Slashdot Mirror

http://www-uxsup.csx.cam.ac.uk/

Most university computer lab machines dual-boot Windows XP and customised openSUSE. All students get ssh shell access to multiple Linux servers in various departments and public webspace. For Maths, Physics, and CompSci courses most of the programming is strongly recommended to be done on Linux machines, and the university gives free classes to staff and students on everything from basic Python to LaTeX.

Well, if you buy the hardware today, you are locked-out from future advances, and you need to recover that sunken cost. Sure, someday we might harness the energy of flying pigs, and at that point we can have a chat on the economic viability of pigovoltaics. It's in no way foolhardy to pass on investing in something that's a money looser now and might never become viable, especially when we are running pretty close to a fundamental physical limit of pig buoyancy.

Please stop propagating lies.

It is not the same with the Prius. At least not as far as energy consumption is concerned.

In case of a car the energy consumption in manufacturing is on average an order of magnitude smaller than the energy consumption during its use. We are talking 10% of total consumption vs more than 80%. You can refer to page 10 of these notes (pdf) to see the figures for an average family car.

In case of assessing energy impact of various stages of product manufacturing common sense will never help you. You just have to do the calculations.

As the subject states, this is the initial roll-out of "Trusted Computing".

http://www.cl.cam.ac.uk/~rja14/tcpa-faq.html

It's simply had the political name-change game played with it. If at first it doesn't fly, rename it and try again. Rinse & repeat until passage/implementation/adoption.

As has been commented elsewhere here, this is something that will be slowly expanded with government and corporate pressure from initially being voluntary to eventually become necessary to connect to your ISP.

This thing is a tyrant's dream. It *must* be prevented from being implemented. There's no way that much power will escape being abused and corrupted.

Strat

Probably varied between colleges, but a quick search turns up a document indicating that in 1869 St John's College, Cambridge required "[a] certificate of attainments from some M.A. of Oxford or Cambridge, together with a certificate of Baptism".

Right tool, right job, etc.

Wish they would release a new version, but this one is free and works decently http://www.inference.phy.cam.ac.uk/opengazer/

read this:
http://www.inference.phy.cam.ac.uk/sustainable/book/tex/cft.pdf

solar sounds lovely but until we have self replicating machine plating the worlds deserts with free pannels it's going to remain a toy source of power for niche uses and status symbols like those ones you see on peoples roofs in canada.

Cambridge claims it was a murder in this history.
Cambridge clarifies that it was a murder of a woman, as a prelude to a book review

Ok, I'm going to have to say I was mistaken on the crime itself, but not of the effect.

Cambridge claims it was a murder in this history.
Cambridge clarifies that it was a murder of a woman, as a prelude to a book review

Ok, I'm going to have to say I was mistaken on the crime itself, but not of the effect.

Renewable is med and long term much much much cheaper than anything we do and have right now.

Citation required. One with the numbers and sources.

To get the ball rolling he is one that is against this assertion. Sustainable Energy -- without the hot air [PDF warning] or the main website and i think the book in html.

Says the guy with an engineering degree and a MBA?

So you have *at least* these qualification then? Since you have made a statement in the same vein, which according to the book of angel'o'sphere is a minimum requirement for making such a statement.

The first problem with Tor is that its threat model assumes a non-global adversary. If only there were some way for government to eavesdrop on Internet communications.

The second problem is that not even that assumption is sufficient.

We estimated that a car driven 100 km uses about 80 kWh of energy.

Your cited website is entirely bogus for this discussion since it applies only to gasoline powered vehicles.

80 kWh/100km translates to 0.75 mi/kWh. This is absurd for electric vehicles - if a modern EV gets less than 3.0 mi/kWh it's either a pile of shitty engineering or you're driving it on a drag strip. Hell, even on a drag strip you'd probably get better than that.

In reality you're looking at about 20kWh/100km, which is consistent with real-world driving experiences as reported by Nissan Leaf owners. ~120-240kW. That's still way too much power for untrained personnel to be handling, and certainly more than any typical household service could provide... but still, your numbers are way off.
=Smidge=

We estimated that a car driven 100 km uses about 80 kWh of energy.

80kWh / 5-10 minutes ~= 1000-500kW.

Hmm. That's roughly the power draw of a small electric passenger train (e.g. an old subway train).

100 km in 5 minutes implies 1200 kph which is roughly the speed of sound at sealevel.

("km/h" is preferred, "kph" is confusing.)

