is at disadvantage in the global marketplace because of the over-reaching U.S. laws. Google isn't the only one, of course: I'm sure U.S. companies have been losing lots of contracts because international businesses are wary of storing or moving data through U.S. systems since it is now well-known that such systems are not immune to U.S. government monitoring or interference.
The article doesn't make it clear whether using optical communications is intended to reduce latency or increase bandwidth.
With respect to latency: the electrical signals travel at ~30% the speed of light, whereas the optical signals travel at ~70% the speed of light (it depends on refractive index, etc.). Over the distances we're talking about (as you said, mm to dm), that's only fractions of a nanoseconds delay savings. This is on the order of a modern computer's switching time. All this complexity to get rid of a one or two processor cycles of latency?
I suspect instead they are looking to increase bandwidth. An optical fiber can carry very high data rates. Moreover a single physical fiber can carry multiple simultaneous channels (e.g. different wavelengths of light). So the intention may instead be to create high-bandwidth links between various processors. Using on-chip lasers can make the entire assembly smaller and faster than the equivalent for electrical wires.
Really what they want, I think, is to implement the same kind of high-speed optical switching we use for transcontinental fiber-optics into a single computer or computer cluster. If you can put all the switching and multiplexing components directly onto the silicon chips, then you can have the best of both worlds: well-established silicon microchips that interface directly into well-understood high-speed optical switching systems.
To use the beloved transportation analogy: it's like moving your cargo off of trucks and onto a high-speed train. Yes it takes time to move cargo, but it's worth it if the time savings of the high-speed train are big enough (for long enough distances, the savings can be significant).
In this case, there may be a delay associated with signal processing, but if the optical transmission is sufficiently faster than an equivalent electrical one, then it's worth it. Considering that electrical signals themselves need to undergo various kinds of switching and processing anyway (data written or read from a bus), I don't know that converting to laser signals will add much of a delay.
It'd be nice to actually have a downloadable video that has menus and chapters and the other niceties that we've grown used to. Including niceties like unskippable FBI warnings and adverts...
I agree that having additional functionality (soundtracks, subtitles, chapter icons, menu system) grouped with video files can be great... however a raw video file has the advantage of being easier to play on a myriad of devices and being under the user's control.
I know nothing about HDi, so I don't know to what extent it locks out the user from accessing the internal data directly... but I really hate data containers that companies use to force user-hostile features (like unskippable content), and so I'm wary in this case.
What did they expect? They probably expected to pay the list price for the quoted product.
The company is taking a loss on each box at $50. That's the company's problem, of course. They are of course free to charge more or less for the devices whenever they want.
So Meraki then does all it can do at that point, force the HW to only run the special software and try to get back into the market. Ah... so I see you missed the part where Meraki pushed firmware upgrades to existing units? They basically forced new software onto older units which lock them out. So, in effect, they sold a device with certain promises (namely, "open!") at a certain price, and then afterwards log into the devices and load new software to prevent the owners of the hardware from exercising the rights that were granted to them under the original contract terms. As far as I know, logging-into someone else's hardware (and then changing the software so that the hardware is now under your control) without their permission is illegal.
The hackers (especially those who put some kind of trust in "openness") are the ones who ruined the municipal network for everyone. They showed a clear lack of political savvy and it ended up turning what could have been a boon for both the city and Meraki into a political morass which ends up with no one at all happy. I disagree. If the company was indeed selling the units at a loss, then that is their own stupidity. Customers taking advantage of what you offer ("open, hackable, access point for $50!") is their legal right and frankly is sensible. I disagree that giving into corporate demands at every turn is "political savvy". The company screwed them (and possibly broke the law), so they are warning others not to deal with that company, and it seems like they are going to try to find other hardware suppliers in the future.
Part of the problem is that the company (Meraki) pushed firmware upgrades to all the units, including older boxes purchased before their revised licensing model. The new firmware locks down the units, making it impossible to hack them and impossible to load custom firmware and bypass the new locks.
That's the really sleezy part--changing your licensing terms for new sales is annoying for loyal customers, but obviously can't apply retroactively to goods you've already sold. But this company is doing just that--trying to retroactively impose their new licensing and payment model onto units that were already sold under an open, permissive terms.
So even though they still have the free code, they are now blocked from loading the code onto their own purchased hardware. It's probably not impossible--a talented hacker can maybe bypass the firmware and load custom code again... but of course they shouldn't have to. It seems to me that Meraki has more or less broken into customer devices without permission and made unrequested changes--rather illegal as far as I know.
