Posts by John Hughes

Randall Munroe spoke to The Reg again. We're habit-forming that way

One for El Reg

1379

AMD sued: Number of Bulldozer cores in its chips is a lie, allegedly

Re: Reread the Article

So AMD "cores" are like Intel "threads".

80286?

Or maybe a 1904?

Kids of today, no historical perspective.

Has Voyager 1 escaped the Sun yet? Yes, but also no, say boffins

Re: @1980s_coder

You think BASIC and FORTH are obscure? You think VAX is "vintage".

Get off my lawn.

Re: old git alert

CORAL-66? Sorry, no can do.

Now, if you wanted some PO-CORAL doing I might be able to dredge my memoriies (NOT).

The story of .Gay: This bid is too gay! This bid is not gay enough! This bid is just right?

The Economist

Does anyone else find irt a bit odd that the go-to guys for deciding if the application is gay enough is the Economist?

Accidental homicide: how VoLTE kills old style call accounting

Re: Indeed

A lot of people are unemployed.

How? Voice networks haven't used people to switch calls for quite some time you know.

Top boffin Freeman Dyson on climate change, interstellar travel, fusion, and more

Re: (continuation of earlier response)

We currently have, for example, scientists starting to freak out about a great cold spot which has apparently formed in the North Atlantic, and which is probably going to play hell with their most dire global warming predictions for Europe and elsewhere.

You shouldn't believe everything you read in el-Reg, especially stuff about global warming.

Re: (sigh)

The observable fact that both Arctic and Antarctic sea ice extents are at or beyond long run seasonal averages would lead me to be sceptical of your claim.

"Arctic sea ice extent is at or beyond long run seasonal averages"?

Are you blind?

The last figure, as of posting (day 284) is 6.063 million km^2. The 1981-2010 average for the same day is 7.684 million km^2. What's 1.6 million km^2 between friends? It's just inside the 2sigma region, but only just.

The Antarctic is currently more or less at the 1979-2008 mean, 14.835 million km^2 vs a mean of 14.830 million km^2. That's likely to be around its high point, and is less than it was last year.

Re: Emeritus

@marshaltown

Try this.

https://www.york.ac.uk/news-and-events/news/2015/research/climate-models/

Seriously? You think I don't know how to follow a link? You're using Cowton et al (2015) to say models are wrong? A paper that demonstrates that most model/data differences are caused by incorrect comparisons?

The ensemble mean of the various models have ALWAYS been higher than empirical data, despite consistent adjustments that push modern temps upward and historical (pre-WWII) data downward.

Except that the adjustments don't do that. Most of the planet is sea, not land, and the SST adjustments show lower warming than the raw data.

When will you guys figure out that the continental US is a tiny part of the planet?

Re: @John Hughes: Emeritus

Are you unable to follow a link?

How could adjusting the ARGO data change the sea surface temperature before 1940?

The link I gave refers to the adjustments made in the HadSST3 dataset, refered to by Judith Curry as "In my opinion, the gold standard dataset for global ocean surface temperatures is the UK dataset, HadSST3".

(P.S. the changes made to the NOAA data still warm the past more than the present -- the adjusted data shows less global warming than the "raw" data).

Re: @John Hughes

Hur Hur, I replied to your message in English, rather than your "American with funny punctuation" and the mods rejected it.

El-Reg out-nannies the Graun.

Seriously, why do you think it's "being an ass" to ask someone to justify their ridiculous statements?

Re: Emeritus

The key word here being "look", models "look" as if they're running off target. What that paper says is simply that natural variability can, over short periods, hide the trend, something we already know. (I.E. MSTC 2015 definitely doesn't say "models are getting worse"). (*)

Other recently published papers were able to show this rather more clearly, for example by choosing model runs where the randomly chosen ENSO signal happened to match the observed ENSO signal, or by imposing the ENSO signal on the model.

