That system runs at 1W@400MHz, although has no video-accelerating northbridge to add to the heat, it can play that MPEG4 video just fine (I am playing something similar now). We've designed it so the 2.5" hard disk actually sits about 5mm from the top of the CPU - if anything we're making cooling harder, and there is NO heatsink. The CPU does NOT power manage into SpeedStep style states - it just runs at 400MHz or "standby" (where it cannot run code until an external interrupt).
It runs fine. Mine's been on 24/7 for nearly a year, barring moving it around and connecting it up to things like new hard disks, changing power strips or measuring the power it uses. It never overheats.
What's the challenge meant to be? Just how crappy Via's chip needs to be that it CAN'T run at 500MHz on a 90nm process, and do without a heatsink of some kind?
Why would they buy a chip design company to make a few systems for their back end store?
There's no real advantage to using the current PASemi line-up for AppleTV or Mac Mini for those ends, and adding the components to shore it up to those requirements would take 18 months at least. At the very least. To what benefit? The current models work just fine.
More likely they have a plan for PA6T in server and storage, and will tap the company for talent in improving the iPhone such as a custom ARM SoC which does *exactly* what they want and not much else would be good - not better performing and probably not better power consumption as XScale already has those design goals sewn up, but if they can reduce the iPhone and iPod Touch to a single chip, that would be a cost goal and enable them to include more flash storage for the same price.
But mostly my bet is it's for servers and storage to start. Then Apple will branch into designing everything that doesn't come from Intel.
Power Architecture is not a chip design, it's an ISA design license. You pay IBM and you get to make a chip which runs code off the Power ISA.
They didn't "make design/manufacturing changes to make chips more energy efficient", they started the core and all the peripheral units from scratch, from the ground up, and just made sure it ran 64-bit PPC code at the end of it. Far more than an IBM licensed chip:)
What is hard to get to grips to is what plans Apple have for a company that makes PPC chip designs, when they have just thrown PPC away and moved to x86, and are making inroads to ARM. It may be they want the ARM (StrongARM/XScale) design experience that PASemi's staff have, and not the PPC chip. But that's like throwing the baby away and drinking the bathwater..
Apple did not not like PPC processors. They loved PPC processors.
Steve Jobs just felt let down by the guys at IBM and Freescale when he wanted a chip that did this in this power envelope and they gave him THAT, 6 months or a year late.
There is no way you will find a PowerPC processor in an Apple mobile device like iPod or iPhone. It just doesn't make any sense to try and shoehorn PPC into a market where it's never gained a foothold. It's true that the only reason PPC isn't there right now is because nobody wants to throw away billions of lines of ARM code, ARM binaries and ARM support with ARM operating systems after 10 years of using ARM, but that just makes it harder to change. Apple don't have that "legacy" (after all they run MacOS X on the iPhone, and MacOS X is already done for PPC..) but I still think it would be a wasteful thing to buy a company like PASemi and roll them into doing in-house iPhone chips. iPhone is about as cheap and power-friendly as it's going to get for a long time, so there is no point expending all these resources on a PPC iPhone.
Of course if they bought one out I'd be first in line; just I think it's unlikely.
PASemi's big markets are currently in the server storage market. I think this is more likely to be a play for the next XServe RAID, SAN software and even to bop IBM on the head given the release of POWER6. If you can't afford or justify a POWER6 system, you could probably buy an Apple PASemi rack with 16-64 cores per 1U for a fraction of the price (and greater aggregate performance).
What is missing here is some sense on the part of the news reporters, who obviously don't understand the difference between highly embedded portable devices and a low power consumption network processor. PASemi certainly do NOT specialise in low power chips for "small devices", they specialise in low power chips for *communications infrastructure* like storage, advanced image processing, cryptography and the like. I am finding it hard to imagine that a chip with capability to support 10Gbe ports and a huge amount of comms bandwidth, transitions to "it's the next iPod processor".
1) widescreen panels do not come much cheaper than 4:3 panels for a lot of reasons 2) scrolling does not mean lower productivity
So, laptop manufacturers are not being "selfish" and they are certainly not trying to beat down coders. What was your point again? Oh, ranting on Slashdot's front page! Of course!
Next time get a provable argument and some verifiable facts under your belt before asking Slashdot why Lenovo, Apple and Dell are trying to make your life a living hell by making you use a scrollbar.
If Google goes down, Windows Live Search still works. Not as well or as unobtrusively, but man.. if you're in a pinch and your books just burned in a fire and you need to get a map to the hospital burns unit.. I think we would all go there.
With Ubuntu and SuSE and KDE and GNOME releases going on at a pace, Intel dealing with desktop/laptop powermanagement, wireless and so on drivers being the hot topic in the kernel, why does the Linux Foundation need to bother with organising more development on the desktop?
On the other hand, servers are getting fancier every day. Infiniband, 10Gbit/100Gbit ethernet, clustering are all real important to get a hold on or Linux is going to be left behind in favour of something else. If you want to run a datacenter, are you going to wait 12 months to use the latest and best technology for your needs, while some hobbyist hacker reverse engineers the Windows implementation for OpenBSD and then someone slaps a GPL license on it and ALL HELL BREAKS LOOSE?