This technology is essentially what is used in atomic force microscopes, and was being investigated by IBM for data storage under the name "Millipede". It basically involves a huge array of cantilevers that have very sharp tips on them (typical tip size is 20 nm but smaller is possible). The tips are used to read and write dots on the surface.
So yes, this system has moving parts. The tips have to scan across the surface, and the cantilevers are basically springs that bend up and down as the tips move over the surface. This definitely has some wear issues to consider, but it's nothing like the large-scale and high-speed movements of a hard drive (where a >2" disk is rotating at >7,000 rpm). Instead, the tips are moving laterally by micrometers at most (the huge array is what allows a large surface to be probed), and the cantilevers are springing up-and-down by only nanometers. The movement in an AFM is controlled using piezoelectric deformation of quartz actuators. This small-scale movement is very robust and reproducible. Quartz oscillators can vibrate/move thousands of times a second continuously for years without much problem (think of oscillators used for clocks, etc.). Moreover this technology has been used in commercial AFMs for years, so it's well-understood.
The thousands of tips are probably all actuated together by a single piezo-motor. They move in unison which would actually allow for high-speed reading/writing (since thousands of bits are read/written at once). You're right that each tip is in principle a point of failure. However, with the right error-correction algorithm, the device could be built so that even if a few tips break, no data is lost.
I agree that the access time isn't going to be as fast as modern RAM, but it could very well be faster than modern hard-drives. I think this is intended as permanent storage, not volatile memory.
Everything except KDE packages will have long-term support.
After all, you can install Ubuntu 8.04 LTS, and then install the package "kubuntu-desktop" to add all the required KDE packages. Most of the system (kernel, servers, most apps, GNOME, etc.) will be covered by LTS and receive patches for years. The KDE components will only receive patches for 18 months (of course after that you can just upgrade to the next supported version of Ubuntu or Kubuntu anyway).
Release available through CDs for groups who need it (ie. LoCo teams, conference teams, etc.) as well as downloading
So, basically, the "fully supported" version of Kubuntu 8.04 will use KDE 3.5. You will be able to purchase commercial support from Canonical if you like, but in any case can always download and use it for free. Or, you can use the more experimental Kubuntu 8.04 Remix, which uses KDE 4.0. In this case you will receive the usual community updates and community support.
Both versions are free, but if you use the more stable version that includes KDE 3.5, you have the option of paying Canonical for commercial support, just like every other official Ubuntu and Kubuntu version. You don't have to pay Canonical to get updates or unlock features or anything... but for businesses who want support contracts the option is there.
And, as I said before, Canonical is opting not to consider Kubuntu 8.04 as an "LTS" release... which means that they will officially provide updates to it for "only" 18 months.
I saw on the Kubuntu pages the other day they are forking into 2 versions, one free ( 'with limited functionality') and a commercial version Can you provide a link? Seriously, I haven't heard anything of the sort, and can't find any corroboration on the Kubuntu site.
Perhaps you're referring to the fact that although Ubuntu 8.04 will be considered "Long-Term Support" (LTS), the corresponding Kubuntu 8.04 will not be LTS (it will still be supported, just not for as long). The reason for this decision being that KDE 4.0 is still "too fresh" for Canonical to guarantee that it will be stable-enough (and unchanging-enough) to warrant the LTS label. However Kubuntu will still be available, will still receive consistent patches and updates, and will still be Free and free.
Or perhaps you're referring to the effort to get Linspire's "Click 'n' Run" to work on Ubuntu, which would allow users to install commercial/proprietary software from repositories?
In any case, I think you're mis-remembering what you've read. As far as I know, Kubuntu and Ubuntu will remain Free and free.
However I'll note that the double standard partly arises from a "Windows vs. Linux" myth. That is, Windows is supposedly "compatible with everything" and there are "drivers for every device." According to the myth, Windows isn't supposed to have those kinds of problems; only Linux has trouble with "strange hardware."
For those of us who know that it's a myth (and that both operating systems support a plethora of devices, though obviously not every single one), it's at least interesting to see a concrete example. Windows has driver problems too. In both Windows and Linux, non-existent or buggy drivers can ruin the user experience. And in both cases, if a user loads potentially unsafe software, they must accept the consequences.
There was a time when the Internet was more like a novelty or hobby project. Those of us using it were on the fringe, and nothing that we did on the 'net was vital.
That is no longer the case. The Internet has grown to become a vital infrastructure. Just about every business relies on the Internet to get their work done. It is an indispensable tool for students and academics. It has risen nearly to the status of roads or electrical power in terms of being depended upon by billions of people.