(* It's amusing to note that the so called "pause" wasn't the first time the model ensembles have looked different from the observed climate, but nobody complained when they looked like they were running cold.)

Re: Emeritus

If an elderly but distinguished scientist says that something is possible, he is almost certainly right; but if he says that it is impossible, he is very probably wrong.

-- Arthur C. Clarke

Science doesn't work by argument to authority.

If Dyson had some real argument about model quality he'd be pointing at published research saying it was falling, rather than unfounded assertions.

Re: @John Hughes: Emeritus

You mean those adjustments that reduce the global warming signal?

What, you don't know that the largest adjustments made to the historical temperature data had the effect of warming the past and cooling the present?

I wonder why.

Re: Bob Forward? Try Larry Niven

You change the properties of the laser to take energy out of the ship/sail system - creating a low energy vacuum behind the sail thus slowing it down by "sucking it" backwards.

Got some kind of citation or source for that, because it sounds like bollocks.

It's all cycles -- or is it.

Well, I hope Dyson was just trying to be polite here. If he really agrees with this nonsense he may be losing it.

http://www.woodfortrees.org/plot/rss/from:1980.1/plot/rss/from:1980.1/to:2003.6/trend/plot/rss/from:2003.6/trend

Oh, RSS is broken, as if we didn't already know that.

Re: Emeritus

I'd believe published papers, not what some guy not involved in the field says.

Can you cite any published research that says models are getting worse? No. Can you cite published research that says models are fitting the data pretty well? Yes.

Re: Bob Forward? Try Larry Niven

You mean he didn't spot the minor problem that it wouldn't work? (Niven had to write l The Ringworld Engineers to "fix" the problem).

Emeritus

Maybe at work, but not bothering to keep up with the research.

What has happened in the past 10 years is that the discrepancies between what's observed and what's predicted have become much stronger.

That's simply rubbish.

Linux kernel dev who asked Linus Torvalds to stop verbal abuse quits over verbal abuse

Re: The problem is, usually Linus is right

Have you *seen* systemd?

Yes. It's pretty good, a shitload better than the junk it replaced. What's your point?

Re: The problem is, usually Linus is right

Poettering usually just claims things which are quickly disproved.

Example?

Ubuntu 15.10: More kitten than beast – but beware the claws

Re: SystemD

But with systemD the one daemon that "dynamically handles device management, mount points partition discovery, and power management" is udevd, as it has done on most Linux systems for many years now. You want to go back to HAL? devfs?

As for whether that stuff is useful for a server, I would remind you that RedHat mostly concentrate on servers, and explicitly want systemd for server applications.

And you still won't come clean on the "interrelated dependencies" that worry you.

Re: SystemD

Edit: You started by asking two appropriate questions, but later in the thread appear to be angry about this.

I got a bit angry because my in my hung-over state I found the FUD being flung around a little indegestible. I don't think another beer would be a good idea.

The use of interrelated dependencies of systems that should be kept separate encourages "standard" distributions

Purest FUD. What "interrelated dependencies" are you talking about?

Being a cute kitty is hard work

Poor little thing -- expected to be cute 100% of the time. Sometimes it gets pissed off, who could blame it.

Seriously, it's a tremendously cute kitty. There must be something wrong with you.

Re: I think it is also a PR issue

Secondly, the way it was introduced to Debian was a pure disaster along with arrogant, rude acting developers.

How are people supposed to act when told: "Hey, that work you are doing for free, you'd better do it this way, even though it will be a lot more work for you and you think it's a waste of time".

If the various dodgy "anti-systemd" moves had got through Debian would be dead by now. As it was the shitstorm thrown up by the anti-systemd trolls lost us some good developers.

Re: SystemD

I am so old that I remember the Berkely Distributions, and still use it.

Don't try to get into an old fogey war with me - I started programming Fortran on 80 column punched cards.

[SystemD] breaks one of the main strengths of UNIX - That every component stands by itself and can be managed separately.