Of course Unix/Linux developers have been suggesting that you setuid your programs if they need root access and can't be given it (Squid auth daemons like pam_auth, are a modern example). This doesn't require any user interaction.
What was that CD writer software on Linux that required setuid to burn to a CD? It was only 2 years ago, maybe even less? That didn't require any user interaction, and in fact only a kernel change forced them to rethink the way they did it.
Linux has all the same problems as Windows, just when an app fails on Linux, they nag the author that crashed (even if it's a kernel bug or a bug in some other app, they nag the author of the thing they were using that went A-SPLODE). On Windows, users instantly blame Microsoft for every nVidia driver bluescreen, every slowdown and crash caused by shitty Antivirus software or AOL Instant Messenger.
I have an apartment where the only ISP available is Grande Communications (a little Texas cable ISP). This is being regulated out by the FCC who say that apartment-deals like this are really bad for business and competition. I am glad of that.
So, if ComCast are the only ISP serving your area.. bad luck. But you're a very low percentage of the population. Unfortunately American runs on the benefit of the majority - where people have 5 ISPs to choose from. Personally, when I chose an ISP, I ignored the apartment complex deal and chose Speakeasy instead - ADSL rather than cable.
I'm kind of lucky where the apartment is, because it's literally around the corner from the old SBC (now AT&T) HQ, the telecoms provision is really excellent. But, most people in cities are like that. They do have 5 ISPs to choose from. So, they can choose different.
If you're unlucky, well, move somewhere better, if downloading torrents on the internet is SO important to you. I can't imagine you'd be doing it for a job.. so I don't see why I should be sympathetic to you:)
Yep, ComCast own the little bit of the internet between you and the rest of the world that isn't on ComCast, and probably operate a hell of a lot of hardware between you and the border router. What little of that they DON'T own is copper owned by some telcom carrier company between your house and your local phone exchange and ComCast are paying to license the use of that anyway:)
So yeah, ComCast can run the horse race however it likes. It can also run foxy boxing in the middle of the track too, while it does it. Their network. If you don't like it, get another ISP.
Net Neutrality is for morons. Only in the USA could you legislate that networks and routers that you own CANNOT be flow controlled, while strengthening physical international borders using biometrics, reshuffled border agencies and more strict immigration..
It's definitely spreading. The market for embedded developers in the UK is totally empty - the vast majority of people coming out of college and university learn Java, or god forbid something even more useless like C# which depends on Windows. If you're not looking for someone to develop a Windows app, if you're someone who's experienced VxWorks or QNX or Integrity, or even that has poked around in Linux, you have a really hard time of it..
I think the term engineer is overused, and overrated. I agree with Canada's stance on it - engineers have a degree in some engineering discipline or they're not engineers. In the US, you might call yourself a "software developer", but I have always gone with the far more flouncy "Consultant". After all, you may "engineer" software or "develop" in another, but you're ALWAYS the consultant between "guys in suits" and "customers" for what code can and can't, will and won't do.
It also seems to pay more than "engineer" jobs, I think in some countries it goes like the Army, Combat Engineers who.. dig holes and put tents up, maybe deploy a prefabricated bridge, maybe blow one up. It is hardly a requirement for an 10-year stint getting your second masters degree in an engineering discipline to do that. After all, if you needed it, it'd be a dead loss to the Army because you couldn't get enough of the guys to the right place to put up that IKEA "Brïgge".
And that - the "can't get enough" problem - is probably diluting the meaning of the word in the corporate world too. There never seem to be enough "software engineers", as in ones that could be stated an engineer in Canada, but everyone is employed as one in the US and UK. I wonder if we actually need more differentiation in our jobs.. a few more job titles, so everyone knows where they stand and what their job would really entail..
Just tell me how that is different to having an open source microkernel with the same drivers?
There is nothing to say a microkernel can't be built as a monolithic *image*, the definition of microkernel is the abstraction (message passing, module loading etc.), not the binary blob it comes in (after all, a lot of microkernels, plus their entire platform support on a certain hardware configuration, are loaded from ROM or flash..)
One could argue that with the exception of the things Linux really relies on being in the kernel (TCP/IP stack etc.), nothing really requires being built in to the kernel. If it can be built as a module, it should be, so that it can be loaded and unloaded manually or automatically on the basis of a problem. It could also be rebuilt without the full kernel. Building it into the kernel as a static and unmovable entity, I think, is a bad idea. Linux has far and away more than enough abstractions to keep modules seperate from kernel nitty gritty. However, Linux also has the problem of having no reliable interface for drivers - symbols change from version to version. Modules generally depend on very specific kernel versions (support going away from one version to a few higher). This makes modular kernels hard to maintain, although efforts like SuSE and RedHat's to keep their distro kernels binary compatible (part of the effort to keep many modules out-of-kernel-packages for external packaging).
Sure you can do filesharing on a dialup connection, or a 256kbit connection. The difference is getting a whole album worth of MP3s in an hour, or 2 hours, basically, compared to the megabit and higher guys who would be getting it in ~10 minutes. The people sharing files TO you need the high bandwidth connection, and let's be honest, most people don't care about their upload ratio on PirateBay..