What's my point? My point is that with respect to most utilities (roads, water, electricity, phone) we wouldn't tolerate much interruption in service... and we certainly wouldn't accept companies squabbling as a decent excuse for degrading the infrastructure. Can you imagine driving to work one day and finding roads blocked because of a contract dispute?
I'm not sure what the answer is. Turning the Internet into a government utility has its own problems. Similarly, laws which require certain norms for the utility may be over-reaching or impotent. But, ultimately, we need to push for this critical infrastructure to no longer be treated as a best-effort hobby/entertainment service. We need companies (and possibly legislators?) to acknowledge that the Internet is critical, and that this means that uptime/bandwidth/QoS must be maintained at a high-level.
I think the difference is that oversight/transparency is required in cases where someone wields power. We don't need cameras watching government employees because they work for us, but because they have power over us. No one is suggesting that we watch the employees while they are driving home from work, eating lunch, or even doing routine paperwork. However, filming government agents as they wield tremendous power (cops on duty, meetings between government officials and lobbyists, etc.) is useful to the extent that it can help curtail abuses of power.
Similarly, an employer who wants to monitor all employees with cameras at all times is over-stepping their bounds and infringing on basic privacy. However I think most people would agree that there are times when an employer can justifiably record employee actions (with their knowledge, of course). For instance if an employee is assessing millions of dollars worth of diamonds, a record of their actions seems reasonable. One should also note that casino employees are recorded for similar reasons.
Finally, it's worth noting that when properly implemented, such systems serve to protect both the employer and employee. Taking the diamond assessing example again, the cameras not only help the employer employees who are stealing: they also allow an employee to exonerate themselves by using the footage ("they were all accounted for when I left the room").
To summarize: it's not a question of mere "employment," but rather a question of "oversight when people wield power."
remember that before these licences came along, individuals would often release software under their own (often poorly worded) licences... When I see something that's licenced under "GPL" or "CC", I know exactly what I'm getting, and don't have to worry if I can or can't do something Furthermore, it's worth noting that when it comes to proprietary software, the licensing landscape is very confused and inhomogeneous. Each software package has its own custom EULA. Though they often share similar features (e.g. "no liability"), there are often all kinds of ridiculous clauses buried among the boilerplate legalese. Volume license agreements are no better: you have to be very careful when selecting them (are we talking per-user? per-installation? per-processor? per-active-instance?).
As you point out, at least with open-source licenses, there are only a handful of major ones that cover the vast majority of software. Once you know about them, you can very quickly know how much control you'll have over the code, and can confidently download/install/use/modify as required.
There is no proprietary equivalent to this kind of well-organized and relatively homogeneous licensing landscape. (Of course not! Having "named" proprietary licenses would make it too easy for a customer to compare different product licenses and select the less onerous ones.)
However no one is forcing dell to use the stickers Do you know that for a fact, or are you just assuming?
The reason I ask is because it's possible that Dell's contracts with Microsoft did, in fact, obligate them to promote Vista by the stickers on computers. For instance their bulk discounts are tied to various deals, such as having "Dell recommends Windows Vista" on their website.
Also worth noting is that many OEMs were shipping systems with "Vista Ready" stickers long before Vista was finalized. They had no way of knowing how well Vista would ultimately run on the machines, other than what Microsoft was telling them. Still, the OEMs share the blame to the extent that it was irresponsible of them to trust Microsoft and put stickers on systems without being sure that their claims were correct.
Actually I would argue that there is plenty of worthwhile content on YouTube (and I'm not talking about unauthorized uploads of TV shows). The thing is that YouTube is like the Internet at large: there is lots of crap so if you just randomly poke around you will of course see a predominance of crap.
If you browse YouTube a bit, and subscribe to the channels that are actually worthwhile, you will quickly build up a feed of interesting stuff with new videos every day. You can use featured videos to get some ideas of new channels to consider. On the other hand, using "most viewed" and "currently watching" to find good stuff is a waste of time. As a random example of something "worthwhile" (in my opinion), consider Wallstrip--a show that does profiles on companies and stock trends, and is infused with sarcasm and wit. There are also channels that discuss science, that do decent original comedy, there is a national geographic channel, etc.
