Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror

Comment: Re:Computers are making everyone's life easier (Score 1) 212

by nzac (#48357187) Attached to: New Book Argues Automation Is Making Software Developers Less Capable

If you feel it's just engineering, then maybe it's time for a change or perhaps some home project that reawakens the magic and art of programming for you again.

I don't think you understand the meaning of word engineering. Engineering and art are not mutually exclusive.

From a professional engineering perceptive what you are calling art is far closer to engineering than what you are calling engineering and once you start repeating processes without redesigning them you are no longer engineering at all.

Comment: Re:Non-system Admin Here (Score 1) 863

by nzac (#48258939) Attached to: Debate Over Systemd Exposes the Two Factions Tugging At Modern-day Linux

Except we have to do this in bash as our top level language. Bash is good for calling executables in a control sequence, but lacks the features to make it easy to use a standard c api.
It is not reasonable to expect the average admin to mix together all the features the kernel can offer (that list is just the beggining) in an imperative language.

Comment: Re:Non-system Admin Here (Score 1) 863

by nzac (#48248721) Attached to: Debate Over Systemd Exposes the Two Factions Tugging At Modern-day Linux

That aside, will people please stop this constant masturbation about startup times? There are way, way, way more important things to deal with than edging out a few more seconds.

systemd has was not developed for start-up times. The only way a Linux project gets this kind of funding and momentum is for the server. Better startup times are just show how shit sysV is.
systemd enables distros and admins to start using features in the kernel (cgroups, constrains ....) that would otherwise be unnecessary prohibitively difficult to implement in a generic imperative language.

Another way of looking at the situation is sysV is constraining everyones options to use otherwise available features to users.

... We need more scripting not less...

When that scripting is boilerplate, this is an unnecessary waste of time for the user and maintainer. If you really need some scripting for something nonstandard you can still call it inside an systemd service file.

Comment: Re:consistency more important (Score 1) 374

by nzac (#43639257) Attached to: Why US Mileage Ratings Are So Inaccurate

If the variation among how people drive is high, it doesn't matter what you pick, the measurements will always be inaccurate for most users.

It is most accurate in the middle, rather than as an extreme outlier, it would think the middle 80 percent fit to a bell curve. It is biased towards cars that are designed for the max speed of the 60s rather than today. You encourage designing cars to be geared for a lower max speed.

Looking at the variation it must cause some cars efficiency to be rated wrongly showing them incorrectly worse or better than another car.

Comment: Re:consistency more important (Score 1) 374

by nzac (#43636049) Attached to: Why US Mileage Ratings Are So Inaccurate

So you get the CAFE mileage for a trip with the same urban non-urban split without causing congestion?

There are too many ways in which people drive, and any choice of a standard is arbitrary.

Yes but you can arbitrary chose the average or median trip from samples.

If you have a middle trip then people can judge for themselves if they will be under or over and at the actual interstate limits which car is the most efficient. An urban, non-urban split in the rating would be useful as well.

Comment: Re:consistency more important (Score 0) 374

by nzac (#43633631) Attached to: Why US Mileage Ratings Are So Inaccurate

That's just stupid. It makes sense only if there is a (commonly known) conversion factor (or more generally a function of CAFE) that gives a reasonably accurate of what you can actually expect (knowing that tests expect you to drive somewhat conservatively).
It looks even worse when the CAFE value is so inaccurate for real world driving that it can not be used to compare different cars fuel efficiency.

When you have such a poor estimate of fuel efficiency that consumers should not be using to make decisions with, consumers can no longer chose cars based on their fuel efficiency and it improves at a slower rate.

Comment: Re:Intel to compete against Chinese $9 ARM chips? (Score 1) 319

by nzac (#43572461) Attached to: $200 Intel Android Laptops Are Coming

Atoms are still comparable in price, its an issue but not the major one. They probably have spare old fab tech to make these.
The main problem with this is that A15s are more powerful than atoms, both absolute and per watt. Also, the graphics in these is likely to be terrible to top it off.

Comment: Re:The King is dead (Score 1) 391

by nzac (#43380467) Attached to: Apple Devices To Outsell Windows For First Time Ever In 2013

The race to the bottom as complained about by tech reviewers (get the product for free), is the quality and value of the high end is reduced because the market is smaller due to people choosing a cheap option. With less competition and economy of scale in the high end there are less options with lower value.

Hypothetically if the cheep options were removed from market and you could only buy laptops that meet the ultabook spec, you would see a lot more diversity and value of Ultabooks.

Mercedes, BMW, Smeg fridges or Learjet aircraft.

These are strong brands like Apple that have universal brand recognition, OEMs don't really have this. Yes, some are more reliable than others but to the general public its still a windows laptop.

Just in case it did not come through, I think the race to the bottom is a good thing for the average consumer and me personally.

Comment: Re:The King is dead (Score 1) 391

by nzac (#43377341) Attached to: Apple Devices To Outsell Windows For First Time Ever In 2013

It's Win (x86) vs ChromeOS and Android vs RT. ChromeOS is cheaper because ARM hardware is much cheaper than x86 and Android is cheaper because it has lower hardware requirements and no office bundling.

MS has to do something about lowering the storage and other requirement for windows if it wants to compete at the 250 dollar point with a useable offering.

If you're not careful, you're going to catch something.

Working...