This chicken and egg problem is not that hard to overcome.
Start off with "traditional" filtering techniques, they are quite accurate and I suspect give a good enough sample size to get you started.
A second option may be to ask users to mark their spam manually for a day or so. That should also be manageable.
Lastly when there is one group up and running, as I understand it new users can be added without any problems. Just keep them out of the statistical pool (only check their incoming mails on spaminess against the rest of the existing pool) until this user has received enough mails to be included. Such as "seed pool" apparently exists already.
Make it illegal and fine the people who profit from it.
Easier said than done. First start with a legal definition of spam e-mail, that does not cover things like mailing lists. Personally I am sending out many mass mailings, on an opt-out basis (I harvest interesting mail addresses myself) - and get very few opt-outs and many reactions. I specifically send mails to people that may be interested in buying my goods. This should definitely be legal, it's a great marketing tool and helps my business very well.
What should be illegal (and I suspect is already) are the attempts at identity theft, selling prescription drugs (real or fake), selling fake brand products, etc. Sellers of this kind of products should be tackled by the police first, and secondly those helping them in their marketing - most notably e-mail spammers. That is at least partly where the real solution lies.
but it also doesn't work on non-Intel machines since you can't recompile for your architecture.
That's really not much of an issue though since you can always wrap the binary codec in an x86 emulator or disassemble and reassemble for your architecture.
Technically, maybe you can do this. The first may be the most workable option. However I have never heard of the second option having been done - if it had been done successfully it would certainly have been posted here on/. - and that means to me that it is so hard it's not worth it. Even for determined geeks with way too much time on their hands. The assembly instructions probably vary way too much over the various architectures.
What most people seem to forget about, we're here talking about a design patent. Now IANAL, however I have learned a little (really a little) about patents.
There are a few issues here, that do not make sense. This is about a Nigerian patent, and patents are regional. A Nigerian patent is not valid in the US and the other way around. Within each country one has to apply for a separate patent. I wonder how they think they can sue over a Nigerian patent in a US court. They should sue in a Nigerian court instead.
Furthermore as it is a Nigerian patent, it will not influence the alleged patented product sold outside of Nigeria. Again, US courts do not come into the picture.
And then it is a design patent. This is not an invention as such, and quite close to copyright. I have to say I forgot how a design patent works exactly, but if it looks different, even though it works following the same technical principle, then it is no problem. Machines are typical objects that are patented for design.
This whole story sounds like a big mess of FUD to me, from someone who has done presumably great work to develop some input method for the Nigerian language, and now tries to cash in on that via dubious methods. I really hope the OLPC team is not held up too much by this, and that if there is a court case filed, that the courts simply do not accept it.
A Dutch invention, from 1994. And then to think that in The Netherlands the problem has never been that bad!
The counter-script it's called, and it's here: http://www.xs4all.nl/~egbg/counterscript.html
From the website:
The Direct Marketing sector regards the telephone as one of its most successful tools. Consumers experience telemarketing from a completely different point of view: more than 92% perceive commercial telephone calls as a violation of privacy.
Telemarketers make use of a telescript - a guideline for a telephone conversation. This script creates an imbalance in the conversation between the marketer and the consumer. It is this imbalance, most of all, that makes telemarketing successful. The EGBG Counterscript attempts to redress that balance.
I'm not affiliated with the site, I just happen to know about it. I never even tried it, when a telemarketer calls I usually just hang up.
I don't think it should surprise anyone that when a new upgrade comes around for OS X that every Mac user is immediately going to descend on it.
Ahum. Not so. I call myself a nerd, playing with computers as my hobby. I re-install Linux every year or so, mainly just for fun as the previous version is good enough, and I can download it for free anyway so why not try it out with all the new eye-candy and so. But only on a computer that is not essential for my work; work computers follow later or not.
I also own an iBook. Bought it now four years ago. Still running OSX 10.3.9. It does what I want nicely, no complaints. I'm not going to put down something like US$160 or 200 just for the latest eye-candy. Sorry, Apple, you do great things, but this is not for me.
In a year or two I may buy a new iBook or whatever they have on offer then. I love the machine, works great. And then it'll have the latest O/S as well of course.
So long story short: SOME mac users may queue up and shell out for the latest and greatest. SOME mac users may do so even because there is some feature they must have. But then, if you buy a computer (not just a mac) for a FUTURE must-have feature, why buy the product in the first place?
That said, I wonder why there are no CMYK TFT monitors? One layer of CMY that sets colour, and a K layer behind that for brightness.
Light mixes differently from ink. This has to do with the light receptors in the eye: we have receptors for red, green and blue - all three of them together are interpreted by our brain as white. That's why monitors, TV's, and the old-fashioned beamers have three colours. Those old beamers had three beams: one for each colour.
