Forgot your password?
typodupeerror

Comment: Re:Nobody else seems to want it (Score 1) 601

by Grishnakh (#47721711) Attached to: Linus Torvalds: 'I Still Want the Desktop'

How much device driver programming have you done? It doesn't sound like you've done any, or know what you're talking about.

>Its true that the kernel maintainers (Torvalds & gang) maintain some drivers with the kernel, but not all.

Most drivers are part of the kernel. Any that aren't are "out-of-tree", and are either in the process of being put in the tree (most of these are in the "staging" area), or are not for some dumb reason and their maintainers have to waste time maintaining them separately. In practice, in a normal Linux distro, ALL drivers which a normal user uses for his normal desktop/laptop hardware are part of the kernel, except the Nvidia and ATI proprietary ones (IF the user chooses to use those).

>And what do you mean "tightly tied to the kernel"?

Device drivers call exported functions that are part of the kernel. Those functions have particular arguments. If the interface changes in the kernel, then any drivers which make use of those interfaces must also change. In a fixed API which some detractors call for, these interfaces are fixed and never change. In Linux, the maintainers don't believe in this because it limits flexibility and makes improvements later much harder (you end up creating new interfaces, but also keeping around the old ones for backwards compatibility, leading to code bloat). The way it is now, if they decide they want to add an additional function argument for some piece of hardware, it's no big deal, they just add it in, then modify all the drivers which call that function to add that argument. You can't do that with a fixed API, you have to create a whole new API (e.g., "function_call_V2(a, b, c, ...)").

>Oh, and don't get me started on GNU Hurd...

HURD is a microkernel, which is an entirely different architecture than Linux which is a monolithic kernel.

Comment: Re:I'd love to be in his class (Score 1) 153

by Grishnakh (#47721157) Attached to: Professor Steve Ballmer Will Teach At Two Universities This Year

>Then please, do name one. Please don't say "it's easy to do". If it's that easy, feel free.

I don't have to, just go read through the comments. You'll find a MS-lover sooner or later. No, they aren't nearly as numerous or loud as Apple lovers, but they are out there. If you think there isn't a single MS fan out there in the world somewhere, you're seriously delusional.

Comment: Re:Nobody else seems to want it (Score 1) 601

by Grishnakh (#47720627) Attached to: Linus Torvalds: 'I Still Want the Desktop'

No, it doesn't. Yes, it does use the Linux kernel, however many drivers in Android are closed-source. It's a real problem for community versions of Android like CyanogenMod, because this means you can only run some hardware on a certain kernel version, because you only have access to a closed-source binary driver. It's been a problem with a lot of other embedded ARM hardware too.

Comment: Re:Nobody else seems to want it (Score 1) 601

by Grishnakh (#47720579) Attached to: Linus Torvalds: 'I Still Want the Desktop'

They don't consider that a problem, they'll just tell you to buy newer hardware which supports >= Vista.

Remember, hairyfeet makes his living running a computer store, so it's entirely in his interest for people to simply throw away perfectly good but older hardware and buy new hardware (from him) to replace it.

But you're entirely correct: Vista introduced a new driver interface which was not compatible with XP, so users with XP-era hardware frequently found that their vendor didn't bother to issue updated drivers to make that hardware work on Vista, and just told customers to buy new hardware. This was not a problem with Linux since it doesn't do such things as it's open-source and community-maintained. If Linux adopted the MS driver model, Linux users would be constantly having to throw away perfectly good hardware whenever a vendor decides to not bother issuing updated drivers for new kernel versions.

Comment: Re:Nobody else seems to want it (Score 1) 601

by Grishnakh (#47720545) Attached to: Linus Torvalds: 'I Still Want the Desktop'

There's nothing wrong with drivers in Linux; hairyfeet always bitches about drivers in Linux but he really doesn't know what he's talking about.

The main controversy with Linux drivers is that they're tightly tied to the kernel, and maintained with it; there is no standardized API or ABI for Linux drivers, so a driver for one kernel version will likely not work with other versions; drivers have to be ported between kernel versions. In addition, drivers are absolutely not binary-compatible between kernel versions, and have to be compiled for them. In practice, this isn't a problem except for proprietary vendors who don't want to open-source their drivers. For everyone else, they just contribute their driver code to the kernel tree, it's integrated into the kernel, and maintained with the kernel. If an interface changes in a new kernel version, the maintainers update the associated drivers accordingly. When distros build kernels, the build the kernels and drivers together. So the only problem is a few proprietary vendors who don't like this model, and want to keep their drivers closed-source. Windows itself is a prime example of why this approach is bad: Windows has a terrible reputation for reliability, and much of the reason for this isn't actually Windows itself, it's shoddy drivers from various vendors, especially small Asian vendors of low-cost peripherals, but also big ones had problems. Drivers operate at the highest privilege levels, so if there's a bug in them, it can easily crash your system, so code quality is paramount. Crappy drivers crashing gave Windows a terrible reputation, and MS had to fix this by instituting their "WHQL" program, whereby vendors had to submit their drivers to Microsoft to be certified. This is a big reason why Windows is now fairly reliable. However, it's not something Linux can do; there's no single company behind Linux, and certainly not with the financial resources, or marketshare clout, to force vendors to go through such an expensive and onerous process. No company is going to pay millions of dollars to have someone certify their Linux drivers, but to run on Windows, they will.