Frankly I think YouTube is dropping the ball a bit by not providing a more useful method of finding the best content. An Amazon-like "people who subscribe/rate like you also like..." would help alot. Just as Slashdot uses various tricks (moderation, friends/foes, etc.) to bring attention to the quality material, YouTube should work harder to bring the good material to the top. The current star-ratings, comment-ratings, and ranking-by-viewing are not working very well. Frankly I don't care about the ratings of YouTube at large; I care about the ratings of a finite subset of like-minded users.
although [the XO] has a Linux-based OS, it doesn't have a regular kind of filesystem. It does have a regular filesystem. The sugar UI organizes things based on activities (a.k.a. programs) and has a journal (a.k.a. search system) that shows you all your documents (a.k.a. files). Despite this abstraction, a normal filesystem hides beneath.
'Hides' is probably the wrong word. One of the activities is a terminal, with which you can browse the conventional Linux filesystem normally. You can SSH into the XO, and use terminal commands to install new software. You can even install a new desktop environment (e.g. xfce) to replace sugar if you prefer. It's a low-power machine, but it's running a full-featured Linux distro.
Lately I've been entertaining the idea of moving to somewhere in the developing world where all the kids have XOs, and teaching them to code. That sounds like a fantastic (and altruistic) thing to do. If you're used to coding in Linux, and using Python in particular, you'll find coding on the XO to be a fun. Personally I find the built-in keyboard hard to use, so I usually connect a USB keyboard and mouse if I'm working on it for an extended period.
Since the original poster seems to prefer permissive licensing, he should also check out the GNOME Human Interface Guidelines 2.0. It's an extensive set of best-practices and guidelines, licensed under the GFDL. Thus he can tailor the guidelines to his needs and redistribute them without worrying about copyright issues (another poster suggested setting-up a wiki for his users, which could also work).
The KDE Usability Guide also has some good material, although at this time it looks much less mature than the GNOME docs.
Slashdot Mirror
This question is not nearly as theoretical as the question suggests: there are many countries where various forms of metered or tiered access are the norm. You just have to look at what these countries offer (and how consumers react) to get an idea of what works and what doesn't.
Here's an example: Videotron cable internet (Montreal, Canada)*. They have packages that run from $30/month to $80/month, depending what you want. They all have usage limits (2 GB/month to 100 GB/month), and charge a fee per additional GB beyond this basic usage.**
Does it "work"? Of course. Customers buy the package they want. If they are routinely going over their monthly limit, they either cut back on usage or upgrade their package. Yes, it is slightly more complicated for the customer than just having a single "unlimited!" package, but then again it's also more honest. In fact the unlimited packages have hidden terms and limits, which makes them more complicated... or at least more annoying.
I'm a heavy internet user (as most Slashdotters probably are). I don't mind paying a premium to get the speeds and usage limits I need: as long as that service level is actually delivered! This isn't rocket science: just provide a variety of packages and let the customers pick. Importantly, price the packages so that you won't go out of business if a sizeable percent of your customers actually use the service you sold them.
[*] Note that I was a Videotron customer when I lived in Montreal. I'm not endorsing their service; merely using them as an example.
[**] Note also that if you really want unlimited usage, you can upgrade to business class service. Again, you pay a premium if you want that level of service, which is fine.
One model for secure communication uses quantum cryptography to exchange a key that is actually pairs of entangled photons. In rough terms, you have a source that generates entangled photon pairs, and you keep one and send the other whoever you're trying to communicate with. You use this stream of photons to generate cryptographic keys, with the added quantum bonus that you can detect whether someone else has intercepted the key exchange (because, if so, the entanglement will be broken so the correlations between the two sets of photons will be "wrong").
For this to work, you need a way to reliably generate single photons or single photon pairs, and a way to transmit these photons without them losing their entanglement. This paper helps address the first part, by generating single photons on demand. Better yet, they generate 'indistinguishable' photons, which is necessary because the objective is to interfere two photons with each other to generate entangled pairs.
1. While searching, you pick up tangential tid-bits of knowledge that are related to what you're trying to find (even if not a perfect match).
2. While searching, you begin to gain an appreciation for how widespread vs. novel (or established vs. fringe) the data you're trying to find is.
3. While searching, you learn to modify your search criteria (e.g. based on 1 and 2), which helps you iteratively discover what you're looking for.
So these trained searchers would need to be highly trained in the subject area, in which case they wouldn't really tolerate being low-paid. Put otherwise: it would be just as expensive to pay you to search for yourself, rather than to outsource the search.