Ink colour blending works different. There you need cyan, magenta and yellow to create all the colours, and black to make it darker. White is not included as paper is the white, the fifth ink you need to be able to create any colour. Apparently there is some colour intensity included in the CMYK stuff. I don't know about that.
Well, for starters, I have written GPL'ed software and released this on a.o. Sourceforge - but that does not make me a member of the "GNU company", whatever that may be. So I doubt that will allow me a copy of the documents that GNU may buy.
That said, I do understand the idea of GNU or so buying this documentation, and maybe write a reference implementation. Is this encouriging the use of closed source software? Maybe - maybe not. On one hand one may argue "it legitimises the software" by making it work with OSS, on the other hand, OSS is also made to work with Windows better, making it a better alternative. And there is nothing wrong with that.
In the eight or so years I've been reading/. (only recently got an ID), I found here a strong tendency of "windows must be destroyed, closed source software must be destroyed". Parent's remark also has that air. I don't agree with this - for me the most important part is that software works together nicely. I can read my e-mail on my iBook, on my Linux computer with Evolution, and on another laptop running Windows and Outlook Express. All thanks to the IMAP protocol, running on my Linux server.
Compatibility, that should be the final goal. Open specs is a very important step towards that; having those specs implemented into standards (like now ODF for documents) is an even more imporant step. Honestly I do not care who made the standard, and how it came into being. Let the Exchange protocol be a standard - I hope this time it's a decent one, not like OOXML. Let it be documented thoroughly, and let there be several good servers and numerous clients that implement it. Then we are there: there is choice, and there is a standard.
And embrace-and-extend can also be tackled, as soon as it is a true standard.
Now imagine there are half a dozen clients with good implementation of the Exchange protocol. Some companies use Microsofts clients, others run OSX with an OSX client, others LInux, others Windows with a competing client. Now if there are enoug of those, Microsoft can not change the protocol without pissing off many companies, who then may consider switching to a competing server implementation.
And that is the world I hope to see sooner or later. Ah well I'm afraid it will remain dreaming for now.
As it is line of sight: how about people walking around the room while watching? There must be a certain line in between the transmitter and the receiver - if someone walks or worse, sits in between those two points: no signal anymore! That doesn't sound very convenient to me. Wifi at least goes through a person. No need to turn around to get a better signal because the access point is behind you.
Sounds like quite a concern to me. Especially as the aerials will be really small, so it will be really just a line in between the two points.
It's simple, you just drill a small hole in each wall to let the signal through; about 5.1mm should do it.
And at that point it's better than using a cable because... ?
It saves you the cable of course! I mean, it's not that easy to push a cable through a 5.1 mm hole in the wall.
...the resellers have to compete with each other to buy their music. Which likely gives rise to higher prices for the labels. This is based on the law of supply and demand. Of course, given the supply is infinite, demand will have little impact on its value.
I do not agree that music supply is absolutely infinite. There is a limited number of suppliers (record labels), and a limited number of songs available. Admittedly each song comes in unlimited copies, but you will normally not buy more than one copy of the same song.
So for example if you want to buy the product "Metallica songs", or "Britney Spears songs", then there is a limited number of songs from only single suppliers. If supply was truly infinite, then everyone could start producing "Britney Spears songs" - they can not, partly thanks to copyright (and no, this is not necessarily something bad: I am a copyright proponent, albeit not in it's current form), partly thanks to every artist being unique.
Now when for example five e-tailers want to sell Metallica, the record label has the right to simply say "ok, who is offering me the best price for those songs?" and limit the sales to only one or two of those competitors. That way the e-tailers have to compete, and record labels can raise prices. Record label (also independents can do this) wins.
On the other hand, if more than one e-tailer sells Metallica, then the consumer will go to the one with the lowest price, or the best service, or whatever. Thus again competition: customer wins. And the fact that each copy of the song can be multiplied infinitely, scarcity of the product will indeed not allow an e-tailer to take a too large profit on the sales.
It seems to me that DRM is digging it's own grave, thanks to the immense popularity of the iPod (I heard in the US like 80% market share).
The iPod uses DRM, but only Apple's DRM. And it can of course play unprotected songs.
Apple does not license it's DRM to other vendor: in effect becoming the only vendor selling DRM'ed songs to 80% of the market of digital music players. The rest of the players can fight of the left-overs.
Apple gets a lot of market power: the labels want to sell music, but only music with DRM. To reach the majority of the market, they must play together with Apple. And Apple has proven not to be very easy a business partner.
Thus the only way the music labels can tap into that 80% of the market, without going through Apple, is by selling non-DRM'ed songs. And there is a good reason for a music label to have multiple resellers for your product: then the resellers have to compete with each other to buy their music. Which likely gives rise to higher prices for the labels.
