Forgot your password?
typodupeerror

Comment: Re:Real-time market approach (Score 1) 485

by DamonHD (#48367833) Attached to: Denmark Faces a Tricky Transition To 100 Percent Renewable Energy

People already DO react to time of use pricing and it is in use in many places round the world at retail level and most places at commercial and industrial level.

One thing to note at the retail level is that typically not even the majority of the retail cost of a unit of electricity is the energy itself; infrastructure costs and so on are folded in too, so the price swings are a lot less dramatic than at the wholesale (or balancing) level which can range from 3:1 to 10:1 in the GB market for example.

The GB's best-known ToU retail tariff (sadly we don't have a national real-time one) is Economy 7, designed originally to soak up power at night from nukes that couldn't be turned down. (It's not only renewables that have stubborn timing problems.) I think the unit price ratio is about 3:1 between day and night. One third of domestic electricity use is on that or closely related tariffs, or 10% of all GB electricity consumption roughly. (The definitive source of this data is DUKES https://www.gov.uk/government/... .)

Rgds

Damon

Comment: Re:Home storage (Score 1) 485

by DamonHD (#48367747) Attached to: Denmark Faces a Tricky Transition To 100 Percent Renewable Energy

Average UK household use when I last looked ~3,300kWh/y.

Ours, ignoring PV, 1,500kWh/y (family of 4, end of terrace house).

I see a (cheap) solution...

However, we do generate a mean of ~10kWh/d from our PV also, down to ~1kWh/d in the depths of winter. A 4kWh battery would mean that we would not have to import from the electricity grid at all for 9 months of the year, only export. A complete inverter+battery system to cover us would currently cost about GBP8k. If we switched from natural gas heating to heat-pump (doubling our annual electricity demand, primarily in winter) that would still be about 6 months. We'd need a bigger inverter though.

Rgds

Damon

Comment: Re:Nice and all (Score 1) 107

by DamonHD (#48354413) Attached to: Eben Upton Explains the Raspberry Pi Model A+'s Redesign

I have my RPi B+ running all my Internet facing services and running off-grid in gloomy London at under 2W, nearer 1.5W when I can fix some transient issues.

http://www.earth.org.uk/off-gr...

And if I need something lower power I have Arduino-like boards that I run on microwatts, eg for battery-powered remote sensing.

http://www.earth.org.uk/out/ho...

I see all the complaining about memory and speed but as someone who grew up with a Z80- and 6502- based set of home computers, then used Sun Workstations with a few MB of RAM and tens of MHz clock, the Arduino/ATMega328P matches the Z80/6502s in performance at a millionth the power and has non-volatile storage built in, and the RPis are and order or two of magnitude better than the workstations and a couple of orders of magnitude cheaper.

I can fit almost all that I need to run with careful use of resources into the RPi/Arduino world and still have lots of elbowroom left over.

Yes, I've worked on big systems for (eg) big banks, but most of us really need GHz and GB on every machine? Some yes (like my MacBook), but most no.

Rgds

Damon

Comment: Re:DMARC and Google: multiple foobars (Score 1) 139

by DamonHD (#48299531) Attached to: Ask Slashdot: How Useful Are DMARC and DKIM?

That would be a very good step and eliminate most of the issue with Google, I agree.

However, G also insists on, for example, sending out G Calendar notifications forged with my sign-in email, which is SPF/DMARC protected, and which other systems (and gmail) thus entirely correctly often reject.

But I can't get anyone at G to even acknowledge the issue. (Would be nice if a Googler was reading this and would pass it on.)

And, BTW, G also seems to ignore email other than from the a/c's login address, eg for AdSense/AdWords, which is a Catch-22, and one small reason why I am using them far less.

Rgds

Damon

Comment: DMARC and Google: multiple foobars (Score 2) 139

by DamonHD (#48294679) Attached to: Ask Slashdot: How Useful Are DMARC and DKIM?

DMARC would work a lot better if Google for one didn't wrongly try to internally forward as-is *and then bounce* email from DMARC-controlled domains, thus making it impossible for example to get through for many support queries, and causing spurious problems with (say) Google Calendar when the account ID is in a DMARC-controlled domain.

Left hand vs right hand Google? You guys are meant to be smart!

That and randomly chucking email from DMARC-controlled domains in SPAM folders...

Rgds

Damon

Comment: Re:Redistribution (Score 1) 739

by DamonHD (#48281145) Attached to: Statisticians Study Who Was Helped Most By Obamacare

Hmm, I hear what you say, but it wasn't generally stand-alone IT that would be much use outside finance, and I was the CTO of the virtual credit card start-up.

But I assure you that there is no gold-filled vault, no not even slightly full of warm lovely glowing metal; nothing to see here, move along.

Rgds

Damon

Comment: Re:Redistribution (Score 1) 739

by DamonHD (#48280765) Attached to: Statisticians Study Who Was Helped Most By Obamacare

It's called "contracting" or "consulting"; I have spent the best part of 20 years working in IT in the City of London, none of it permanent.

I have also founded a retail finance start-up. I'm not sure if start-ups count as freelancing; there's matching insecurity and worry and no steady income!

Rgds

Damon

Comment: Re:You shouldn't need insurance for most things (Score 1) 739

by DamonHD (#48276727) Attached to: Statisticians Study Who Was Helped Most By Obamacare

For most things maybe you just shouldn't be paying, at all denying treatment to those who happen not to have spare cash, eg students and young adults in general getting going in their jobs. I couldn't have paid anything much as a teenager for my epilepsy diagnosis and treatment; should I have just rotted before I even got to uni? I had left home, BTW.

(In the UK I do pay for a few things at point of use under the NHS, but often even then fairly small fixed/tiered charges.)

Rgds

Damon

Comment: Re:Just don't try to write an OS in Java (Score 1) 511

by DamonHD (#47744815) Attached to: If Java Wasn't Cool 10 Years Ago, What About Now?

Hi,

That is a different and also interesting case, and just by bringing it up you'd pass my test.

No, I meant something like:

public static final String CRITICAL_ID = "whatever"

in a secure API. If I use char[] instead what happens if a miscreant overwrites the content of the array; what BadThings might happen?

Rgds

Damon

Comment: Re:Performance improvements have helped it survive (Score 2) 511

by DamonHD (#47744793) Attached to: If Java Wasn't Cool 10 Years Ago, What About Now?

Except that once JIT/Hotspot is involved all or most of the executions *are* of compiled machine code, optimised to the particular CPU on hand and the particular data set for this particular job. So things like dynamic inlining *for this particular job* can allow the JVM to produce *better* machine code which will execute more quickly than statically-compiled code.

The translation takes time, but for long-running tasks that may well be easily amortised away.

So single-threaded Java can beat single-threaded C++.

But I can also bring more CPUs to bear on the code safely with Java for a given level of code complexity (well, now C++ finally has some sensible volatile semantics, that's a little less true).

And there are other factors such as the generally forced synchronous nature of C++ heap handling which can work against it.

Rgds

Damon

Maternity pay? Now every Tom, Dick and Harry will get pregnant. -- Malcolm Smith

Working...