Okay, maybe I'm being a bit extreme. I certainly agree that there is a wide middle-ground of searches that are sufficiently complex that they are not on the Google top-10, but sufficiently simple that they can be explained to someone else. A trained searcher could, at least, scour the literature for things that seem to match what you want (and reject obviously bad data, spam, etc.). This would still require the search requester to check through the top results, but would save time if the trained searcher knew of many search tricks that others do not (e.g. reference librarians are often able to find data much faster than anyone else).
I guess I'm partially convinced. Trained searchers could be useful in some cases, but in other cases it's more useful for the intended user of the data to be part of the discovery process, since they will learn more and find the best answer faster.
In the short term, Apple is a powerful force for eroding Microsoft's monopoly. But in the long term, Apple is just as prone to lock-in as Microsoft. In fact, Apple has been hugely successful at creating lock-in and selling incremental upgrades. Moreover, if trusted computing scares you, then the idea of a single vendor being in control of both the hardware and OS design is very scary indeed: Apple can make sure that they only ship devices where the trusted-computing enforced lock-in is already firmly in place (and not removable).
In the long term, the only salvation from such initiatives is software that is truly open and Free. In cases where the OS actually makes the capabilities of the trusted computing chip available to the end user, then everything changes: we can use the chip as a guard against viruses and malware by authorizing trusted keys (e.g. Debian's or Red Hat's). As long as the end user is in control, they can bypass the authentication when required (e.g. to run code they wrote themselves).
Obviously this all breaks down if the hardware manufacturers only ship computers with pre-loaded certifications for big-name vendors (Microsoft) and no way for the user to add new certs. In such a world, the end-user can't be said to "own" the hardware they buy in any meaningful way. Luckily I think that dystopia isn't realistic: running Linux servers is important (and profitable!) so there will always be a market for commodity gear that we can use to run Free operating systems.
Another 'danger' (from Microsoft's perspective) would be "the WINE effect", a.k.a. reverse engineering. If they separate their OS into well-defined modules, then others can create replacements for those modules. Even if the interfaces are secret and there is no public documentation (which is likely to be the case), the partitioning into modules will mean that at some level there is a well-defined API (even if it isn't publicly disclosed). So people can reverse-engineer that API and write their own drop-in replacement modules.
This would be great for lots of people: other companies could write competing modules to replace Windows functionality (why pay for Microsoft's system-wide search module when Google's is so much better?). Also, free and open-source modules will probably be created for many of those features.
Of course, it may be that Microsoft intends to create a complicated system of internal certificates and code signing so that only MS-approved modules can use these hidden APIs. It seems like that would add a considerable performance penalty, but then again I guess that's not too different from the decisions they made in designing Vista.
This is a perfect example of why Windows looks less appealing over time, and why I started shifting everything I do to be inherently platform agnostic (e.g. if I'm going to learn a new piece of software, I preferentially find programs that work on any OS).
Basically, they are redesigning Windows not so much for engineering reasons, but for marketing reasons. Having a modular OS could be a very good idea, if done properly. However if this is just artificial fragmentation so that they can nickel-and-dime their customers, then this means the software is going to get worse.
Why? Because it's hard enough to optimize software just to do its job properly. You can't optimize for every constraint simultaneously, so if you add requirements like "separated into marketable modules" or "resistant to user tampering" then the coders will necessarily have to compromise on other optimizations, like "speed of operation" or "flexibility" or "reliability" or whatever.
Software is becoming more and more of a commodity... which means that open-source software is rapidly catching up (in those fields where it wasn't already the leader), and also that companies like Microsoft who are still caught in the "must sell proprietary code in boxes" mindset have to add more and more user-hostile features.
Just to speculate a bit here... It seems to me that people naturally, inherently try to ascribe 'personality' and 'morality' to entities like corporations. I guess it's an extension of the natural human desire to assess people's character, and use this assessment to determine trust relationships. In a normal human-to-human interaction, you can determine a person's character (whether they will treat you well or not) and use this to decide whether to trust them. It works because other people tend to be relatively consistent and constant over time.
The problem is that people then unconsciously port this methodology into the domain of assessing a corporation. In this case it doesn't work: you can have a positive experience with one part of the company, but that actually says little about how other parts of the company will treat you (e.g. a nice salesman versus a rude phone support person a week later). This confusion is very much intentional on the part of the company: the marketing departments are very good at creating the image of friendlieness, or trustworthyness, or hipness, or whatever... but this bears no correlation to the actual engineering or sales departments.
It's been said before that if corporations are persons then they are surely insane persons. Indeed. The problem is that corporations 'behave' in inconsistent ways. It's like they have mental disorders (bipolar? multiple personalities?), and hence violate the normal rules we would like to use for consistency and trust.