This way I see DRM having dug it's own grave: one DRM scheme became very popular, giving one player a very powerful virtual monopoly over online music sales. The label-mandated DRM now locks everyone in to that one player: Apple with their iTunes Music Store. And the only way to break this monopoly is to drop DRM, and that is exactly what is happening now.
And already we see the fruits of this development: iTunes forced to lower their prices, other stores offering flexible pricing options ('priced between 89 and 99 cents' - not much of a difference but there is flexibility), and certainly this will start opening the market for more online music resellers. This can not be a bad thing.
Getting even more off-topic: here in Hong Kong recently retail chain HMV started to sell tracks through ATM-style kiosks. Digital sales, but not online. These kiosks are in their retail outlets, offering buyers a huge collection (about half a million tracks or so; that requires quite a large brick 'n mortar store to house), and instant downloads to their digital music player. Again they use DRM: in this case Microsoft's Plays For Sure scheme. Now without DRM I'm sure HMV would have a much bigger market. I have no idea on the market share of Plays For Sure devices, though it's for sure less than half. So DRM free can instantly double one's market. If PFS devices are only 20% of the market (just a guess), they could increase their market five times just by dropping the DRM.
I doubt the record labels will ever agree that DRM limits their sales; confirming the R in DRM stands for Restrictions. Not Rights. Restricting not only what the user can do, but restricting your own market even more in the process.
Traditional wind farms (using wind turbines) are apparently not much of a problem for birds. I recall a study done years ago in The Netherlands, where some environmental protection group wanted to see how much damage the wind turbines were doing. The startling result: nearly nil! The explanaition: birds will not fly into the turbines because they are warned by the noise.
Now how that would hold up with the kites I don't know. My only experience is with kiting at the beach: we could sometimes see seagulls make sudden movements to avoid our line. I don't think they can see it, possibly they can hear it. I've never had a seagull or other bird actually hit my line when kiting.
two production Windows servers left, out of 15 or so
Is this "Mindbridge" a real company? I know geeks with 15 servers in their basement...
I don't know what business they are in (Safari crashes on TFA), but then: I have a very real company, two of them even, and I have only one server. It's doing what I need. But then I'm not in the business of selling web access, or server space, or so. Most companies have only one or two servers, because most companies are not in the business of selling server space. Besides, modern servers can handle a huge lot of work, one server now can easily handle what 10 servers did a decade or so ago.
I'm as a hobby also programming some web apps. Nothing fancy, no AJAX, mostly database interfacing. The widgets a browser provides are limited, but do the job for that. Programming the same in Gnome would be faster for the end user and look fancier, but then it's stuck to one computer, and one operating system. My web apps run on one server, and can be accessed from anywhere. And that is cool.
Web apps are ok but I prefer to have my apps with me, and not need to be connected to access them.
You are talking about a photo editor: that is an example to me of an application that is better off stand-alone. Just like a word processor. Anything that loads data, edits it, and later saves it. The software I'm thinking of (e-mail, accounting, crm,...) often requires multi-user access and a database to store it's data. It will always do calls to that database.
It depends on the application which method is best!
Web apps I think are very cool. The idea at least, they do seriously lack in functionality. But then web apps, it's nothing new. It's what the whole browser war was started about a decade or so ago. And even further back in time, we were talking about dumb terminals. The concept is the same.
I'm as a hobby also programming some web apps. Nothing fancy, no AJAX, mostly database interfacing. The widgets a browser provides are limited, but do the job for that. Programming the same in Gnome would be faster for the end user and look fancier, but then it's stuck to one computer, and one operating system. My web apps run on one server, and can be accessed from anywhere. And that is cool.
Also I started fiddling with sql ledger, again a web based interface. Again the widgets provided by the browsers do the job quite nicely. Heavy word processing, wysiwyg, etc are not necessary in that application. The fact that it is web based, so effortlessly supports multiple users, and is accessible even when I am not in office, are really great things.
Admittedly for e-mail applications it's a bit limited. Mostly the text processing is limited, but now Firefox supports spell checking already, and actually not much more I can think of for an e-mail application that one would need. We're talking about e-mail people! No need for fancy stuff.
The main issue is responsiveness, which is quite moot when you're in a small office on a simple, 100 Mb LAN. And when away from office, that's a sacrifice I'm willing to take for the convenience of it.
A new OS or fork that fails to gain more than 4% of the user base in 9 months could only be considered a success in Redmond.
4% in nine months looks a hell of a lot better than the ten years it took Linux and the twenty-three years it took the Mac to reach and hold the same market share.
You are mixing up market share and user base. The user base of Windows is over 90% of the market. The user base of Apple is about 4% of the market, that of Linux a little less.
