Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror

Comment Re:Capitalism (Score 1) 77

Even if, in general, Apple and Google are competitors in many sectors, they certainly aren't here because Apple doesn't have a competing search product. Of course, when you search on the iPhone is has to go somewhere (Bing? Yahoo!) and those queries are worth money, so it's good sense to come to a revenue-sharing arrangement. That's not collusion or conspiracy, it's just a regular business deal for something of value that another party cannot produce on their own.

By comparison, Apple cannot fab their own SOCs so they buy them from Samsung, for billions of dollars yearly. Those two are also competitors (and legal adversaries) in many sectors distinct from chip supply. That's not "collusion" either -- it's the fact that a large business has many interests and generally can keep them separated so each part of the company can function.

Comment Tools are judged ... (Score 3, Insightful) 329

Tools are judged by their ability to do the job repeatedly and without fail.

That's not how anything is judged -- they are judged by expected TCO. And that TCO includes initial cost, minus expected performance but plus the expected value of failures multiplied by the cost of each failure. All of these vary by the job that the tool is being asked to do.

To give an example, if the wrench is going onto a deep-sea oil platform where replacement will be very costly and will cause very expensive delays, the last factor is very high and so reliability will be at a premium.

On the other hand, the local auto mechanic probably has a dozen wrenches and a parts truck that comes around every other day that can bring a new one in for nearly zero overhead. So she might be willing to accept a higher failure rate.

On yet another (third?) hand, someone working in aerospace or other sensitive area will likely need a wrench that can accurately deliver a set amount of torque. In this case, the accuracy of the tool will be the most important concern, since failure of the product (satellite, jet engine, space shuttle booster rocket clamp attachment) will be far more costly than failure of the tool.

So there you have it, three examples of how making general statements about how we judge things is complete bollocks. The "right tool for the right job" might be cliché but the lesson is less about picking the right tool and more about thinking about the properties that are priorities for the job.

Comment Re: What the fuck has happened to Slashdot?! (Score 1) 176

* The fall of the GPL thanks to people wanting to use truly free licenses like the BSD and MIT licenses.

This is still news. It's an ongoing issue. It's causing the FSF and software like GCC to become irrelevant.

This is silly, GCC is just a technologically inferior option to LLVM/clang -- ask anyone working in the compiler space about it. Even the folks working on GCC admit that's not aging gracefully.

If GCC becomes irrelevant, they will have no one to blame but themselves.

Comment Re:Actual benefit in regular use vs. SATA3 SSD? (Score 1) 117

If you're not doing anything storage intense, the PCIe bandwidth is not going to make much of a difference. Same with NVMe, main advantage is at big queue depths.

Actually besides queue depth, a lot of the benefit comes from reducing host CPU usage, contention, latency and context switching. AHCI has a single global queue (of pretty limited depth) and so multiple threads doing IO need to either block or else incur the overhead of bouncing the IO to another thread. For your hypothetical enterprise application actually saturating on 16 cores with 32 threads, Amdahl's law starts to actually become an issue. In NVMe, each physical CPU core has its own personal NVMe command and completion queues, that it can issue to without waiting/snooping/blocking any other cores.

Finally, there's just a lot fewer driver layers needed. AHCI is bulky and complicated by comparison. Less work for the host driver means more cycles for your enterprisey applications.

[ And of course, you're right about the sports car analogy ... ]

Comment Re:Walls are free? (Score 1) 75

Meanwhile, books such as those from authors opposing the death penalty, favoring gun control, might cost double.

And that's where this facile comparison finally goes off the rails -- no one even remotely hinted that the introduction of this plan would increase (let alone double) the cost of any other plan. In fact, by lowering barriers to entry and increasing demand, it will likely lower the cost of every other plan.

Comment Re:Walls are free? (Score 2) 75

If I told you that you could live totally free of cost, but it would be in a prison, would you accept?

If I had the choice to enter and leave the prison at will? I certainly don't see anyone claiming that once a person uses the plan they can never leave ...

I mean, the salient thing about a prison isn't the lack of freedom to do what you will inside the prison, it's the fact that you can't leave of your own free will. If you can walk out at any time, it's just not a prison -- much the way this plan is not a prison, it's just a kind of not-nice-place.

Comment Re:Another year, another video codec... (Score 1) 285

And now you have x265 that's not hardware accelerated on anything but the most modern GPUs (and even then, only partially -- certainly not suitable for any set-top-box, tablet or mobile phone. Heck, even a laptop that has partial GPU supporting (or none) will burn through tons of battery watching it on a flight with no power plugs.

Netflix has to support all those platforms (and probably worse ones) -- and then you enter the idea of having multiple copies for every asset :-(

Comment Re:Will others follow suit? (Score 1) 175

It's not good for the environment that people keep binning stuff that is more than adequate for their current needs. If people's main use of computers is browsing the internet and watching cat videos, a ten-year-old computer is more than adequate for the task.

A ten year old computer probably uses about 10x the power of a modern machine, even a cheap one. So it might be adequate for the task, but if a newer one can do for 10W what it does for 300W, then that's a savings of more than a nickel an hour or ~$100/year.

Comment Re:Will others follow suit? (Score 1) 175

I have a work computer that has such pathetic computing requirements that spending even one penny on something new is one penny too much

Except that you are paying lots of pennies to power that old inefficient Pentium IV. And if you counted the difference in pennies from powering that versus buying a new efficient one every 5-10 years, you'd come up with lots of extra pennies.

Comment Re:Will others follow suit? (Score 1) 175

Shouldn't this be the other way around? The default should be x86-64 unless there is a specific software/driver/whatever that doesn't work properly in that configuration (in which case, hopefully you don't have to browse the web for it).

I mean, the last time anyone sold processors that didn't support was the Pentium 4, sometime a decade ago. Given Moore's law (and the rising price of electricity), you could replace that with a RaspberryPI of equivalent horsepower that will probably pay itself back in power in under a year (the P4 was a power hog ...).

And if you do have some custom setup that doesn't run on 64bit, I'm really sorry. Legacy support sucks (I know firsthand), but don't expect the rest of the world to keep updating their shit just because you have to.

Comment Re:Need more mature languages (Score 2) 232

Python provides no true concurrency due to global interpreter lock. Java is not suitable for realtime due to unpredictable GC, while C/C++ is not suitable for anything which should never crash or return random results due to memory corruption.

Python has multiprocessing for 'true concurrency' if you need it
Java is not actually used for anything real-time
C/C++ can be written safely if you are willing to be careful and unit-test (also managed memory with C++11/14 constructs helps the drudgery) with tools like ASAN and Valgrind.

Yes, those are hard problems, but it's also 2015 and we can come up with powerful compilers and JIT virtual machines. Going back to less concurrency than plain old shell scripts where '&' starts a true separate process is not an answer.

Good thing no one proposed that.

Slashdot Top Deals

It is difficult to soar with the eagles when you work with turkeys.

Working...