* Posts by Paul Barnfather

40 publicly visible posts • joined 29 Feb 2008

Men in Green step back from GM's 230mpg Volt claim

Paul Barnfather


"...unlike in your car where you'll get almost all of it"

Umm, nope. According to the US DoE you get "about 15%".


Perhaps you've forgotten how inefficient modern cars really are? A good engine only manages around 30%. That's not "almost all".

Compare that with a battery car:

Fossil fuel power stations are 36-40% efficient. Power lines are about 90% efficient. Factor in battery charging (97%) and electric motor (92%) and you get 29-32%. No brilliant either; about the same as a fossil fuel powered in fact.

Thing is, we already know we can get vast amounts of electricity from nuclear & wind. So we might as well see if we can use some of it to power our cars...

Paul Barnfather


Wind resistance is taken into account using a coast down test:


It still doesn't help the published figures get anywhere near real-world driving.

Unfortunately there are many {ahem} tweaks that manufacturers can make to influence the results of the coast down test and hence get much better mpg/CO2 figures...

Please kill this cookie monster to save Europe's websites

Paul Barnfather

Not just cookies

I don't like being tracked without permission. Sadly, the trackers seem to feel thay have a right to track me and have moved on from relying on normal (or 3rd party) cookies.

Have you checked your Flash cookies lately? Do YOU know what your browser's showing the world via GlobalStotage()?

It's turing into a bit of an arms race. Firefox users can regain some control with a combination of the NoScript, RequestPolicy and BetterPrivacy add-ons. But what about IE, Chrome and other browsers? There's no easy way to defend against this cr@p. Publishers certainly aren't interested in asking permission.

I'm no fan of legislation, but has anyone got a better idea?

UK's national grid 'ready' for e-car expansion

Paul Barnfather

National generating capacity

Demand goes pretty low during the night (30GW min), so I can understand that overall it's possible that "no new power stations would be needed".

What I don't quite understand is the claim that it's still OK even with "uncontrolled charging". Presumably, most people will come home from work at more or less the same time, plug the car in and pop the kettle on. When is peak demand? Usually early evening.

Small (overnight) EV chargers apparently pull ~3kW. 3 million of these means up to 9GW extra peak load - right when you don't want it...

Paul Barnfather

Hmmm indeed....

Even if the national grid is up to the job (it's certainly possible), the "research team emphasise that local improvements may nevertheless be necessary", according to the press release.

I'd like to see the costs for that.

That's potentially a lot of new substations, new cables to be laid, new electrical installations in each house (if you want anything faster than an overnight charge). At a modest £500/house the additional infrastructure cost comes in at well over £1billion...

Firefox users caught in crossfire of warring add-ons

Paul Barnfather

@Ian McNee

Whoa, chill out man.

NoScript and AdBlock are pretty wonderful. A little criticism and/or competition isn't going to do them any harm.

My problem with NoScript is that once I've enabled scripts on a page, I don't seem to have much control over what they actually do (and what sites they connect to). RequestPolicy gives me some of that control, which is the reason it is worth a mention. It also works fine with NoScript, if that's important to you.

Paul Barnfather

A solution

Uninstall both, try RequestPolicy instead.

It won't protect you against malicious scripts on a website, but it blocks ads and 3rd party scripts.

Better security and far less annoying than NoScript.

Disclosure to private eyes sometimes legal, says privacy watchdog

Paul Barnfather


"The ICO said that it was a principle of the DPA that people are told when their information is gathered how or when it will be disclosed."

So, why was no action taken against BT when it handed over the private communications data of thousands of individuals to Phorm without consent?

UK.gov to spend £2bn on ISP tracking

Paul Barnfather


Thanks for the replies, all.

I totally agree that there are ways around this - either for the security services or those that are being watched.

I think the main points are:

1) It'll cost a fortune - and we know who'll be footing the bill

2) It won't achieve what the government think it will achieve: an easily accessible repository that lists who talks to whom

3) The very existence of such a repository is probably illegal anyway...

Paul Barnfather

What's the point?

According to ol' Jackboots, all they want to know is "who is talking to who, not what they're saying". And the solution to this is mandatory DPI in every ISP.

Apart from creating an operational and privacy nightmare for ISPs that we, The Governed, end up paying for, this appears to acheive sweet F.A.

Please correct me if I'm wrong, but:

- the ISP still don't know who I'm emailing (I use Gmail over HTTPS);

- they still don't know who I talk to (Skype is P2P and encrypted).