The Linux driver model works just fine for Linux, and it's part of why Linux is so reliable and easy to install on most hardware. Instead of hundreds of drivers for simple devices running on the same chipset, for instance (a common happening with things like SD card readers) there's only one, maintained by the community, with high quality. The main problem is with video drivers (NVIDIA and ATI), because these are far more complex than an Ethernet or USB driver, which a few motivated programmers can write a driver for in a short time on a volunteer basis. This is why Linux has had trouble with video support with high-end video chips. These two vendors have closed-source drivers, but to get them to work with Linux and its frequently updating (due to security fixes) kernels is a little awkward and has caused problems in the past. However, open-source efforts including "Nouveau" have made a lot of progress and are pretty close to replacing the closed-source drivers.

Comment: Re:professor processor (Score 1) 153

by Grishnakh (#47720305) Attached to: Professor Steve Ballmer Will Teach At Two Universities This Year

Anyone can be a professor; you just have to get some college to give you that job. You don't need any kind of degree. It's just that, usually, colleges require an advanced degree (usually PhD) to be a professor, but they can hire whomever they want, so if they want to waive or lessen that requirement because of "industry experience", they can.

Comment: Re:I'd love to be in his class (Score 1) 153

by Grishnakh (#47720279) Attached to: Professor Steve Ballmer Will Teach At Two Universities This Year

>I'll challenge you to find _one_ loyal customer of any of those products, one who actually prefers it to an Iphone, Ipod, cheap notebook, or Windows 7.

I'm sorry to inform you, but it's not that hard to find a loyal customer for any of these products in online forums like this (though that person may just be a shill, it's impossible to tell). There's always some moron who pipes up and talks about how much he loves Windows 8 Metro or Surface or Windows Phone.

As for the Zune, no one uses those any more because, just like no one uses iPods any more: they've been made obsolete by phones. But there's a fair number of people who said they really liked their Zunes just for playing MP3s (back when they used them), they just didn't like the crappy sharing feature or the MS music store or the way MS screwed up "PlaysForSure".

Comment: Re:I'd love to be in his class (Score 1) 153

by Grishnakh (#47720231) Attached to: Professor Steve Ballmer Will Teach At Two Universities This Year

Your analysis is only partly correct; you've missed out on all the other business software they make tons of money on. MS is only highly profitable because of their business software, and the usage of their software in offices: Windows, Office, Sharepoint, Windows Server, SQL Server, etc etc. The place where they're failing abysmally is with consumers: they still sell (desktop) Windows of course, but they probably don't make much money with the home versions, and people aren't buying new PCs that much any more, and instead are buying smartphones and tablets (iOS and Android). MS's consumer offerings are ignored or laughed at: Surface, Windows Phone, etc. haven't done well. Xbox doesn't look like it's doing all that well any more either.

Basically, if MS cut out most of the consumer ventures, they'd be far more profitable. But there's definitely a tie-in there: people like to use software at work that they're familiar with, so if MS abandons the consumer space altogether, it wouldn't be long before companies shift to something else for their desktops, and then the rest of the MS infrastructure would crumble too.

Comment: Re:which turns transport into a monopoly... (Score 1) 267

by BigSlowTarget (#47714169) Attached to: Helsinki Aims To Obviate Private Cars
As with everything it depends on the service level and cost. If you live outside major cities your service level would be like Amtrak in those cities - terrible. This is proposed for a major city where it makes the most sense and I expect will lock the inhabitants (or at least the non-wealthy ones) into those cities by denying transport outside them and preventing them from traveling to less spoiled areas.

Comment: Re:Rolling roadblocks (Score 5, Interesting) 464

by ledow (#47705683) Attached to: Google's Driverless Cars Capable of Exceeding Speed Limit

Dunno about the US, but in the UK there aren't 4 lanes. There is one lane, and other overtaking lanes.

Technically, if you have four cars all at the same speed in all four lanes, at least three of them would be breaking the law (dunno about the US, assume it's similar). If they're overtaking, it's not a problem, because they have to pull back in when they've completed the maneouvure and you can overtake them then.

To be honest, robots obeying rules will make the roads I travel on move faster. It's the dickheads who constantly change lanes and try to "beat" the queues when speeds come down that cause most of the slowdowns and "phantom braking waves" that I witness every day.

And, to be honest, I'd rather get somewhere at 65 predictably than 70 unpredictably, in spasms and spurts and with sudden braking.

Comment: HAHAHA "bare-knuckled" (Score 2) 702

by poity (#47704969) Attached to: News Aggregator Fark Adds Misogyny Ban

It gets image macros from 4chan and still can't decide on what the irony tag is for. Fark has long become a shadow of its former self. The left leaning political correct mob has been in control for many years now -- even being a consistent critic of public officials on both sides of the political aisle gets you into trouble because in those times when you do write anything critical of a left leaning politician, you get 10 replies of people knocking down strawmen and breathlessly posting whataboutisms.

Comment: Re:"Hard redirect" (Score 3, Interesting) 375

by Entrope (#47702097) Attached to: Rightscorp's New Plan: Hijack Browsers Until Infingers Pay Up

The key element of a tortious interference claim is not the existence of a contract, it is third-party interference with a business or contractual relationship. sixoh1 was suggesting that someone might have a cause of action against Rightscorp, not the ISP, so the ISP's prerogative to terminate customer contracts is not relevant.

I am not now, nor have I ever been, a member of the demigodic party. -- Dennis Ritchie

Working...