However you're right, hiking a $2.95 connection by $5 is just fucking mean. Maybe those guys could pay an extra $1 a month, the 256kbitters $2 a month, and 1mbit and above, the full $5.. kind of a "you pay for what you can possibly grab" sliding scale.
So far no Slashdot poster in this thread gave a decent prior art.. it seems Yahoo stumbled on a user interface *idea* and patented it, and let me make a further point; are not attempting to do anything BUT that. Every time IBM or Microsoft or Google or one of the other billionaire software companies patents something, the world goes apeshit because they think somehow this will be used to smash into the ground the world of free software. However it has been proven that it would be just as much a motivation to patent it to protect the world from dickheads like Eolas, Forgent, and the many rampant disputes going on in the East Texas District Courts in the past and even as we speak. Yahoo could patent PNG now and as long as they don't start expecting licensing for it, it doesn't really matter.
There is no defense on software patents that includes complaining about how you could code the same thing in five minutes with a copy of GTK+ - patents are to protect invention, and the only way to invalidate an invention is to show that it is shit or impossible to achieve economically (which does not require patent review) or to show that someone invented it before you. So, I suggest we first investigate Yahoo's intentions for this patent, then we can scramble around being paranoid..
The way I read this, it's describing something similar to.. well, let's take Flickr as a Yahoo example. If you had an interface which you could pick photos on, but really don't have a great deal of space for myriad menus, and click-click-click operation. Imagine you select and drag a photo image, and the user interface darkens and presents a ring of graphical menu items - perhaps a trashcan, or a couple of previews of certain filter effects which you can drag the photo object onto and apply the effect.
That's my take on it. I'd say if that's what it's getting at (although it may be worded a little too generically, for sure), then this is quite a novel use.. not totally unique in a "oh that's so obvious" kind of jealous way, but novel enough to justify a patent.
Now I think of it, I actually think.. no, I am sure.. I have seen this menu operation as described in the patent claims on Nintendo's Wii Menu or Paint app or something..??
But if it does turn out to be useful to release it, and I don't want to release it open-source, I can't simply buy a commercial license and be done with it.
Yes you can. If software does not see a public release, it has no license; the GPL explicitly differentiates between private software for your own use and that which is made available to the public. Trolltech make the same distinction.
You can develop the software using the free library, as long as it does not see the light of day outside of your own use. There is no need to license software that will never leave your desktop or be used by anyone outside of your company. If you need to entertain a commercial release, then you can buy the commercial license and recompile/distribute that software with the commercial library.
I know a couple of projects which have especially done it this way; the software used was internal and used the free, non-commercial Qt library built from source. When it was decided to be released, due to licensing of other parts of the software it needed to stay closed source. So a license was bought.. and the software was released. Trolltech, nor any software company, would fault you for not planning to use commercial software, and there is no API difference in the commercial Qt and free Qt which means you must code for one and not the other.
So, don't be stupid. Just code it for free, now, for your own use (the licenses permit this) and then get a commercial license *IF* you see the need for a public binary release at the time. But you do not need to plan that ahead.
Wikipedia fails for one simple reason; most of the data is without citation and most of the data with citation relies on web links that do not work anymore. The documentation that IS correct has absolutely no attribution and to find out who wrote an article or various portions of it you need to delve into histories or use something like they use to prove that the government is using it for propaganda or companies are removing swathes of information that are disparaging by the IP blocks they're posted from.
Being able to sort information by far better categories (not just an encyclopaedia) and enforcing attribution means the scrupulous among us will be able to publish data on the knowledge base and get the credit for it, and be able to be *congratulated or better yet, corrected* on it.
With Wikipedia, if you don't like what someone wrote, you delete it. You change it. You add insults. Then you can't use any of the data from Wikipedia anywhere else because it's GFDL. The information is *so* free the only place you can read it is ON Wikipedia, or has spidered Wikipedia and presented the data verbatim on another site.. if Google allows authors to select their license themselves (be it a CC variant, GFDL or a true copyright with a restrictive clause) then this will only draw people in.
There is something wrong about trying to free information by putting it under a restrictive, blanket license. Not all content can be licensed the same way. Wikipedia is high maintenance - looking for citations, constant review by editors, vandalism watches, locking, even selecting for the front page..
As for the advertising, even Wikipedia needs to earn it's keep. To be honest I really really object to trying to read an encyclopedia entry and being told that the WikiMedia conference is going to be on a certain date, taking up 1/4 of my screen at the top of the page, or that I need to donate to the cause. Fuck that. I want to turn that damn advert off. I don't care about it. But, it's essential to keep the site going. You can't complain about it, because without impressing it onto people that they need to pay for the upkeep of the service, they won't.
So, how is this any different to advertising using Google down the side? Well, it isn't. Google needs to make money by selling advertising and authors should be given the opportunity to earn money for all the effort they put in, because after all, spending a couple of days writing a 10 page article on something is an action most people would like to be paid for even just a little.
All the evidence from the history of life is consistent with the assumption of an evolutionary past which cannot be actually proven. How do you show that 3 billion years of random mutations came up with certain biological models? It's pretty difficult without a 3 billion year simulation or experiment.