Oh well. At least they can drive the economy out of recession through the Magic Power Of Targeted Advertising {sigh...}

Phorm director advises UK.gov broadband minister

Paul Barnfather


Blimey, isn't that a cool £500k right now? And they claim there's no conflict of interest!!

I'd say Kip Meek has absolutely no place whatsoever advising anyone on Digital Britain.

Isn't there some kind of law against this behaviour, or are we just seeing a small glimpse of what they're all up to?

Sod the conspiracy theories about Jackboot Jacqui, this lot are just trying to line their own pockets. Greed, pure and simple.

Hybrid Jaguar XJ en-route?

Paul Barnfather

@Frank Bough

If the Prius does it for you, then that's nice. But please don't think they're a solution for everyone.

Last time I drove one I found it painfully slow, poor handling and disappointingly thirsty (i.e. sub 50mpg). In short, I hated the damn thing.

For my usual driving (which doesn't involve sitting nose-to-tail in endless traffic jams), existing diesels do the job cheaper, quicker and with less effort.

Having said that, I rarely drive > 250 miles, so put me down for a lottery win and a Tesla! :-)

Paul Barnfather
Thumb Up

Re: They fail to explain one thing

Yes, you are right. If your electric car is charged from fossil fuels, there's no net gain (apart from a reduction in urban pollution and noise).

The reason they're pushing plug-in EVs is that they know how to generate lots of electricity from low-CO2 sources: wind, wave, nuclear, solar, etc. They don't know how to do the same for petrol or diesel. The only other low-CO2 option (biofuels) didn't work and the politicians got their hands burned, so using electricity to power cars is seen as the Next Big Thing.

Of course, there's always some exciting new technology on the horizon (synthetic fuels, cheap hydrogen, etc), but for now no one's really come up with a better idea. Heck, most cars only last a maximum of 10 years anyway so none of this is permanent...

Brown pledges to make Britain's drivers greener

Paul Barnfather
Thumb Up

Re: A big wind turbine for every 4 square km?

Your power requirement looks somewhere near. Total UK transport energy consumption in 2007 = 60 million Toe (http://www.berr.gov.uk/energy/statistics/publications/ecuk/page17658.html)

1 Toe = 42GJ, so that works out at a transport total of 80GW energy from fuel.

Only just over 1% is currently electric (rail), so ignore that.

Assume fossil fuel vehicles ~30% efficient (fuel -> motion). So there's 24GW of "motion energy" needed. I have no idea what real-world efficiency of EVs is - no combustion losses, so guess at 80% (generated electricity -> motion).

So to power the UK's entire transport from electricity would need another 30GW.

Paul Barnfather

Re: Electricity - the "green" energy

If you're talking about the environmental (i.e. CO2) impact of driving a fossil fuel car vs. driving an electric car then the figures come out at...{tada.wav}...guess what?

"About the same."

It's no big deal either way, given the current mix of electricity generation in this country.

The mysterious *attraction* of EVs to the policy wonks is that, once you've got people used to driving different cars, they can generate the 'leccy from nuclear, wind and CCS coal. All of which substantially reduces the CO2.

It's kinda hard to make that kind of change if we're all still running around on diesel. Biofuel was the last Great White Hope and look where that got us...

Phorm moves beyond privacy - except when slating rivals

Paul Barnfather

I don't get it

If I'm an online publisher or trader, how does scraping content from my site help me? All that will happen is that traffic will be driven away from me and towards Phorm affiliate sites.

I lose out - unless I sign up with Phorm.

And if everybody does that, we're no further forward. There's no net gain, the web is not "saved". The only thing that happens is everyone is a little poorer and Kent becomes very rich.

Oh, wait...

Microsoft cries netbook victory against Linux

Paul Barnfather

How much...

...is Microsoft charging for an OEM XP license on a netbook these days? Is it by any chance less than the usual £50?

Methinks that this sudden switch to XP didn't come about because "customers want it". I'm sure there was panic in Redmond when they realised that 90% of these things were shipping with Linux...

A solution? Why not use your OS monopoly to cross subsidise the XP license down to a couple of dollars? Easy!

Though of course that would be illegal so Microsoft wouldn't do that.

Tesla unwraps Model S

Paul Barnfather
Thumb Up


That is all.

North East, Nissan to explore e-car promo plan

Paul Barnfather

@Gary F

Where on earth did you get your figures from?

I don't have absolute figures to hand, but just think about what you're claiming for a minute.

If I put fuel in a car & burn it, 30%(-ish) of the energy produced is turned into useful forward motion.