Damn, could you be less explicit if you tried? What are you waffling on about.

The use of interrelated dependencies of systems that should be kept separate encourages "standard" distributions

Ah, so libc is a bad thing because everyone depends on it. You're against people writing useful software because people might use it. Gotcha.

[Systemd] will allow organizations like, say, Canonical to distribute a "premium" commercial product (like Red Hat) that will tend to limit user and developer choice; and encourages loading unnecessary insecure cruft.

So just use Debian then. Trusting a commercial company to provide a free system has always seemed to be a mugs game to me -- you just end up as an unpaid beta tester.

Re: SystemD

Please elucidate.

1. Why don't you want it.

2. How on earth could it be part of a "conspiracy" to "monetize Linux"?

The Steve Jobs of supercomputers: We remember Seymour Cray

Virtual Memory

Memory is like an orgasm. It's a lot better if you don't have to fake it.

He was just annoyed that ATLAS was delivered before the CDC 6600

How to build a server room: Back to basics

managed PDU's

Metered and managed PDUs will save you bacon one day. Buy them.

And if you buy APC be prepared for them to die after 2-3 years.

And the bloody plugs will fall out if you breath on them.

Global warming stopped in 1998? No it didn't. If you say that, you're going to prison

Re: Oh no not again...

All the models assume solar output is constant. It isn't. And it despite denials from the Warm Mongers, it's variance matches well with the variance in temperature.

Does it bollocks.

Any time some clown makes the argument "scientists haven't thought of this" you know he's a D-K sufferer.

Re: I hope they are ready...

I don't think that our ability to predict the movement of the sun is based on modelling,

Uh, yes it is. There was this guy called Isaac Newton, you may of heard of him, he made a model, we call it "the laws of motion". Some other bugger called Albert Einstein fixed a few problems later on and now we use his version of the model if we want better accuracy.

Re: If one is

Don´t know how this figure relates to whatever else releases warmth into the enviroment, but makes me think.

Incredibly tiny.

The sun is giving us around 250 W per M2

certain places I monitor have a rallying cry for the believers to come and make a stand against the article known

Where? The only place I've seen Lewis getting noted was on WUWT.

Re: Oh no not again...

all the climate models discount the sun as having any influence on the temperature of the earth

What on earth are you going on about? The sun is where the heat comes from, everyone knows that, the models are based on that.

The problem is that the sun is not providing more heat, so it can't be the cause of the warming.

Vanished global warming may not return – UK Met Office

Re: Spin your first sentence, as usual

The conclusion of the paper states that the bad sites do indeed skew the results high by a "significant" amount.

No, it doesn't.

If it does, please quote the part that says that.

Re: Article based on report from a government agency

All the raw data I've seen (far from a complete set, but comprising of some very important factors) indicates a strong possibility that we're in for cooler weather over the next few years.

What do you mean by "we"? If you're talking about the UK, then as the Met office point out you might be right, if you're talking about global average temperatures over a reasonable period then not a fucking chance.

Re: Spin your first sentence, as usual

Fall, Souleymane; Watts, Anthony; Nielsen-Gammon, John; Jones, Evan; Niyogi, Dev; Christy, John R.; Pielke, Sr., Roger A. (2011). "Analysis of the impacts of station exposure on the U.S. Historical Climatology Network temperatures and temperature trends" (PDF). Journal of Geophysical Research 116 (D14120). Bibcode:2011JGRD..11614120F. doi:10.1029/2010JD015146.

From the summary:

Temperature trend estimates vary according to site classification, with poor siting leading to an overestimate of minimum temperature trends and an underestimate of maximum temperature trends, resulting in particular in a substantial difference in estimates of the diurnal temperature range trends. The opposite ‐ signed differences of maximum and minimum temperature trends are similar in magnitude, so that the overall mean temperature trends are nearly identical across site classifications.

