Please create an account to participate in the Slashdot moderation system


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: Re:Economics (Score 1) 129

by MrKaos (#49343353) Attached to: First Nuclear Power Plant Planned In Jordan
It sounds like a cop out because it is a cop out.

Who knows our civilisation may be at its peak right now and we will never reach these technological heights again. For all we know our selfishness will drive humanity back to nomads with some crazy old man poking a stick in a fire saying 'We used to have great machines that could fly'. Not what I want, but just as likely.

Comment: Re:Economics (Score 1) 129

by MrKaos (#49343229) Attached to: First Nuclear Power Plant Planned In Jordan

Could someone fill me in on the economics of nuclear power generation? I'd like to know what the usual payback period for a plant is, and how much it costs to operate a plant over that period.

Absolutely. Here is a link to the peer reviewed science that details net energy return after factoring input costs.

Have a great day!

Comment: Re:Hang on a minute (Score 1) 120

by MrKaos (#49340691) Attached to: Uber To Turn Into a Big Data Company By Selling Location Data

Comparing your mild annoyance at the thought that a company that you don't have to do business with could sell your data to a third party to slavery is incredibly offensive.

People that hide behind the freedom of anonymous speech I fight for to criticize me, offend me. Go an write a letter to your duly elected representative you are wasting your time here.

The Military

How Nuclear Weapon Modernization Undercuts Disarmament 223

Posted by Soulskill
from the perfecting-the-planetcracker dept.
Lasrick writes: John Mecklin details exactly how nuclear weapons modernization is kick-starting a new arms race, and how modernizing these weapons to make them more accurate and stealthy puts the world at even greater risk of nuclear war: "[T]his is precisely why the U.S. Congress rejected the Air Force’s requests for low-yield, precision-guided nuclear weapons in the 1990s: Their very accuracy increases the temptation to use them." The issue is not getting very much attention, but the patience of the non-nuclear states is wearing thin, and a breakthrough in public awareness may be on the horizon: "The disarmament debate is likely to make this spring's NPT conference a contentious one and just might be loud enough to make the public aware that a new type of nuclear arms race is unfolding around the world."

Comment: Re:The stupid is strong with these people! (Score 2, Informative) 146

by MrKaos (#49333949) Attached to: Draconian Australian Research Law Hits Scientists
Any chance we can talk about the meta data laws while there is still a chance to stop it? I know it's my submission but there is a slim chance that maybe we can do something if enough people know please please let there be some hope

Comment: Re:noatime,nodiratime (Score 1) 204

by MrKaos (#49332407) Attached to: Apple Doubles MacBook Pro R/W Performance

What benchmark made you conclude that HFS+ is faster than NTFS when using big block sizes ?

None, NTFS is a crap filesystem. I was just pointing out the comparison there is wasteful vs sluggish.

Does anybody still use reiserfs and what makes it "enterprise grade" ?

It's the fastest filesystem I've tested vs ext(s), xfs, and a few others. I had to do a lot of throughput testing on different filesystems so I wrote a battery of tests that helped me figure it out years ago.

Windozes server 2012 uses some of the principles from reiserfs, I don't know if that counts and I can't speak to who users reiserfs commercially but I use it whenever I need something fast and reliable.

The guy may be a killer but he knows how to write a filesystem and I doubt the US military has given up on their investment in it.

Mac OS X still depends on old mac system 6/7 filesystem functionality like resource forks, these are not that easy to "retrofit" in ufs/zfs.

Interesting - I didn't know that - but still sucks for mac users.

A lot of the IO schedulers are implemented mainly to have some IO fairness because mechanical hard drives are very easy to saturate.

and also to make it look like all the processes running behave smoothly - if you have a dedicated application though your still s.o.o.l on a mac

These aren't that useful anymore when you can push 500.000 IOPS to a set of SSD's. And don't diss the FreeBSD storage subsystem: ufs allows for consistent backups without having to use volume management and creating a snapshot beforehand (LVM2+ext4).

Now way would I dis FreeBSD - I'm might be a linux guy but I still think BSD is a solid offering - and good on them for having apple use their work - they deserve more credit. Though ext4 suffers its own issues if you need to have big directory structures.

I'll stick with my guns here though, any limitations on linux disk performance is a function of how well the controller drivers implement the hardware functionality. Configurable I/O, CPU scheduling, software and hardware raid coupled with filesystem choice make Linux reign supreme in terms of achievable I/O performance.

Comment: Re:noatime,nodiratime (Score 1) 204

by MrKaos (#49332181) Attached to: Apple Doubles MacBook Pro R/W Performance

Nah, I neglected to consider the on board controller quality of laptops - Apple have usually been pretty good in that regard - filesystems though - blech. It might be interesting to do a comparison with linux on a T series lenovo - which is a similar build quality an a mac but in reality mine was an observation about filesystem performance over hardware.

Even more interesting would be a comparison of similar macs with one hosting linux, but even that would just be limited by how well linux supports the apple controller under linux.

I have done extensive testing of filesystem throughput though, however those have been on higher end kit than laptops.

+ - Fukushima compensation recovery could take 30 years

Submitted by AmiMoJo
AmiMoJo (196126) writes "Japan's Board of Audit says it could take the government up to 30 years to recoup the funds it provided to help compensate victims of the Fukushima Daiichi nuclear accident. The government issued bonds worth nine trillion yen, or about 75 billion dollars, to help the plant's operator, TEPCO, pay compensation. The money is mainly intended to help those who have been forced to evacuate and to cover the costs of decontamination work. The board says that even if TEPCO pays the government half of its current profits every year, recovery of the funds will take until fiscal 2032 at the earliest. Interest on money that the government borrowed may amount to $1bn alone."

Comment: Re:noatime,nodiratime (Score 1) 204

by MrKaos (#49325311) Attached to: Apple Doubles MacBook Pro R/W Performance

I'm still waiting for the next laptop to even meet 2 years ago Apple's model.

Well, I'm a linux guy too, but to be fair to the windows folk HFS+ only achieves that performance by using a 16kb block size, not by having a performance filesystem and thus was very wasteful with disk space - especially when you consider how many small files exist on an fs.

No the disk performance crown still resides with Linux users that have access to enterprise grade filesystems like murderFS, ahem I mean reiserfs, xfs and other performance kings of that ilk.

I'm sure that there are macOS users out there who know how to retrofit such a filesystem into their BSD kernels, but that is hardly a stock MAC, it's BSD functionality. Even then I'm not sure if BSD supports configurable IO and CPU schedulers or a pre-emptable kernel - not that 99% of mac users would even understand why that is important.

Comment: Re: Be fair (Score 1) 178

by MrKaos (#49325001) Attached to: WHO Report Links Weed Killer Ingredient To Cancer Risk

people opposing GMO does not act as they would if they had reasonable cautions about the ecosystem.

Ok, so what has been done by these organizations to protect the contamination of the ecosystem's genome from GMOs. I'm kind of on the fence about GMOs however it seems to me there is as much ground to be cautious about deploying them when they can introduce species extinction by interfering with the germination of seeds.

It seems to me that whilst there are great benefits there are also great risks, especially when there is an abundance of food and the real issue is attempting to manipulate commodity prices through the practice of grain dumping at sea.

If we need more food, wouldn't it be more logical to end grain dumping and improve food distribution than to grow more food that will just be dumped at sea anyway? Maybe we should be trying to understand the genome better before we go modifying it in the ecosystem.

RAM wasn't built in a day.