For all we know, the very start of life was placed intelligently and the random mutations aren't random at all - they're simply designed to happen over the course of life on Earth. While it may employ random chance there is nothing to say that nothing is predefined or intelligently placed by some other force. Random chance usually manifests itself
Richard Dawkins makes a good point in that resolving unexplainable phenomena and complexity to "God did it" only delays science in finding the correct solution to the problem and working out the intricacies and the depth of the complexity inherent in biological models. It's not really wise to just shrug it off - I really believe in stem cell research and genetic engineering. We can play God all we like. But that is not to say that just because we can, that some "Hand of God" did not do anything in the first place.
To limit it to a single entity or deity is a little naive; after all plenty of religious are polytheistic. Maybe an alien with four arms and blue skin did it, with a lot of little avatar helpers including one that looks suspicously like an elephant.. who knows? You don't. You can't prove it. Just like you can't really prove that Darwin's *THEORY* of evolution is actually correct. It only looks that way, statistically, and the concept is sound - that random mutations may occur and if they are of no true benefit or of a specific disadvantage then they are eventually coded out. However, is this selection of nature, just a general "life sucks" mentality to it all, or is it specifically designed to route the evolution of certain things into certain other things, by way of not having them mutate too much out of the norm, and die out before reaching some unknown future purpose?
You can't disprove intelligent design any more than you can prove that random numbers are truly random. You could collect data forever. Sometimes a coin flips heads 20 times in a row, that doesn't seem random, but in a million years your coin flips will eventually plateau out.
I know a lot of geneticists who delve deep into the dark secrets of DNA and coding and none of them believe this is a random accident of protein shuffling. They are not of particularly religious backgrounds, do not go to church at all, perhaps a few are what you might call spiritual.. but they do not believe that it is God as in Yahweh as in The One That Is In The Bible who did it, just that.. it's a bit too convenient even for billions of years of work.
I think it's important that there is a distinction between Creationism (in the context of specific religion) and Intelligent Design. The possibility that it is NOT just random chance has to be entertained. In that light, you can't - as a director of a science curriculum - say "I don't support the teaching of intelligent design for these reasons". You can't just cut it out of the curriculum. You can however teach it without targeting a specific religious framework for it and in fact the same way you can't have an opinion on a neutral matter like this if you're working for the state department as director of science curriculums, teaching it from a neutral religious aspect is also enshrined in the same 1st Amendment that got her fired. Teaching evolution and random chance as the ONLY way life can have been created and have been on Earth can and should be considered narrow-minded.
After all what is science if you refuse to accept one possibility over another? Stephen Hawking could be wrong about the black hole information paradox, and he came up with the theory underlying the reason why the paradox exists. Why can't evolutionary science be random chance on top of some underlying intelligent design, by God, Gods, Aliens, or Predetermined Laws or some other unknown?
What she did - as a government employee - goes right against the 1st Amendment, but her right to do it is enshrined in that same amendment. What will be interesting here is whether her freedom of expression is justifiably restricted in the interests of allowing the citizens of the state she works for the same rights of freedom of expression, thought, and to make their own opinions.
After all, you can't say evolution is proved, any more than you can say no intelligence had a hand in the creation of life on Earth. You have to take both sides. You have to TEACH both sides.
Going into detail like "fossils are God messing with us to prove our faith" and "the world is 9000 years old and was created in 7 days" is just bollocks, this isn't intelligent design, this is the Christian Bible. You could sum up Intelligent Design by showing a Star Trek TNG episode ("The Chase") which is a wonderful exponent of the intelligent design principle. But it doesn't involve God, it doesn't involve Bibles.. and it's a perfectly valid theory of the origins of life:)
especially considering how well Sony offered complete backward compatibility so perfectly on the PS2
It's not. The PS2 ran the same code (common MIPS architecture processor) and had the same IO controller for audio and controller access as the original Playstation. Compatibility was assured through hardware. The graphics controller was also superficially similar - enough at least to let your games run, just faster and with some blending tacked on top.
Nintendo hit the same concept with the Wii being able to play Gamecube games - it practically *is* a Gamecube, just faster.
The PS3 also includes some of the original components of the PS2 for compatibility's sake although they are starting to get rid of those right now, and they never worked so well. The Xbox 360 ain't even slightly like the original Xbox. New CPU, new IO, new graphics controller.. the whole thing is software emulation. It should be said that Sony and Microsoft have different compatibility goals - Microsoft want to keep Xbox owners happy by letting them use their old games and back catalogue (Nintendo have the same goal). After all in a world where Halo and Halo 2 (or Metroid Prime and Echoes) did so well, do you really want gamers to start from the third game and lose the ability to play the saga from the start?:)
Sony have decided nobody wants to buy a $500 console to play 8 year old games so they're ditching the feature. Since they continue selling the PS2 at ridiculously affordable prices, there's no point making the PS3 compatible. It's not like you can still buy a Gamecube or Xbox brand new, though, these days.