I was suggesting that in order to drive 100km someone might wait 5-10 minutes for the battery to charge, so the power needed is 500-1000kW (roughly many hundreds of electric room heaters or kettles, or a single old electric subway train -- the latter obviously using special cables etc, and supporting only a few trains at once in a small area).

We estimated that a car driven 100 km uses about 80 kWh of energy.

80kWh / 5-10 minutes ~= 1000-500kW.

Hmm. That's roughly the power draw of a small electric passenger train (e.g. an old subway train).

100 km in 5 minutes implies 1200 kph which is roughly the speed of sound at sealevel.

There have been supersonic cars, and they require well over a mere 1000 horsepower or so.

1000 HP will barely achieve 250 MPH.

We estimated that a car driven 100 km uses about 80 kWh of energy.

80kWh / 5-10 minutes ~= 1000-500kW.

Hmm. That's roughly the power draw of a small electric passenger train (e.g. an old subway train).

Rescaling, the figures become 0.5 to 1.0 MW. That's a highly non-trivial amount of power to transfer electrically (ignoring the massive electromagnetic fields that level of power transfer creates). Not something that's going to be done in the home.

Recall, a consumer-grade hair drier is in the 1.0 to 1.5 kW range. We're talking about operating about a thousand of those at the same time for 5-10 minutes. Personally, I don't want to be anywhere near that. Moreover, even if it's wildly efficient at 99% transfer to the batteries, that's 0.01 x 1 MW = 10 KW of loss that needs to be dissipated. I am not familiar with materials found in the home that can provide safe, reliable, tamper-proof thermal isolation from grasping a cable / connector package that is glowing hot.

We estimated that a car driven 100 km uses about 80 kWh of energy.

80kWh / 5-10 minutes ~= 1000-500kW.

Hmm. That's roughly the power draw of a small electric passenger train (e.g. an old subway train).

Re: Energetics of solar panels. Modern panels "pay" for themselves about 10-15 times over.

See Permanently dispelling a myth of photovoltaics via the adoption of a new net energy indicator

In the case of a building which requires windows I would suggest only counting the solar cell manufacturing cost as the glass and installation cost happen regardless. Unless said building didn't require the windows to be replaced, in which case it's valid to count it.

There is a lot of fundamentals in distributed systems/algorithms/computing.

Agreed. As an undergrad I learnt from her stuff. Someting's wrong with America if concurrent, distributed systems theory isn't considered a legitimate branch of computer science.

But "cloud computing" is something very different, being a marketing term for "stuff being stored and processed somehow somewhere else so I don't have to think much about it". The term is usually used to refer to particular APIs or "web apps" on top of little more than a modern storage VAXcluster or a classical server instance which can be migrated transparently between physical hosts. A true concurrent, distributed transaction processing/operating system (from the PoV of processing and storage) is rarely provided.

And as far as I understand it, it doesn't depend on any established branch of "mathematics".

How are you modelling? How are you proving resilience? How are you measuring performance, asymptotically and theoretically? You require both mathematical maturity and grasp of all the elementary branches of mathematics in order to give yourself the opportunity to fully explore any topic in computer science. I'm not really sure what the counterargument here is beyond, "I want to be lazy and get by with the minimum to pass 'cos some mathematics is hard."

Please don't twist my words. Small incremental changes won't work. Doing a lot of little scales up to be just that -- little. Small reductions do nothing much besides making us feel better. They have no real impact. See Sustainable Energy -- without the hot air for numbers to back it up.

I love that they keep trying to bring this up. It's their Pinky and the Brain-style take over the world plan. The TCPA FAQ, while somewhat old by now, is still relevant (and shows just how long they've been trying this).

Their sample image is here: http://www.cl.cam.ac.uk/~abl26/copy-background.jpg
Their sample image after recompressing is here: http://www.cl.cam.ac.uk/~abl26/copy-background-recomp.jpg
According to them, the sample recompressed image is saved "with particular quality settings".

So I opened the original image file in GIMP, went to save it as a JPEG, made sure the preview was turned on, and saw nothing.
Of course, this is because I save every jpeg at 1x1, floating point, 100 quality.

So I reduced the quality. After a while, the image appears. A change of 1% (from "82" to "83", for example) can render the message completely visible, and another change can render it completely invisible.

I did a low strength blur (imperceptible to my eyes) on the image, and went through the same experiment, and the message was rendered completely gone.
I suspect that same could be done by adding a small amount of noise, etc.