This 4% of the user base is in case of Windows almost the same as total market share - that's because they are a monopoly. However you could compare it to the uptake of OSX in the Apple user base - within that 4% of the market. Not the total market.
The point: user base and market share are different things.
If those price cuts are a result from piracy, then I'd say piracy does have some positive effect. And the artists it won't hurt either, as these albums are primarily from major labels, where artists get a certain amount per album sold (if anything at all), not a percentage of the selling price. So for the artists it may even be good: at a fixed commission per album sold, more albums sold will see their revenues increase. And we all know that a lower priced product will normally result in more sales.
Sounds like a win-win situation to me! Cheaper CD's for the music buyers, more income for the artists!
You are talking about a mesh network. Should go fine. Maybe add a directional aerial to connect to it's peers over slightly longer distances, saves a 100% overlapping network.
But not to my living room...
on
Solar Powered Wi-Fi
·
· Score: 3, Informative
City-wide or country-wide WiFi sounds very cool to me. But - walls are a serious problem for WiFi. This in contrast to GSM signals, however in the concrete jungle called Hong Kong (with like 6 or 8 networks), even GSM is not everywhere available, particularly indoors.
I've wireless at home. It has a problem sometimes penetrating the two concrete walls between my living room (where the access point is) and my bedroom (where I sometimes use my laptop as well). I live on the 16th floor, a wireless access point on the ground level will never reach my living room. The penetration is too poor, and the distance is too long. So for city-wide WiFi, are there better solutions available?
In the countryside the problems are of course different - mountains are in the way and distances are often huge. Yet GSM networks are already fully covering even sparsely populated countries like Sweden and Norway. Is there a way do do so for WiFi without setting up repeaters every 500m? Is there a way to penetrate walls like GSM signals do?
The technology is nice, I love it. But at this moment for wireless networking on the go I will continue to use my mobile phone, over GPRS (yes we have UMTS available but that is mighty expensive, not worth it for me). It ain't fast, but it is virtually everywhere available, and has no problem keeping a connection when sitting in the train (try that with WiFi that is not in the train itself).
All and all I wonder, why not use the existing GSM networks? Most developed countries have UTMS available everywhere (USA is a developing country when it comes to digital technology, sad as it may be). Isn't that much more convenient, and cost efficient to use than a newly built WiFi network? There are more and more unlimited wireless plans (in Hong Kong you pay about US$80-100 per month for unlimited UTMS, add say US$200 a month for unlimited UTMS/GPRS roaming in mainland China). It's there, it's ready, and it's getting cheaper fast.
I may joke that in Google Sky, Rigel appears to be "(c) google" and Sirius will be a hotlink for digital radio, but there's a more serious concern of incomplete, poorly matched, patchwork quality, license-encumbered imagery that will blunt the value of Google Sky if they're not careful. Since Google's an ad company, I fail to see how this will actually bring them more revenue.
Google Sky, like Google Earth, will cost them money to set up. However, not that much, as the main infrastructure (huge distributed databases) they have in place already. It only costs them the labour to do so. But that's not bad for Google anyway, because now we're talking about them (again), they get press, more people (not everyone uses Google) use their search, and that's where they make their money.
Google is a young, rich, sorry very rich company. They can experiment a lot. They're not just about search anymore, they are about data management and distributed computing. Huge datasets they time and again prove to manage effectively and reliably. Earth, images, movies - all huge datasets, that require specialised database infrastructure. I have more and more the feeling that they do all this partly for fun, partly because they can, and partly simply as experiment. Images are relative large sets of data, especially when you have millions of them, and they are in high resolutions, possibly stored even with limited compression to make the stitching together part easier.
I've been playing with Google Earth now and then, and I love the street view. It's truly impressive how one can turn around in the street, with the images following. Borderlines between images may be a bit patchy at time, but considering it is all done automatically it's quite impressive. There is a lot of processing power behind that (they probably borrow some of my computer as well, but still).
Google by now has probably the most computing power of any company in the world. I wouldn't be surprised if full percentages of the world's computing power are in hands of Google already. Most of all I hope they stay true to their "do no evil" mantra, as I'm sure there is a lot of good that can come out of these experiments.
Arguing a company doesn't have to follow laws of a nation it does business in is just stupid. Of course they have to obey the laws, however the laws may or may not have exceptions (i'm not sure in this case).
The GP argued that a company based in a foreign country, doing business in that foreign country, but being owned by a US company, would still fall under US laws.
I totally agree with you - business one does in USA falls under US law. But if one ALSO does business in another country, that part of the business should NOT fall under US law, as it is conducted in the other country. According to GP, no matter what, as soon as a business is linked to the USA, all business done even abroad falls under US law.