All of that to say that we should be very careful about assigning personality to corporations. A statistical analysis of a company is meaningful (e.g. "I use this company because 80% of customers who call the support line get a satisfactory solution within 5 minutes"), but we should not fall into the (natural) trap of treating the company as a single personality (e.g. "I use this company because it's always been nice to me").
Just in case you're serious (or someone else suffers from the misconception embodied in your post):
The issue here is not with users voluntarily using Google services (search, gmail, etc.). Rather it is with companies who want to outsource their data needs to Google. In addition to the visible public products that Google has, it also offers corporate solutions: for instance if a company wants to outsource their email system, or have Google run search and collaborative software for use inside the company.
Google is trying hard to make these new kinds of products work. But unfortunately U.S. laws mean that any data that ends up on Google servers can be snooped by U.S. authorities. Many companies don't like the idea that the U.S. government will have such broad access to their data. In many countries where strong privacy laws exist (Canada, U.K., etc.), allowing the data to be managed by a U.S. company would then actually be illegal--since the company couldn't guarantee integrity or privacy of the data.
The end result of this is that Google is at disadvantage in the global marketplace because of the over-reaching U.S. laws. Google isn't the only one, of course: I'm sure U.S. companies have been losing lots of contracts because international businesses are wary of storing or moving data through U.S. systems since it is now well-known that such systems are not immune to U.S. government monitoring or interference.
The article doesn't make it clear whether using optical communications is intended to reduce latency or increase bandwidth.
With respect to latency: the electrical signals travel at ~30% the speed of light, whereas the optical signals travel at ~70% the speed of light (it depends on refractive index, etc.). Over the distances we're talking about (as you said, mm to dm), that's only fractions of a nanoseconds delay savings. This is on the order of a modern computer's switching time. All this complexity to get rid of a one or two processor cycles of latency?
I suspect instead they are looking to increase bandwidth. An optical fiber can carry very high data rates. Moreover a single physical fiber can carry multiple simultaneous channels (e.g. different wavelengths of light). So the intention may instead be to create high-bandwidth links between various processors. Using on-chip lasers can make the entire assembly smaller and faster than the equivalent for electrical wires.
Really what they want, I think, is to implement the same kind of high-speed optical switching we use for transcontinental fiber-optics into a single computer or computer cluster. If you can put all the switching and multiplexing components directly onto the silicon chips, then you can have the best of both worlds: well-established silicon microchips that interface directly into well-understood high-speed optical switching systems.
To use the beloved transportation analogy: it's like moving your cargo off of trucks and onto a high-speed train. Yes it takes time to move cargo, but it's worth it if the time savings of the high-speed train are big enough (for long enough distances, the savings can be significant).
In this case, there may be a delay associated with signal processing, but if the optical transmission is sufficiently faster than an equivalent electrical one, then it's worth it. Considering that electrical signals themselves need to undergo various kinds of switching and processing anyway (data written or read from a bus), I don't know that converting to laser signals will add much of a delay.
I agree that having additional functionality (soundtracks, subtitles, chapter icons, menu system) grouped with video files can be great... however a raw video file has the advantage of being easier to play on a myriad of devices and being under the user's control.
I know nothing about HDi, so I don't know to what extent it locks out the user from accessing the internal data directly... but I really hate data containers that companies use to force user-hostile features (like unskippable content), and so I'm wary in this case.
Part of the problem is that the company (Meraki) pushed firmware upgrades to all the units, including older boxes purchased before their revised licensing model. The new firmware locks down the units, making it impossible to hack them and impossible to load custom firmware and bypass the new locks.
That's the really sleezy part--changing your licensing terms for new sales is annoying for loyal customers, but obviously can't apply retroactively to goods you've already sold. But this company is doing just that--trying to retroactively impose their new licensing and payment model onto units that were already sold under an open, permissive terms.
So even though they still have the free code, they are now blocked from loading the code onto their own purchased hardware. It's probably not impossible--a talented hacker can maybe bypass the firmware and load custom code again... but of course they shouldn't have to. It seems to me that Meraki has more or less broken into customer devices without permission and made unrequested changes--rather illegal as far as I know.
This technology is essentially what is used in atomic force microscopes, and was being investigated by IBM for data storage under the name "Millipede". It basically involves a huge array of cantilevers that have very sharp tips on them (typical tip size is 20 nm but smaller is possible). The tips are used to read and write dots on the surface.