If I put fuel in a power station, I get around (say) 35-40% out in the form of electricity. That's only to be expected - these things run mostly at constant output, so efficiency gains are possible. If I want to power a car with this, I might lose 10% of that energy in transporting it over the wires and charging the battery. Maybe another 10% lost in the motor if it's really inefficient. So I'm back down to around the 30% mark.

Guess what? Electric vehicles usually work out *about the same* as internal combustion vehicles in terms of energy efficiency. That's hardly surprising, given that they do more or less the same thing and there's no spectacular waste of energy anywhere along the way.

While it's certainly true that a wholesale switch to EVs with the current mix of generation won't save the planet, I can't see how your figures could possibly make sense.

Supercar maker pitches 'leccy sportster with stunning spec

Paul Barnfather
Thumb Up

Flow Batteries?

Could be interesting: put new electrolyte in and - hey presto - your battery is charged!

i.e. just like putting petrol in ('cept you also drain the old "fuel" out)

When I last checked, these things had quite poor energy density, but maybe things have moved on? On the plus side, you could store the liquid (almost) anywhere in the car, which is probably easier than cramming every available space with Li-Ions...

Obligatory link:


Watchdog mauls billboard sex ads

Paul Barnfather
Thumb Up


The last thing we want is medicines being advertised. These adverts are legal in Australia and these huge "Want longer lasting sex" adverts are *everywhere*, almost to the exclusion of all other advertising. Whatever your views on the morals of this it just *feels* wrong to me somehow.

Meanwhile in Germany, it's now legal to advertise medicine on the telly. Result: during a Casualty-style drama show you get 10 mins of back-to-back drug adverts. Which makes the show pretty much unwatchable. Presumably the drug companies have money to burn when it comes to pushing their wares...

Online campaign urges MPs to block secret expenses

Paul Barnfather
Thumb Up

Done - letter sent.

So, after months of argument and having the proposals comprehensively rejected, our Dear Leaders just decide to change the rules anyway. With a few days notice, of course - gotta have some time for "consultation"!

What *is* it with these people?? Can you imagine the rest of us submitting a huge expenses claim with just the final figure and a note saying "I'm sorry, I can't tell you what I spent the money on. It's a secret."

Thanks for alerting readers to this.

Century-old hydropower plant to run on fudge

Paul Barnfather
Thumb Down

Oh, come off it!

The typical electricity consumption for a UK household is around 3,000 - 5,500kWh per year (ranging from single person at the low end up to a family house).

This hydro plant produces 510,000kWh each year.

So yes, it's enough to "power over 100 homes [with electricity]" .

I'm all for criticism of unnecessary greenwash. But what is the problem here, exactly?

Europe's Tesla will be first with full performance

Paul Barnfather


No, I'm following Ian Johnstone's suggested calculation.

- You've multiplied that the wrong way (as indicated by the ridiculous 450 MPG figure) - I calculated the power from burning fuel @30MPG, then multiplied that by 27% to get the power at the wheels.

- Yes, I *know* it's not 100% efficient through the drivetrain (it's more like 90%), but this assumption is enough to show that the "1c/mile" figure quoted by Tesla isn't that wide of the mark.

Yes, it assumes you can charge it on cheap electric.

Yes, it's a bit optimistic.

But it's not wrong, and it's not a lie.

The overall energy used by an electric car is *close* to that used by a fossil fuel car. Of course it is - there's not massive wasteage at any point so that's what you'd expect. Conventional cars and electric cars are both pretty good solutions.

Claiming that one is *magnitudes* better than the other is not going to fly. It's just not true. [That goes for the eco-weenies as well as the petrol-heads.]

Paul Barnfather

Re: It doesn't add up

OK, I'll bite:

[FYI, Tesla quote "around 1c/mile", assuming off-peak electricity]

1) Off-peak electricity is arounf 2.7c/kWh in the states, so 1c buys 0.37kWh

2) 0.37 kWh = 1,332,000 J/mile

3) Let's take a petrol car:

30 MPG is 9.4l/100km

1 litre of petrol contains 10 kWh per litre

So we're using 94 kWh/100km

Or 1.5 kWh/mile

Or 5,400,000 J/mile

Conversion efficiency is around 27%, so at the wheels we need

0.27 x 5,400,000 = 1,458,000 J/mile

4) Scaling that back up to get a "MPG equivalent", that makes

1,458,000/1,332,000 * 30 = 33 MPG

Sounds perfectly reasonable for a little 2-seater car.

How on earth do you get to 160 MPG??

Paul Barnfather
Thumb Up

@Stephen Gray

"Why do we need electric cars?"