Business done by Skype within the US will fall under US laws. But that does not mean that the complete company falls under US law. If they are incorporated in Luxembourg, they will have to do their accounts under that law, not under US law, for example.
The allofmp3.com case is in a way interesting. Of course it's a question who is doing the importing in such a situation. But that's a totally different issue.
Slashdot Mirror
This chicken and egg problem is not that hard to overcome.
Start off with "traditional" filtering techniques, they are quite accurate and I suspect give a good enough sample size to get you started.
A second option may be to ask users to mark their spam manually for a day or so. That should also be manageable.
Lastly when there is one group up and running, as I understand it new users can be added without any problems. Just keep them out of the statistical pool (only check their incoming mails on spaminess against the rest of the existing pool) until this user has received enough mails to be included. Such as "seed pool" apparently exists already.
Easier said than done. First start with a legal definition of spam e-mail, that does not cover things like mailing lists. Personally I am sending out many mass mailings, on an opt-out basis (I harvest interesting mail addresses myself) - and get very few opt-outs and many reactions. I specifically send mails to people that may be interested in buying my goods. This should definitely be legal, it's a great marketing tool and helps my business very well.
What should be illegal (and I suspect is already) are the attempts at identity theft, selling prescription drugs (real or fake), selling fake brand products, etc. Sellers of this kind of products should be tackled by the police first, and secondly those helping them in their marketing - most notably e-mail spammers. That is at least partly where the real solution lies.
That's really not much of an issue though since you can always wrap the binary codec in an x86 emulator or disassemble and reassemble for your architecture.
Technically, maybe you can do this. The first may be the most workable option. However I have never heard of the second option having been done - if it had been done successfully it would certainly have been posted here on /. - and that means to me that it is so hard it's not worth it. Even for determined geeks with way too much time on their hands. The assembly instructions probably vary way too much over the various architectures.
What most people seem to forget about, we're here talking about a design patent. Now IANAL, however I have learned a little (really a little) about patents.
There are a few issues here, that do not make sense. This is about a Nigerian patent, and patents are regional. A Nigerian patent is not valid in the US and the other way around. Within each country one has to apply for a separate patent. I wonder how they think they can sue over a Nigerian patent in a US court. They should sue in a Nigerian court instead.
Furthermore as it is a Nigerian patent, it will not influence the alleged patented product sold outside of Nigeria. Again, US courts do not come into the picture.
And then it is a design patent. This is not an invention as such, and quite close to copyright. I have to say I forgot how a design patent works exactly, but if it looks different, even though it works following the same technical principle, then it is no problem. Machines are typical objects that are patented for design.
This whole story sounds like a big mess of FUD to me, from someone who has done presumably great work to develop some input method for the Nigerian language, and now tries to cash in on that via dubious methods. I really hope the OLPC team is not held up too much by this, and that if there is a court case filed, that the courts simply do not accept it.
Not knowing the details, but:
- there were two types
- one was certified
- the other was sold
So the states bought knowingly an uncertified machine?
A Dutch invention, from 1994. And then to think that in The Netherlands the problem has never been that bad! The counter-script it's called, and it's here: http://www.xs4all.nl/~egbg/counterscript.html
From the website:
The Direct Marketing sector regards the telephone as one of its most successful tools. Consumers experience telemarketing from a completely different point of view: more than 92% perceive commercial telephone calls as a violation of privacy.
Telemarketers make use of a telescript - a guideline for a telephone conversation. This script creates an imbalance in the conversation between the marketer and the consumer. It is this imbalance, most of all, that makes telemarketing successful. The EGBG Counterscript attempts to redress that balance.
I'm not affiliated with the site, I just happen to know about it. I never even tried it, when a telemarketer calls I usually just hang up.
Ahum. Not so. I call myself a nerd, playing with computers as my hobby. I re-install Linux every year or so, mainly just for fun as the previous version is good enough, and I can download it for free anyway so why not try it out with all the new eye-candy and so. But only on a computer that is not essential for my work; work computers follow later or not.
I also own an iBook. Bought it now four years ago. Still running OSX 10.3.9. It does what I want nicely, no complaints. I'm not going to put down something like US$160 or 200 just for the latest eye-candy. Sorry, Apple, you do great things, but this is not for me.
In a year or two I may buy a new iBook or whatever they have on offer then. I love the machine, works great. And then it'll have the latest O/S as well of course.
So long story short: SOME mac users may queue up and shell out for the latest and greatest. SOME mac users may do so even because there is some feature they must have. But then, if you buy a computer (not just a mac) for a FUTURE must-have feature, why buy the product in the first place?
In Soviet Russia, Google searches you!
Light mixes differently from ink. This has to do with the light receptors in the eye: we have receptors for red, green and blue - all three of them together are interpreted by our brain as white. That's why monitors, TV's, and the old-fashioned beamers have three colours. Those old beamers had three beams: one for each colour.