(My emphasis).

NOAA produced the same result by throwing out all the stations that Anthony said were "bad" and finding that it made little difference to the trend (in fact they found that the "bad" stations were slightly lowering the temperature trend).

As to why you see no mention of that on Anthony's site, I can't imagine.

Re: record temperatures every year, who said it stopped and why the biased article title?

Why are we concerned?

Do you live in the lower troposphere? I don't.

Do you "measure" temperature by examining microwave radiation from a satellite then running huge spaghetti FORTRAN programs to calculate a number? I tend to use a thermometer.

Do you not know that we've already passed one degree, so it wouldn't take 200 years even if you assume the dodgy satellite data is the true picture.

Re: Spin your first sentence, as usual

Of course otherwise you could do some science and check whether urban heat islands were distorting measured trends - but science is soooo boring.

It's ok, some guy called Anthony Watts did the science.

He managed to prove that the UHI effect made no difference to warming measurements at all, not quite the result he was expecting.

Re: record temperatures every year, who said it stopped and why the biased article title?

And the satellite temperature records continue to show no warming throughout their entire record.

Why bother posting rubbish like this? It's so easy for people to check it out.

RSS: Trend: 0.122 ±0.067 °C/decade (2σ)

UAH: Trend: 0.142 ±0.068 °C/decade (2σ)

Why are you so interested in a dodgy calculated proxy for a temperature in the lower troposphere anyway? I happen to live on the surface of the earth, not in the troposphere.

Ich nicht bin Charlie: Facebook must crack down on racists, says Germany's Merkel

So people who have risked their lives to escape ISIS have "radical islamist tendancies"?

Clown.

I suspect your disk is broken.

Re: The 90's

I do have a fear that the high influx of Syrians will change the feeling of Germany...

A high influx of people who have risked their lives to reject authoritarian dictatorship and obscurantist murdering slave takers could indeed change the feeling of any country -- for the better.

Re: freedom of speech

Yes most countries don't have freedom of speech as the basis on their whole constitution like the US

Free speech is obviously not "the basis of the whole [US] constitution" as it was added as an amendment after the fact.

In the German constitution human rights are defined by the first articles of the constitution -- they are the basis of the whole constitution. Freedom of expression is article 5.

Wikipedia’s biggest scandal: Industrial-scale blackmail

Re: I think Captain Scarlet would go under

"The Captain Scarlet page was created on 8 January 2003."

And Wikipedia existed long before then.

But it didn't, did it.

Wikipedia was created in 2001.

So you're all bent out of shape because wikipedia didn't have a Captain Scarlet page for two whole years.

WTF?

Re: I think Captain Scarlet would go under

Try reading what I posted again; and bear in mind that Wikipedia is being constantly edited. Just because the information is there now, that doesn't in any way "disprove" what I wrote.

The Captain Scarlet page was created on 8 January 2003.

What are you wittering on about?

Debian upgrades Wheezy and Jessie with a combined 372 updates

Who's advice?

You might like to do the update one version at a time if you were paranoid (I am) but a reinstall? Why?

Re: Wheezy remains in production, as system-d still not fully stable compounds this

Well, no.

AFAIK (please correct me if I'm wrong) syslogd is started and stopped at more or less the same time with sysvinit and systemd.

However with systemd messages are logged to journald, not syslogd. journald will pass them on to syslogd when syslogd starts up, so more messages are logged at startup.

Re: Bloody systemd!

Well a lot of us are waiting for the "promised"* SystemD free next version after Jessie. Lets just say I'm not holding my breath.

Are you talking about Devuan? Why bother waiting, just install sysvinit on Jessie instead of systemd.

Re: Bloody systemd!

/lib/systemd/system, add "After=nfs-common.service" to remote-fs-pre.target and all should then be well.

Shouldn't you be pitting a new file in /etc/systemd/system instead of modifying the existing one in /lib/systemd/system?