Slashdot Mirror
Since when was running a 1 Watt CPU without a heatsink regarded as a challenge?
http://www.genesi-usa.com/efika.php - plug plug
That system runs at 1W@400MHz, although has no video-accelerating northbridge to add to the heat, it can play that MPEG4 video just fine (I am playing something similar now). We've designed it so the 2.5" hard disk actually sits about 5mm from the top of the CPU - if anything we're making cooling harder, and there is NO heatsink. The CPU does NOT power manage into SpeedStep style states - it just runs at 400MHz or "standby" (where it cannot run code until an external interrupt).
It runs fine. Mine's been on 24/7 for nearly a year, barring moving it around and connecting it up to things like new hard disks, changing power strips or measuring the power it uses. It never overheats.
What's the challenge meant to be? Just how crappy Via's chip needs to be that it CAN'T run at 500MHz on a 90nm process, and do without a heatsink of some kind?
You can't go wrong with Thawte..
Why would they buy a chip design company to make a few systems for their back end store?
There's no real advantage to using the current PASemi line-up for AppleTV or Mac Mini for those ends, and adding the components to shore it up to those requirements would take 18 months at least. At the very least. To what benefit? The current models work just fine.
More likely they have a plan for PA6T in server and storage, and will tap the company for talent in improving the iPhone such as a custom ARM SoC which does *exactly* what they want and not much else would be good - not better performing and probably not better power consumption as XScale already has those design goals sewn up, but if they can reduce the iPhone and iPod Touch to a single chip, that would be a cost goal and enable them to include more flash storage for the same price.
But mostly my bet is it's for servers and storage to start. Then Apple will branch into designing everything that doesn't come from Intel.
If you read my comment, you'll notice I said that.
Power Architecture is not a chip design, it's an ISA design license. You pay IBM and you get to make a chip which runs code off the Power ISA.
:)
They didn't "make design/manufacturing changes to make chips more energy efficient", they started the core and all the peripheral units from scratch, from the ground up, and just made sure it ran 64-bit PPC code at the end of it. Far more than an IBM licensed chip
What is hard to get to grips to is what plans Apple have for a company that makes PPC chip designs, when they have just thrown PPC away and moved to x86, and are making inroads to ARM. It may be they want the ARM (StrongARM/XScale) design experience that PASemi's staff have, and not the PPC chip. But that's like throwing the baby away and drinking the bathwater..
Apple did not not like PPC processors. They loved PPC processors.
Steve Jobs just felt let down by the guys at IBM and Freescale when he wanted a chip that did this in this power envelope and they gave him THAT, 6 months or a year late.
There is no way you will find a PowerPC processor in an Apple mobile device like iPod or iPhone. It just doesn't make any sense to try and shoehorn PPC into a market where it's never gained a foothold. It's true that the only reason PPC isn't there right now is because nobody wants to throw away billions of lines of ARM code, ARM binaries and ARM support with ARM operating systems after 10 years of using ARM, but that just makes it harder to change. Apple don't have that "legacy" (after all they run MacOS X on the iPhone, and MacOS X is already done for PPC..) but I still think it would be a wasteful thing to buy a company like PASemi and roll them into doing in-house iPhone chips. iPhone is about as cheap and power-friendly as it's going to get for a long time, so there is no point expending all these resources on a PPC iPhone.
Of course if they bought one out I'd be first in line; just I think it's unlikely.
PASemi's big markets are currently in the server storage market. I think this is more likely to be a play for the next XServe RAID, SAN software and even to bop IBM on the head given the release of POWER6. If you can't afford or justify a POWER6 system, you could probably buy an Apple PASemi rack with 16-64 cores per 1U for a fraction of the price (and greater aggregate performance).
What is missing here is some sense on the part of the news reporters, who obviously don't understand the difference between highly embedded portable devices and a low power consumption network processor. PASemi certainly do NOT specialise in low power chips for "small devices", they specialise in low power chips for *communications infrastructure* like storage, advanced image processing, cryptography and the like. I am finding it hard to imagine that a chip with capability to support 10Gbe ports and a huge amount of comms bandwidth, transitions to "it's the next iPod processor".
1) widescreen panels do not come much cheaper than 4:3 panels for a lot of reasons
2) scrolling does not mean lower productivity
So, laptop manufacturers are not being "selfish" and they are certainly not trying to beat down coders. What was your point again? Oh, ranting on Slashdot's front page! Of course!
Next time get a provable argument and some verifiable facts under your belt before asking Slashdot why Lenovo, Apple and Dell are trying to make your life a living hell by making you use a scrollbar.
If Google goes down, Windows Live Search still works. Not as well or as unobtrusively, but man.. if you're in a pinch and your books just burned in a fire and you need to get a map to the hospital burns unit.. I think we would all go there.
what if he's into geek girls? :)
With Ubuntu and SuSE and KDE and GNOME releases going on at a pace, Intel dealing with desktop/laptop powermanagement, wireless and so on drivers being the hot topic in the kernel, why does the Linux Foundation need to bother with organising more development on the desktop?