Ink colour blending works different. There you need cyan, magenta and yellow to create all the colours, and black to make it darker. White is not included as paper is the white, the fifth ink you need to be able to create any colour. Apparently there is some colour intensity included in the CMYK stuff. I don't know about that.
Wouter.
Well, for starters, I have written GPL'ed software and released this on a.o. Sourceforge - but that does not make me a member of the "GNU company", whatever that may be. So I doubt that will allow me a copy of the documents that GNU may buy.
That said, I do understand the idea of GNU or so buying this documentation, and maybe write a reference implementation. Is this encouriging the use of closed source software? Maybe - maybe not. On one hand one may argue "it legitimises the software" by making it work with OSS, on the other hand, OSS is also made to work with Windows better, making it a better alternative. And there is nothing wrong with that.
In the eight or so years I've been reading /. (only recently got an ID), I found here a strong tendency of "windows must be destroyed, closed source software must be destroyed". Parent's remark also has that air. I don't agree with this - for me the most important part is that software works together nicely. I can read my e-mail on my iBook, on my Linux computer with Evolution, and on another laptop running Windows and Outlook Express. All thanks to the IMAP protocol, running on my Linux server.
Compatibility, that should be the final goal. Open specs is a very important step towards that; having those specs implemented into standards (like now ODF for documents) is an even more imporant step. Honestly I do not care who made the standard, and how it came into being. Let the Exchange protocol be a standard - I hope this time it's a decent one, not like OOXML. Let it be documented thoroughly, and let there be several good servers and numerous clients that implement it. Then we are there: there is choice, and there is a standard.
And embrace-and-extend can also be tackled, as soon as it is a true standard.
Now imagine there are half a dozen clients with good implementation of the Exchange protocol. Some companies use Microsofts clients, others run OSX with an OSX client, others LInux, others Windows with a competing client. Now if there are enoug of those, Microsoft can not change the protocol without pissing off many companies, who then may consider switching to a competing server implementation.
And that is the world I hope to see sooner or later. Ah well I'm afraid it will remain dreaming for now.
Wouter.
As it is line of sight: how about people walking around the room while watching? There must be a certain line in between the transmitter and the receiver - if someone walks or worse, sits in between those two points: no signal anymore! That doesn't sound very convenient to me. Wifi at least goes through a person. No need to turn around to get a better signal because the access point is behind you.
Sounds like quite a concern to me. Especially as the aerials will be really small, so it will be really just a line in between the two points.
Wouter.
And at that point it's better than using a cable because... ?
It saves you the cable of course! I mean, it's not that easy to push a cable through a 5.1 mm hole in the wall.
...the resellers have to compete with each other to buy their music. Which likely gives rise to higher prices for the labels. This is based on the law of supply and demand. Of course, given the supply is infinite, demand will have little impact on its value.I do not agree that music supply is absolutely infinite. There is a limited number of suppliers (record labels), and a limited number of songs available. Admittedly each song comes in unlimited copies, but you will normally not buy more than one copy of the same song.
So for example if you want to buy the product "Metallica songs", or "Britney Spears songs", then there is a limited number of songs from only single suppliers. If supply was truly infinite, then everyone could start producing "Britney Spears songs" - they can not, partly thanks to copyright (and no, this is not necessarily something bad: I am a copyright proponent, albeit not in it's current form), partly thanks to every artist being unique.
Now when for example five e-tailers want to sell Metallica, the record label has the right to simply say "ok, who is offering me the best price for those songs?" and limit the sales to only one or two of those competitors. That way the e-tailers have to compete, and record labels can raise prices. Record label (also independents can do this) wins.
On the other hand, if more than one e-tailer sells Metallica, then the consumer will go to the one with the lowest price, or the best service, or whatever. Thus again competition: customer wins. And the fact that each copy of the song can be multiplied infinitely, scarcity of the product will indeed not allow an e-tailer to take a too large profit on the sales.
Wouter.
It seems to me that DRM is digging it's own grave, thanks to the immense popularity of the iPod (I heard in the US like 80% market share).
The iPod uses DRM, but only Apple's DRM. And it can of course play unprotected songs.
Apple does not license it's DRM to other vendor: in effect becoming the only vendor selling DRM'ed songs to 80% of the market of digital music players. The rest of the players can fight of the left-overs.
Apple gets a lot of market power: the labels want to sell music, but only music with DRM. To reach the majority of the market, they must play together with Apple. And Apple has proven not to be very easy a business partner.
Thus the only way the music labels can tap into that 80% of the market, without going through Apple, is by selling non-DRM'ed songs. And there is a good reason for a music label to have multiple resellers for your product: then the resellers have to compete with each other to buy their music. Which likely gives rise to higher prices for the labels.