1. Big combined-cycle power generators are pretty damned efficient (50%+) because they run at more or less constant load. Even taking into account the transmission losses, it's still better that achieved by a fossil-fuelled car (<30%).

2. Can you run your car on nuclear power or solar? Thought not. Using electricity allows you to use a whole variety of raw fuels. There's not enough land available to produce enough alcohol/cooking oil to power all the cars.

That's why.

Research: Wind power pricier, emits more CO2 than thought

Paul Barnfather


"is very expensive electricity as the energy used to pump the water to the upper lake is greater than is gained from the generators"

You're thinking about it from the fossil fuel point of view. Fossil fuel power costs money to make, but you can have it whenever you want it.

Wind power costs "nothing" to make, but you get it regardless of when you want it.

It makes sense to store it when you don't need it ready for when you do. The fact that some of the energy is lost during the storage/recovery process is irrelevant, because (provided you didn't need the electricity for something else) you might as well store it.

It's the same for nuclear. Nukes are happiest producing power 24/7, whether you want it or not. So when demand is low, you store it. That is why we built Dinorwig. And it's why we'll continue need to build expensive complexities into the system - whether we have nuclear or wind or both.

It's just physics, man.

Paul Barnfather

Poor article

The assumption that wind power is a substitute for peak gas plants is a pretty basic error. The whole market/pricing for renewable power is completely different:

Fossil fuel = cost per megawatt-hour. Plant is relatively cheap. But you pay for the fuel

Renewable = cost per megawatt. Once you've built it, the energy is effectively "free"

At the moment, electricity is more or less the same price regardless of when you want it (Economy 7 excepted), mostly because we get most of it from fossil fuel. If we get a lot of electricity from renewables, this changes. Want power when it's windy? Heck, have all you want. Want it on a calm day? Sorry, gonna cost ya.

Over time, you'll see new & interesting uses for the "spare" power e.g. charging plug-in hybrid cars, together with intelligent appliances that modify their behaviour to minimse your electricity bill e.g. fridges.

As fossil fuel prices go up (hello, Gordon), renewables become more attractive - despite all the drawbacks. Nothing is going to get solved overnight, which is why some currently "uneconomic" technologies are being pursued anyway. This is just sensible planning - it's not rocket science!

PS @david: PV cells can last from around 10 years (early stuff) to at least 30 years (current thin-film). They don't stop working at this point, but they do produce less power (~80% of the original output). Other things to consider: there's a tiny amount of cadmium in them, so you ought to recycle them when you do finally chuck them away. Even if you don't it's a lot better than burning coal for electricity and putting the cadmium up the chimney (~10x less). Greenhouse gas emissions are around 30g/kWh (current tech), vs. 10g/kWh (nuclear), 400g/kWh (combined cycle gas), 900+ g/kWh (coal). Solar cells don't appear to be very efficient (10-20%), but this is approx 10x better than sun -> biomass -> fuel -> electricity. So, PV is a long way from perfect but probably a useful improvement on some of the ways we currently produce power.

Ex-Logica boss to teach UK.gov how to identify crap IT

Paul Barnfather
Thumb Up

What actually works

Maybe easier/quicker to look at what government IT isn't crap.

My vote is for http://www.transportdirect.info

It seems to do exactly what it was designed to do. Heck knows how they managed that. Any more out there?

Phorm launches data pimping fight back

Paul Barnfather
Thumb Up

TalkTalk may have a solution

As well as confirming it'll be opt-in, they seem to have realised the weakness of Phorm's cookie-based opt out:

"We had a meeting yesterday and based on customer opinion we decided to use a different method, yet to be decided, to split the traffic so it doesn't hit a WebWise server at all for those that opt out."

Excellent news - and kudos to TalkTalk for listening...

Paul Barnfather

@ Lyndon Hills

If you look at the contract with your ISP, you'll find they have every right to do deep packet inspection. No problem with that.

You'll also (probably) find that they promise to keep this data private and never sell it on to a third party.

Now we see ISPs are planning to sell this data to a third party. Even if you're happy to have your data sold on, surely you see the problem here?

Paul Barnfather

Alarm bells...

Full webpage data *apparently is* stored, despite the denials above.

From last night's Webwise chat with Kent:

[Archived here: http://www.badphorm.co.uk/page.php?10]

"MBurgess: Pages are not tagged (or modified), and the keyword analysis process is offline so it can't affect response times.

narcosis: If the keyword analysis process is offline then in order to scan for keywords would you not have to have a copy of webpage in order to analyze it offline ?