On the other hand, servers are getting fancier every day. Infiniband, 10Gbit/100Gbit ethernet, clustering are all real important to get a hold on or Linux is going to be left behind in favour of something else. If you want to run a datacenter, are you going to wait 12 months to use the latest and best technology for your needs, while some hobbyist hacker reverse engineers the Windows implementation for OpenBSD and then someone slaps a GPL license on it and ALL HELL BREAKS LOOSE?
Of course Unix/Linux developers have been suggesting that you setuid your programs if they need root access and can't be given it (Squid auth daemons like pam_auth, are a modern example). This doesn't require any user interaction.
What was that CD writer software on Linux that required setuid to burn to a CD? It was only 2 years ago, maybe even less? That didn't require any user interaction, and in fact only a kernel change forced them to rethink the way they did it.
Linux has all the same problems as Windows, just when an app fails on Linux, they nag the author that crashed (even if it's a kernel bug or a bug in some other app, they nag the author of the thing they were using that went A-SPLODE). On Windows, users instantly blame Microsoft for every nVidia driver bluescreen, every slowdown and crash caused by shitty Antivirus software or AOL Instant Messenger.
I have an apartment where the only ISP available is Grande Communications (a little Texas cable ISP). This is being regulated out by the FCC who say that apartment-deals like this are really bad for business and competition. I am glad of that.
:)
So, if ComCast are the only ISP serving your area.. bad luck. But you're a very low percentage of the population. Unfortunately American runs on the benefit of the majority - where people have 5 ISPs to choose from. Personally, when I chose an ISP, I ignored the apartment complex deal and chose Speakeasy instead - ADSL rather than cable.
I'm kind of lucky where the apartment is, because it's literally around the corner from the old SBC (now AT&T) HQ, the telecoms provision is really excellent. But, most people in cities are like that. They do have 5 ISPs to choose from. So, they can choose different.
If you're unlucky, well, move somewhere better, if downloading torrents on the internet is SO important to you. I can't imagine you'd be doing it for a job.. so I don't see why I should be sympathetic to you
Yep, ComCast own the little bit of the internet between you and the rest of the world that isn't on ComCast, and probably operate a hell of a lot of hardware between you and the border router. What little of that they DON'T own is copper owned by some telcom carrier company between your house and your local phone exchange and ComCast are paying to license the use of that anyway :)
So yeah, ComCast can run the horse race however it likes. It can also run foxy boxing in the middle of the track too, while it does it. Their network. If you don't like it, get another ISP.
Net Neutrality is for morons. Only in the USA could you legislate that networks and routers that you own CANNOT be flow controlled, while strengthening physical international borders using biometrics, reshuffled border agencies and more strict immigration..
It's definitely spreading. The market for embedded developers in the UK is totally empty - the vast majority of people coming out of college and university learn Java, or god forbid something even more useless like C# which depends on Windows. If you're not looking for someone to develop a Windows app, if you're someone who's experienced VxWorks or QNX or Integrity, or even that has poked around in Linux, you have a really hard time of it..
I think the term engineer is overused, and overrated. I agree with Canada's stance on it - engineers have a degree in some engineering discipline or they're not engineers. In the US, you might call yourself a "software developer", but I have always gone with the far more flouncy "Consultant". After all, you may "engineer" software or "develop" in another, but you're ALWAYS the consultant between "guys in suits" and "customers" for what code can and can't, will and won't do.
It also seems to pay more than "engineer" jobs, I think in some countries it goes like the Army, Combat Engineers who.. dig holes and put tents up, maybe deploy a prefabricated bridge, maybe blow one up. It is hardly a requirement for an 10-year stint getting your second masters degree in an engineering discipline to do that. After all, if you needed it, it'd be a dead loss to the Army because you couldn't get enough of the guys to the right place to put up that IKEA "Brïgge".
And that - the "can't get enough" problem - is probably diluting the meaning of the word in the corporate world too. There never seem to be enough "software engineers", as in ones that could be stated an engineer in Canada, but everyone is employed as one in the US and UK. I wonder if we actually need more differentiation in our jobs.. a few more job titles, so everyone knows where they stand and what their job would really entail..
Just tell me how that is different to having an open source microkernel with the same drivers?
There is nothing to say a microkernel can't be built as a monolithic *image*, the definition of microkernel is the abstraction (message passing, module loading etc.), not the binary blob it comes in (after all, a lot of microkernels, plus their entire platform support on a certain hardware configuration, are loaded from ROM or flash..)
One could argue that with the exception of the things Linux really relies on being in the kernel (TCP/IP stack etc.), nothing really requires being built in to the kernel. If it can be built as a module, it should be, so that it can be loaded and unloaded manually or automatically on the basis of a problem. It could also be rebuilt without the full kernel. Building it into the kernel as a static and unmovable entity, I think, is a bad idea. Linux has far and away more than enough abstractions to keep modules seperate from kernel nitty gritty. However, Linux also has the problem of having no reliable interface for drivers - symbols change from version to version. Modules generally depend on very specific kernel versions (support going away from one version to a few higher). This makes modular kernels hard to maintain, although efforts like SuSE and RedHat's to keep their distro kernels binary compatible (part of the effort to keep many modules out-of-kernel-packages for external packaging).
Isn't it weird to think of a time in the world when there was no such thing as a helicopter?