This way I see DRM having dug it's own grave: one DRM scheme became very popular, giving one player a very powerful virtual monopoly over online music sales. The label-mandated DRM now locks everyone in to that one player: Apple with their iTunes Music Store. And the only way to break this monopoly is to drop DRM, and that is exactly what is happening now.
And already we see the fruits of this development: iTunes forced to lower their prices, other stores offering flexible pricing options ('priced between 89 and 99 cents' - not much of a difference but there is flexibility), and certainly this will start opening the market for more online music resellers. This can not be a bad thing.
Getting even more off-topic: here in Hong Kong recently retail chain HMV started to sell tracks through ATM-style kiosks. Digital sales, but not online. These kiosks are in their retail outlets, offering buyers a huge collection (about half a million tracks or so; that requires quite a large brick 'n mortar store to house), and instant downloads to their digital music player. Again they use DRM: in this case Microsoft's Plays For Sure scheme. Now without DRM I'm sure HMV would have a much bigger market. I have no idea on the market share of Plays For Sure devices, though it's for sure less than half. So DRM free can instantly double one's market. If PFS devices are only 20% of the market (just a guess), they could increase their market five times just by dropping the DRM.
I doubt the record labels will ever agree that DRM limits their sales; confirming the R in DRM stands for Restrictions. Not Rights. Restricting not only what the user can do, but restricting your own market even more in the process.
Wouter.
Traditional wind farms (using wind turbines) are apparently not much of a problem for birds. I recall a study done years ago in The Netherlands, where some environmental protection group wanted to see how much damage the wind turbines were doing. The startling result: nearly nil! The explanaition: birds will not fly into the turbines because they are warned by the noise.
Now how that would hold up with the kites I don't know. My only experience is with kiting at the beach: we could sometimes see seagulls make sudden movements to avoid our line. I don't think they can see it, possibly they can hear it. I've never had a seagull or other bird actually hit my line when kiting.
Wouter.
Is this "Mindbridge" a real company? I know geeks with 15 servers in their basement...
I don't know what business they are in (Safari crashes on TFA), but then: I have a very real company, two of them even, and I have only one server. It's doing what I need. But then I'm not in the business of selling web access, or server space, or so. Most companies have only one or two servers, because most companies are not in the business of selling server space. Besides, modern servers can handle a huge lot of work, one server now can easily handle what 10 servers did a decade or so ago.
15 servers where I work is barely a ROUNDING ERROR
Not everybody is working at Google!
I'm as a hobby also programming some web apps. Nothing fancy, no AJAX, mostly database interfacing. The widgets a browser provides are limited, but do the job for that. Programming the same in Gnome would be faster for the end user and look fancier, but then it's stuck to one computer, and one operating system. My web apps run on one server, and can be accessed from anywhere. And that is cool.
Web apps are ok but I prefer to have my apps with me, and not need to be connected to access them.
You are talking about a photo editor: that is an example to me of an application that is better off stand-alone. Just like a word processor. Anything that loads data, edits it, and later saves it. The software I'm thinking of (e-mail, accounting, crm, ...) often requires multi-user access and a database to store it's data. It will always do calls to that database.
It depends on the application which method is best!
Web apps I think are very cool. The idea at least, they do seriously lack in functionality. But then web apps, it's nothing new. It's what the whole browser war was started about a decade or so ago. And even further back in time, we were talking about dumb terminals. The concept is the same.
I'm as a hobby also programming some web apps. Nothing fancy, no AJAX, mostly database interfacing. The widgets a browser provides are limited, but do the job for that. Programming the same in Gnome would be faster for the end user and look fancier, but then it's stuck to one computer, and one operating system. My web apps run on one server, and can be accessed from anywhere. And that is cool.
Also I started fiddling with sql ledger, again a web based interface. Again the widgets provided by the browsers do the job quite nicely. Heavy word processing, wysiwyg, etc are not necessary in that application. The fact that it is web based, so effortlessly supports multiple users, and is accessible even when I am not in office, are really great things.
Admittedly for e-mail applications it's a bit limited. Mostly the text processing is limited, but now Firefox supports spell checking already, and actually not much more I can think of for an e-mail application that one would need. We're talking about e-mail people! No need for fancy stuff.
The main issue is responsiveness, which is quite moot when you're in a small office on a simple, 100 Mb LAN. And when away from office, that's a sacrifice I'm willing to take for the convenience of it.
4% in nine months looks a hell of a lot better than the ten years it took Linux and the twenty-three years it took the Mac to reach and hold the same market share.
You are mixing up market share and user base. The user base of Windows is over 90% of the market. The user base of Apple is about 4% of the market, that of Linux a little less.