So yes, this system has moving parts. The tips have to scan across the surface, and the cantilevers are basically springs that bend up and down as the tips move over the surface. This definitely has some wear issues to consider, but it's nothing like the large-scale and high-speed movements of a hard drive (where a >2" disk is rotating at >7,000 rpm). Instead, the tips are moving laterally by micrometers at most (the huge array is what allows a large surface to be probed), and the cantilevers are springing up-and-down by only nanometers. The movement in an AFM is controlled using piezoelectric deformation of quartz actuators. This small-scale movement is very robust and reproducible. Quartz oscillators can vibrate/move thousands of times a second continuously for years without much problem (think of oscillators used for clocks, etc.). Moreover this technology has been used in commercial AFMs for years, so it's well-understood.
The thousands of tips are probably all actuated together by a single piezo-motor. They move in unison which would actually allow for high-speed reading/writing (since thousands of bits are read/written at once). You're right that each tip is in principle a point of failure. However, with the right error-correction algorithm, the device could be built so that even if a few tips break, no data is lost.
I agree that the access time isn't going to be as fast as modern RAM, but it could very well be faster than modern hard-drives. I think this is intended as permanent storage, not volatile memory.
Everything except KDE packages will have long-term support.
After all, you can install Ubuntu 8.04 LTS, and then install the package "kubuntu-desktop" to add all the required KDE packages. Most of the system (kernel, servers, most apps, GNOME, etc.) will be covered by LTS and receive patches for years. The KDE components will only receive patches for 18 months (of course after that you can just upgrade to the next supported version of Ubuntu or Kubuntu anyway).
So, basically, the "fully supported" version of Kubuntu 8.04 will use KDE 3.5. You will be able to purchase commercial support from Canonical if you like, but in any case can always download and use it for free. Or, you can use the more experimental Kubuntu 8.04 Remix, which uses KDE 4.0. In this case you will receive the usual community updates and community support.
Both versions are free, but if you use the more stable version that includes KDE 3.5, you have the option of paying Canonical for commercial support, just like every other official Ubuntu and Kubuntu version. You don't have to pay Canonical to get updates or unlock features or anything... but for businesses who want support contracts the option is there.
And, as I said before, Canonical is opting not to consider Kubuntu 8.04 as an "LTS" release... which means that they will officially provide updates to it for "only" 18 months.
Perhaps you're referring to the fact that although Ubuntu 8.04 will be considered "Long-Term Support" (LTS), the corresponding Kubuntu 8.04 will not be LTS (it will still be supported, just not for as long). The reason for this decision being that KDE 4.0 is still "too fresh" for Canonical to guarantee that it will be stable-enough (and unchanging-enough) to warrant the LTS label. However Kubuntu will still be available, will still receive consistent patches and updates, and will still be Free and free.
Or perhaps you're referring to the effort to get Linspire's "Click 'n' Run" to work on Ubuntu, which would allow users to install commercial/proprietary software from repositories?
In any case, I think you're mis-remembering what you've read. As far as I know, Kubuntu and Ubuntu will remain Free and free.
I agree with you.
However I'll note that the double standard partly arises from a "Windows vs. Linux" myth. That is, Windows is supposedly "compatible with everything" and there are "drivers for every device." According to the myth, Windows isn't supposed to have those kinds of problems; only Linux has trouble with "strange hardware."
For those of us who know that it's a myth (and that both operating systems support a plethora of devices, though obviously not every single one), it's at least interesting to see a concrete example. Windows has driver problems too. In both Windows and Linux, non-existent or buggy drivers can ruin the user experience. And in both cases, if a user loads potentially unsafe software, they must accept the consequences.
There was a time when the Internet was more like a novelty or hobby project. Those of us using it were on the fringe, and nothing that we did on the 'net was vital.
That is no longer the case. The Internet has grown to become a vital infrastructure. Just about every business relies on the Internet to get their work done. It is an indispensable tool for students and academics. It has risen nearly to the status of roads or electrical power in terms of being depended upon by billions of people.
What's my point? My point is that with respect to most utilities (roads, water, electricity, phone) we wouldn't tolerate much interruption in service... and we certainly wouldn't accept companies squabbling as a decent excuse for degrading the infrastructure. Can you imagine driving to work one day and finding roads blocked because of a contract dispute?
I'm not sure what the answer is. Turning the Internet into a government utility has its own problems. Similarly, laws which require certain norms for the utility may be over-reaching or impotent. But, ultimately, we need to push for this critical infrastructure to no longer be treated as a best-effort hobby/entertainment service. We need companies (and possibly legislators?) to acknowledge that the Internet is critical, and that this means that uptime/bandwidth/QoS must be maintained at a high-level.