Yawn.

Their sample image is here: http://www.cl.cam.ac.uk/~abl26/copy-background.jpg
Their sample image after recompressing is here: http://www.cl.cam.ac.uk/~abl26/copy-background-recomp.jpg
According to them, the sample recompressed image is saved "with particular quality settings".

So I opened the original image file in GIMP, went to save it as a JPEG, made sure the preview was turned on, and saw nothing.
Of course, this is because I save every jpeg at 1x1, floating point, 100 quality.

So I reduced the quality. After a while, the image appears. A change of 1% (from "82" to "83", for example) can render the message completely visible, and another change can render it completely invisible.

I did a low strength blur (imperceptible to my eyes) on the image, and went through the same experiment, and the message was rendered completely gone.
I suspect that same could be done by adding a small amount of noise, etc.

Yawn.

So I downloaded their test image here: http://www.cl.cam.ac.uk/~abl26/copy-google.jpg that they claim gets a message on it when compressed by google proxy http://www.google.com/gwt/x/i?u=http%3A%2F%2Fwww.cl.cam.ac.uk/~abl26/copy-google.jpg&wsi=223e8e5df695e99c&ei=6ixQTebOCoPoxQW8rYlv&wsc=yq&whp=012e012f72be

But when I take the original and re-save it in Photoshop CS5 I don't see the void lettering. I reduced the JPEG quality and kept trying and at quality 1, the lowest setting I was starting to see a pattern, but no words appeared.

I'd say their idea is nice, but doomed to failure, not least they mention "If you can’t see the message in the recompressed image, make sure your browser is rendering the images without scaling or filtering." which would be the obvious source of attack on such a method should it actually work in practise.

With Paint.NET, it is visible only at quality 60.

The original image can be resized without showing the watermark, see their demo page (and press Ctrl +/- if you browse with Firefox). But so can the re-encoded picture, it shows the watermark only at the zoom level of 100%.

From this I suppose that there is also one zoom level at which the original picture shows the VOID watermark (you better choose it to be an odd value)!

So I downloaded their test image here: http://www.cl.cam.ac.uk/~abl26/copy-google.jpg that they claim gets a message on it when compressed by google proxy http://www.google.com/gwt/x/i?u=http%3A%2F%2Fwww.cl.cam.ac.uk/~abl26/copy-google.jpg&wsi=223e8e5df695e99c&ei=6ixQTebOCoPoxQW8rYlv&wsc=yq&whp=012e012f72be

But when I take the original and re-save it in Photoshop CS5 I don't see the void lettering. I reduced the JPEG quality and kept trying and at quality 1, the lowest setting I was starting to see a pattern, but no words appeared.

I'd say their idea is nice, but doomed to failure, not least they mention "If you can’t see the message in the recompressed image, make sure your browser is rendering the images without scaling or filtering." which would be the obvious source of attack on such a method should it actually work in practise.

From the original paper: "The technique now needs to be extended to handle arbitrary photographs, not just uniform regions."

I do agree that I was ahead of myself. Now for nitpicks:

Aerogel isn't a daydream. You can buy it. Costs reasonable amounts, even. If you wanted to insulate particularly hot CPU/GPU heatsinks in a laptop from the bottom of the case, a few mm of aerogel would be my choice, at a cost of maybe $10 or so. Maybe not in a $200 netbook, but Apple sure could pull that off if they needed to. Hot heatsinks are much easier to cool.

Superconductors are problematic due to rather theoretical reasons, too: good luck when you lose cooling (see what happened to LHC). Supercapacitors and low ESR don't mix. You get one or the other, and I think the reasons are to do with fundamental properties of the class of materials used in their design.

I don't really think that electric hybrids are very good at what they claim to do. Practical -- sure, but not very good at all. Unfortunately, neither are hydraulic hybrids, even 100% efficient ones, or really any other kind of hybrid, and I think that's where the buck stops.

Here's a helpful diagram

and even that insight I think is less penetrating than what one can get from studying abstract information theory and encoding in computer science (where it is quantitative and measurable in addition to providing one with essentially the same insights). I've mentioned it before, I'll mention it again:

http://www.inference.phy.cam.ac.uk/itprnn/book.html

I've got the hardback, but thanks for the link to the online version.