Sure you can do filesharing on a dialup connection, or a 256kbit connection. The difference is getting a whole album worth of MP3s in an hour, or 2 hours, basically, compared to the megabit and higher guys who would be getting it in ~10 minutes. The people sharing files TO you need the high bandwidth connection, and let's be honest, most people don't care about their upload ratio on PirateBay..
However you're right, hiking a $2.95 connection by $5 is just fucking mean. Maybe those guys could pay an extra $1 a month, the 256kbitters $2 a month, and 1mbit and above, the full $5.. kind of a "you pay for what you can possibly grab" sliding scale.
The devil is in the details.
So far no Slashdot poster in this thread gave a decent prior art.. it seems Yahoo stumbled on a user interface *idea* and patented it, and let me make a further point; are not attempting to do anything BUT that. Every time IBM or Microsoft or Google or one of the other billionaire software companies patents something, the world goes apeshit because they think somehow this will be used to smash into the ground the world of free software. However it has been proven that it would be just as much a motivation to patent it to protect the world from dickheads like Eolas, Forgent, and the many rampant disputes going on in the East Texas District Courts in the past and even as we speak. Yahoo could patent PNG now and as long as they don't start expecting licensing for it, it doesn't really matter.
There is no defense on software patents that includes complaining about how you could code the same thing in five minutes with a copy of GTK+ - patents are to protect invention, and the only way to invalidate an invention is to show that it is shit or impossible to achieve economically (which does not require patent review) or to show that someone invented it before you. So, I suggest we first investigate Yahoo's intentions for this patent, then we can scramble around being paranoid..
The way I read this, it's describing something similar to.. well, let's take Flickr as a Yahoo example. If you had an interface which you could pick photos on, but really don't have a great deal of space for myriad menus, and click-click-click operation. Imagine you select and drag a photo image, and the user interface darkens and presents a ring of graphical menu items - perhaps a trashcan, or a couple of previews of certain filter effects which you can drag the photo object onto and apply the effect.
That's my take on it. I'd say if that's what it's getting at (although it may be worded a little too generically, for sure), then this is quite a novel use.. not totally unique in a "oh that's so obvious" kind of jealous way, but novel enough to justify a patent.
Now I think of it, I actually think.. no, I am sure.. I have seen this menu operation as described in the patent claims on Nintendo's Wii Menu or Paint app or something..??
Yes you can. If software does not see a public release, it has no license; the GPL explicitly differentiates between private software for your own use and that which is made available to the public. Trolltech make the same distinction.
You can develop the software using the free library, as long as it does not see the light of day outside of your own use. There is no need to license software that will never leave your desktop or be used by anyone outside of your company. If you need to entertain a commercial release, then you can buy the commercial license and recompile/distribute that software with the commercial library.
I know a couple of projects which have especially done it this way; the software used was internal and used the free, non-commercial Qt library built from source. When it was decided to be released, due to licensing of other parts of the software it needed to stay closed source. So a license was bought.. and the software was released. Trolltech, nor any software company, would fault you for not planning to use commercial software, and there is no API difference in the commercial Qt and free Qt which means you must code for one and not the other.
So, don't be stupid. Just code it for free, now, for your own use (the licenses permit this) and then get a commercial license *IF* you see the need for a public binary release at the time. But you do not need to plan that ahead.
Wikipedia fails for one simple reason; most of the data is without citation and most of the data with citation relies on web links that do not work anymore. The documentation that IS correct has absolutely no attribution and to find out who wrote an article or various portions of it you need to delve into histories or use something like they use to prove that the government is using it for propaganda or companies are removing swathes of information that are disparaging by the IP blocks they're posted from.
Being able to sort information by far better categories (not just an encyclopaedia) and enforcing attribution means the scrupulous among us will be able to publish data on the knowledge base and get the credit for it, and be able to be *congratulated or better yet, corrected* on it.
With Wikipedia, if you don't like what someone wrote, you delete it. You change it. You add insults. Then you can't use any of the data from Wikipedia anywhere else because it's GFDL. The information is *so* free the only place you can read it is ON Wikipedia, or has spidered Wikipedia and presented the data verbatim on another site.. if Google allows authors to select their license themselves (be it a CC variant, GFDL or a true copyright with a restrictive clause) then this will only draw people in.
There is something wrong about trying to free information by putting it under a restrictive, blanket license. Not all content can be licensed the same way. Wikipedia is high maintenance - looking for citations, constant review by editors, vandalism watches, locking, even selecting for the front page..
As for the advertising, even Wikipedia needs to earn it's keep. To be honest I really really object to trying to read an encyclopedia entry and being told that the WikiMedia conference is going to be on a certain date, taking up 1/4 of my screen at the top of the page, or that I need to donate to the cause. Fuck that. I want to turn that damn advert off. I don't care about it. But, it's essential to keep the site going. You can't complain about it, because without impressing it onto people that they need to pay for the upkeep of the service, they won't.
So, how is this any different to advertising using Google down the side? Well, it isn't. Google needs to make money by selling advertising and authors should be given the opportunity to earn money for all the effort they put in, because after all, spending a couple of days writing a 10 page article on something is an action most people would like to be paid for even just a little.