I think the difference is that oversight/transparency is required in cases where someone wields power. We don't need cameras watching government employees because they work for us, but because they have power over us. No one is suggesting that we watch the employees while they are driving home from work, eating lunch, or even doing routine paperwork. However, filming government agents as they wield tremendous power (cops on duty, meetings between government officials and lobbyists, etc.) is useful to the extent that it can help curtail abuses of power.
Similarly, an employer who wants to monitor all employees with cameras at all times is over-stepping their bounds and infringing on basic privacy. However I think most people would agree that there are times when an employer can justifiably record employee actions (with their knowledge, of course). For instance if an employee is assessing millions of dollars worth of diamonds, a record of their actions seems reasonable. One should also note that casino employees are recorded for similar reasons.
Finally, it's worth noting that when properly implemented, such systems serve to protect both the employer and employee. Taking the diamond assessing example again, the cameras not only help the employer employees who are stealing: they also allow an employee to exonerate themselves by using the footage ("they were all accounted for when I left the room").
To summarize: it's not a question of mere "employment," but rather a question of "oversight when people wield power."
As you point out, at least with open-source licenses, there are only a handful of major ones that cover the vast majority of software. Once you know about them, you can very quickly know how much control you'll have over the code, and can confidently download/install/use/modify as required.
There is no proprietary equivalent to this kind of well-organized and relatively homogeneous licensing landscape. (Of course not! Having "named" proprietary licenses would make it too easy for a customer to compare different product licenses and select the less onerous ones.)
The reason I ask is because it's possible that Dell's contracts with Microsoft did, in fact, obligate them to promote Vista by the stickers on computers. For instance their bulk discounts are tied to various deals, such as having "Dell recommends Windows Vista" on their website.
Also worth noting is that many OEMs were shipping systems with "Vista Ready" stickers long before Vista was finalized. They had no way of knowing how well Vista would ultimately run on the machines, other than what Microsoft was telling them. Still, the OEMs share the blame to the extent that it was irresponsible of them to trust Microsoft and put stickers on systems without being sure that their claims were correct.
Actually I would argue that there is plenty of worthwhile content on YouTube (and I'm not talking about unauthorized uploads of TV shows). The thing is that YouTube is like the Internet at large: there is lots of crap so if you just randomly poke around you will of course see a predominance of crap.
If you browse YouTube a bit, and subscribe to the channels that are actually worthwhile, you will quickly build up a feed of interesting stuff with new videos every day. You can use featured videos to get some ideas of new channels to consider. On the other hand, using "most viewed" and "currently watching" to find good stuff is a waste of time. As a random example of something "worthwhile" (in my opinion), consider Wallstrip--a show that does profiles on companies and stock trends, and is infused with sarcasm and wit. There are also channels that discuss science, that do decent original comedy, there is a national geographic channel, etc.
Frankly I think YouTube is dropping the ball a bit by not providing a more useful method of finding the best content. An Amazon-like "people who subscribe/rate like you also like..." would help alot. Just as Slashdot uses various tricks (moderation, friends/foes, etc.) to bring attention to the quality material, YouTube should work harder to bring the good material to the top. The current star-ratings, comment-ratings, and ranking-by-viewing are not working very well. Frankly I don't care about the ratings of YouTube at large; I care about the ratings of a finite subset of like-minded users.
'Hides' is probably the wrong word. One of the activities is a terminal, with which you can browse the conventional Linux filesystem normally. You can SSH into the XO, and use terminal commands to install new software. You can even install a new desktop environment (e.g. xfce) to replace sugar if you prefer. It's a low-power machine, but it's running a full-featured Linux distro.
Lately I've been entertaining the idea of moving to somewhere in the developing world where all the kids have XOs, and teaching them to code. That sounds like a fantastic (and altruistic) thing to do. If you're used to coding in Linux, and using Python in particular, you'll find coding on the XO to be a fun. Personally I find the built-in keyboard hard to use, so I usually connect a USB keyboard and mouse if I'm working on it for an extended period.
Since the original poster seems to prefer permissive licensing, he should also check out the GNOME Human Interface Guidelines 2.0. It's an extensive set of best-practices and guidelines, licensed under the GFDL. Thus he can tailor the guidelines to his needs and redistribute them without worrying about copyright issues (another poster suggested setting-up a wiki for his users, which could also work).
The KDE Usability Guide also has some good material, although at this time it looks much less mature than the GNOME docs.