MBurgess: Yes, a mirrored copy is analyzed."

How long is the mirrored copy kept? How is it deleted? This isn't just headers - this is a full grab of all web traffic *before* it's profiled and categorised.

BT targets 10,000 data pimping guinea pigs

Paul Barnfather
Thumb Up

Mainstream media getting the message?

The Guardian have a good summary up here (and they've spoken to an ex-Phorm employee):


Nice work, guys - keep it coming!

The Phorm files

Paul Barnfather
Black Helicopters

Statement from TalkTalk

After at first denying any partnership with Phorm, TalkTalk have replied to my complaint, stating:

"I can advise as previously stated PHORM are unable to access any

personal information without your permission. The service they are

offering is called Webwise, although they are able to view your browsing

history, they are unable to recognise who you are through this


Can anybody help with a suitable rebuttal of this nonsense? It's like doublespeak. Surely anybody could be easily identified from their browsing history? What makes Phorm "unable" to the same?

Broadband big boys waiting on data pimping

Paul Barnfather

TalkTalk denies any partnership with Phom!

Just got this back from customer services.

I quote:


Thank you for your e-mail regarding a possible new partnership between

TalkTalk and Phorm.

I can confirm that TalkTalk is not in any partnership with this company

and all data from our customers accounts is kept confidential and is not

shared with any other company.


Interesting, huh?

Paul Barnfather

@ Mark

Yes, having read El Reg's latest analysis you're right - there's no ad "injection" going on here (other than the Phorm cookie, which *is* injected). They seem to be replacing generic OIX ads with targeted ones. So if that is true, it's apparently no worse than DoubleClick.

BUT they are definitely passing your (private) web traffic onto a third party for "processing". Even if they claim to delete this stuff straight away, this is an absolute no-no as far as the Data Protection Act is concerned, and almost certainly violates the ISPs own privacy policy (until they change it, that is...). Even so, you simply cannot pass on copies of private data to a third party for any kind of processing without consent of the individual concerned.

Paul Barnfather

@Mark "There is no injecting going on here"

The Phom website implies otherwise:

"The OIX uses data from ISP pipes to upgrade the generic advertising on websites with more relevant ads. These ads will be viewed by that ISP's subscribers who are most likely to be looking for the advertised product or service based on keyword patterns in their browsing behavior. "

Paul Barnfather
Thumb Up

@ Jon

Yes, an https proxy should should do it.

According to Phorm, "[the] technology does not view any information on secure (HTTPS) pages"

More here: http://www.phorm.com/about/faq.php

Paul Barnfather

@ Mark

Not quite. When you go to Tesco, the data stays with Tesco and their privacy policy (presumably) does not allow them to sell data on your shopping habits to third parties. Plus, the Data Protection Act requires them to take care of this data.

In this case, the ISPs are unilaterally allowing a third party to come in and snoop your surfing habits and sell this private data to who knows. In the contract with your ISP, I would be very surprised if you agreed to give them permission to sell/pass on your surfing habits to third parties.

That is the problem here: they are doing something with YOUR private data without your permission. They are specifically not allowed to do this by law.

Paul Barnfather

Here's how it works

(I think, having read all the blurb on the Phorm site...)

1. All your browsing (URLs, content but not https) is routed via Phorm's server. Presumably due to bandwidth requirements it's physically there in the ISP datacentre (like a wiretap).

2. Phorm's server sends a unique cookie to your browser, analyses your web traffic and "categorises" that browser in real-time based on your surfing habits (actually 3x an hour, maybe). It then associates your browser ID with your assigned "category".

3. The OIX database (ad server?) then watches for that cookie ID and injects "relevant" ads into your http data stream based on your current category - either replacing existing ads or even inserting new ones (not entirely clear on this). It may also pass on your cookie ID and category to third parties to that they can do the advertising directly.

So no, forget any swanky DNS tricks or ad blocking. What the ISPs appear to be doing is allowing a trusted (hah!) third party access to all your surfing content. Phorm claim this "cannot be used to identify you because it's anonymous and we ignore phone numbers and email addresses and IP addresses". Bo||ox. Just one snapshot of a non-https gmail session is enough to identify anyone...

To me this appears to be in blatant contravention of TalkTalk's privacy policy, which states that they will not disclose personally identifiable information to thrid parties except in a very specific set of circumstances.

PS: The service does not appear to be live yet on TalkTalk (you can for yourself by going here: http://www.webwise.com/privacy/can-choose-NA.html). You might want to keep checking, and be sure to let your ISP know what you think about this.