All the evidence from the history of life is consistent with the assumption of an evolutionary past which cannot be actually proven. How do you show that 3 billion years of random mutations came up with certain biological models? It's pretty difficult without a 3 billion year simulation or experiment.
For all we know, the very start of life was placed intelligently and the random mutations aren't random at all - they're simply designed to happen over the course of life on Earth. While it may employ random chance there is nothing to say that nothing is predefined or intelligently placed by some other force. Random chance usually manifests itself
Richard Dawkins makes a good point in that resolving unexplainable phenomena and complexity to "God did it" only delays science in finding the correct solution to the problem and working out the intricacies and the depth of the complexity inherent in biological models. It's not really wise to just shrug it off - I really believe in stem cell research and genetic engineering. We can play God all we like. But that is not to say that just because we can, that some "Hand of God" did not do anything in the first place.
To limit it to a single entity or deity is a little naive; after all plenty of religious are polytheistic. Maybe an alien with four arms and blue skin did it, with a lot of little avatar helpers including one that looks suspicously like an elephant.. who knows? You don't. You can't prove it. Just like you can't really prove that Darwin's *THEORY* of evolution is actually correct. It only looks that way, statistically, and the concept is sound - that random mutations may occur and if they are of no true benefit or of a specific disadvantage then they are eventually coded out. However, is this selection of nature, just a general "life sucks" mentality to it all, or is it specifically designed to route the evolution of certain things into certain other things, by way of not having them mutate too much out of the norm, and die out before reaching some unknown future purpose?
You can't disprove intelligent design any more than you can prove that random numbers are truly random. You could collect data forever. Sometimes a coin flips heads 20 times in a row, that doesn't seem random, but in a million years your coin flips will eventually plateau out.
I know a lot of geneticists who delve deep into the dark secrets of DNA and coding and none of them believe this is a random accident of protein shuffling. They are not of particularly religious backgrounds, do not go to church at all, perhaps a few are what you might call spiritual.. but they do not believe that it is God as in Yahweh as in The One That Is In The Bible who did it, just that.. it's a bit too convenient even for billions of years of work.
I think it's important that there is a distinction between Creationism (in the context of specific religion) and Intelligent Design. The possibility that it is NOT just random chance has to be entertained. In that light, you can't - as a director of a science curriculum - say "I don't support the teaching of intelligent design for these reasons". You can't just cut it out of the curriculum. You can however teach it without targeting a specific religious framework for it and in fact the same way you can't have an opinion on a neutral matter like this if you're working for the state department as director of science curriculums, teaching it from a neutral religious aspect is also enshrined in the same 1st Amendment that got her fired. Teaching evolution and random chance as the ONLY way life can have been created and have been on Earth can and should be considered narrow-minded.
After all what is science if you refuse to accept one possibility over another? Stephen Hawking could be wrong about the black hole information paradox, and he came up with the theory underlying the reason why the paradox exists. Why can't evolutionary science be random chance on top of some underlying intelligent design, by God, Gods, Aliens, or Predetermined Laws or some other unknown?
What she did - as a government employee - goes right against the 1st Amendment, but her right to do it is enshrined in that same amendment. What will be interesting here is whether her freedom of expression is justifiably restricted in the interests of allowing the citizens of the state she works for the same rights of freedom of expression, thought, and to make their own opinions.
:)
After all, you can't say evolution is proved, any more than you can say no intelligence had a hand in the creation of life on Earth. You have to take both sides. You have to TEACH both sides.
Going into detail like "fossils are God messing with us to prove our faith" and "the world is 9000 years old and was created in 7 days" is just bollocks, this isn't intelligent design, this is the Christian Bible. You could sum up Intelligent Design by showing a Star Trek TNG episode ("The Chase") which is a wonderful exponent of the intelligent design principle. But it doesn't involve God, it doesn't involve Bibles.. and it's a perfectly valid theory of the origins of life
It's not. The PS2 ran the same code (common MIPS architecture processor) and had the same IO controller for audio and controller access as the original Playstation. Compatibility was assured through hardware. The graphics controller was also superficially similar - enough at least to let your games run, just faster and with some blending tacked on top.
Nintendo hit the same concept with the Wii being able to play Gamecube games - it practically *is* a Gamecube, just faster.
The PS3 also includes some of the original components of the PS2 for compatibility's sake although they are starting to get rid of those right now, and they never worked so well. The Xbox 360 ain't even slightly like the original Xbox. New CPU, new IO, new graphics controller.. the whole thing is software emulation. It should be said that Sony and Microsoft have different compatibility goals - Microsoft want to keep Xbox owners happy by letting them use their old games and back catalogue (Nintendo have the same goal). After all in a world where Halo and Halo 2 (or Metroid Prime and Echoes) did so well, do you really want gamers to start from the third game and lose the ability to play the saga from the start?
Sony have decided nobody wants to buy a $500 console to play 8 year old games so they're ditching the feature. Since they continue selling the PS2 at ridiculously affordable prices, there's no point making the PS3 compatible. It's not like you can still buy a Gamecube or Xbox brand new, though, these days.