This 4% of the user base is in case of Windows almost the same as total market share - that's because they are a monopoly. However you could compare it to the uptake of OSX in the Apple user base - within that 4% of the market. Not the total market.
The point: user base and market share are different things.
If those price cuts are a result from piracy, then I'd say piracy does have some positive effect. And the artists it won't hurt either, as these albums are primarily from major labels, where artists get a certain amount per album sold (if anything at all), not a percentage of the selling price. So for the artists it may even be good: at a fixed commission per album sold, more albums sold will see their revenues increase. And we all know that a lower priced product will normally result in more sales.
Sounds like a win-win situation to me! Cheaper CD's for the music buyers, more income for the artists!
You are talking about a mesh network. Should go fine. Maybe add a directional aerial to connect to it's peers over slightly longer distances, saves a 100% overlapping network.
City-wide or country-wide WiFi sounds very cool to me. But - walls are a serious problem for WiFi. This in contrast to GSM signals, however in the concrete jungle called Hong Kong (with like 6 or 8 networks), even GSM is not everywhere available, particularly indoors.
I've wireless at home. It has a problem sometimes penetrating the two concrete walls between my living room (where the access point is) and my bedroom (where I sometimes use my laptop as well). I live on the 16th floor, a wireless access point on the ground level will never reach my living room. The penetration is too poor, and the distance is too long. So for city-wide WiFi, are there better solutions available?
In the countryside the problems are of course different - mountains are in the way and distances are often huge. Yet GSM networks are already fully covering even sparsely populated countries like Sweden and Norway. Is there a way do do so for WiFi without setting up repeaters every 500m? Is there a way to penetrate walls like GSM signals do?
The technology is nice, I love it. But at this moment for wireless networking on the go I will continue to use my mobile phone, over GPRS (yes we have UMTS available but that is mighty expensive, not worth it for me). It ain't fast, but it is virtually everywhere available, and has no problem keeping a connection when sitting in the train (try that with WiFi that is not in the train itself).
All and all I wonder, why not use the existing GSM networks? Most developed countries have UTMS available everywhere (USA is a developing country when it comes to digital technology, sad as it may be). Isn't that much more convenient, and cost efficient to use than a newly built WiFi network? There are more and more unlimited wireless plans (in Hong Kong you pay about US$80-100 per month for unlimited UTMS, add say US$200 a month for unlimited UTMS/GPRS roaming in mainland China). It's there, it's ready, and it's getting cheaper fast.
I may joke that in Google Sky, Rigel appears to be "(c) google" and Sirius will be a hotlink for digital radio, but there's a more serious concern of incomplete, poorly matched, patchwork quality, license-encumbered imagery that will blunt the value of Google Sky if they're not careful. Since Google's an ad company, I fail to see how this will actually bring them more revenue.
Google Sky, like Google Earth, will cost them money to set up. However, not that much, as the main infrastructure (huge distributed databases) they have in place already. It only costs them the labour to do so. But that's not bad for Google anyway, because now we're talking about them (again), they get press, more people (not everyone uses Google) use their search, and that's where they make their money.
Google is a young, rich, sorry very rich company. They can experiment a lot. They're not just about search anymore, they are about data management and distributed computing. Huge datasets they time and again prove to manage effectively and reliably. Earth, images, movies - all huge datasets, that require specialised database infrastructure. I have more and more the feeling that they do all this partly for fun, partly because they can, and partly simply as experiment. Images are relative large sets of data, especially when you have millions of them, and they are in high resolutions, possibly stored even with limited compression to make the stitching together part easier.
I've been playing with Google Earth now and then, and I love the street view. It's truly impressive how one can turn around in the street, with the images following. Borderlines between images may be a bit patchy at time, but considering it is all done automatically it's quite impressive. There is a lot of processing power behind that (they probably borrow some of my computer as well, but still).
Google by now has probably the most computing power of any company in the world. I wouldn't be surprised if full percentages of the world's computing power are in hands of Google already. Most of all I hope they stay true to their "do no evil" mantra, as I'm sure there is a lot of good that can come out of these experiments.
The GP argued that a company based in a foreign country, doing business in that foreign country, but being owned by a US company, would still fall under US laws.
I totally agree with you - business one does in USA falls under US law. But if one ALSO does business in another country, that part of the business should NOT fall under US law, as it is conducted in the other country. According to GP, no matter what, as soon as a business is linked to the USA, all business done even abroad falls under US law.
Business done by Skype within the US will fall under US laws. But that does not mean that the complete company falls under US law. If they are incorporated in Luxembourg, they will have to do their accounts under that law, not under US law, for example.
The allofmp3.com case is in a way interesting. Of course it's a question who is doing the importing in such a situation. But that's a totally different issue.