I don't think most of its core conceptual ideas are bullshit per se, or at least they are less so than many things. My primary issues with it are that it became a hammer looking for a nail, going way, way overboard with language and the way it relates to human behavior. And the religion thing. What sort of website tolerates people putting up articles with lines like "This short essay thus honors Korzybski and gives thanks for his work"...?

I actually agree with you (I think), that it works better as the partial basis for an epistemology and not so well as an explanation for human perceptions and filtering (where it is basically more than half bullshit in the precise sense that it is divorced from biological psychology, information theory in computer science, and other sort of hard science that might refute its pretty ideas). If it comes right down to it, Malinowski, Dorothy Lee, Sapir and Whorf, all came up with anthropological of philosophical systems that link language to thought, and much as I love them all -- they were a partial focus of my undergrad major -- they simply aren't as strongly supported by actual unambiguous evidence as one might like. And Wittgenstein has alas always left me cold -- Russell's student he might have been (but then, so was my undergraduate philosophy guru) but he was no Russell.

Ultimately, language and sensory filtering is without doubt important in understanding understanding, but it is far from being the key to understanding it in the sense that the IGS wishes to promote.

Where it is most useful is (again, I think) the insight that it gives one into the whole semiotic/ontology/epistemology chain in logic, math, inference, reason, and even that insight I think is less penetrating than what one can get from studying abstract information theory and encoding in computer science (where it is quantitative and measurable in addition to providing one with essentially the same insights). I've mentioned it before, I'll mention it again:

http://www.inference.phy.cam.ac.uk/itprnn/book.html

is free, online, and amazing. It can change your life (if you can make it through the math:-). Compare this to IGS stuff, which is just plain yucky. It drips.

I'm not sure your Jaynes reservation is apropos. First of all, the ultimate product of Cox or Jaynes is a "plausibility" or "probability". Plausibility or probability of what? That a given proposition is true. In the context of physics and science, that is entirely appropriate. I don't want to know if Newton's law of gravitation is in a more or less accurate correspondence with the observations -- that is the data and it speaks for itself. What I want is to be able to go from the data to some sort of estimate as to how plausible or likely it is that Newton's law of gravitation is true, even though I can casually invent an infinity of ways to explain the data with it or without it and can never be certain that even the most basic premises of my measurement and analysis process are themselves true.

What we thirst for is truth. What we get is a more or less consistent "physical theory", more or less well corroborated by experience and experiment.

Oh, well.

rgb

You message insinuates that the actions of producing a computer chip with some technology is clearly and inexcusably morally wrong.

In this case, that insinuation is considered by many to be correct.

http://www.cl.cam.ac.uk/~rja14/tcpa-faq.html

Once a hugely-powerful system like this is fully-implemented, "stupid DRM tricks" are actually the least worrisome aspect. What government can accomplish in the way of control of everyone's information & digital communications is far more worrisome.

Strat

Yet another reason for the routine use of 95% frequentist statistics to be replaced with bayesian methods.

Frequentist statistics is a reasonable approximation of Bayesian statistics when applied to large numbers. You may well be right above, but my impression is that these sorts of statistical mistakes don't come from inappropriate use of frequentist methods, but rather from conclusions based on poor evidence and a heaping helping of observer bias. Do enough studies and eventually you will see spurious artifacts. If your study is to duplicate someone's study, you may feel pressure (conscious or not) to duplicate the results as well.

I'm with oldbox and others here. It's really time to ditch the null hypothesis and confidence level crap. A geneticist friend of mine once asked me whether he should be using a one-tail or two-tailed test (the test values differ by a factor of two when it comes to working out the significance levels). I'm afraid I didn't know better at the time, so I didn't know what an appropriate Bayesian approach would have given him.

As for doing "enough studies" and picking the ones you like, that's basically fraudulent.

sure, this paper "Optimised to Fail: Card Readers for Online Banking"

Whilst section 2 "protocol description" is fairly good at the logical description of the process - after that they get it wrong; especially the section around the "bit filter" and the way the various card schemes make use of this feature. e.g. they pick up that bank CAP cards have a different bit filter but not "why" and why that makes a card scheme implementation better or worse. Obviously I can't quite remember the exact maths behind the bit masking. it's been a few years.

even after all the work we did, we didn't predict section 4.1, that's something I personally regret that I hadn't designed in better safeguards for the cardholder in a theft situation. I could have used different PINs (it supports multiple PIN) but that might have caused more problems than its worth.

Some parts of the protocol weakness section was off the mark and I think they were using a new CAP card during the tests because it is just wrong.

Now, none of this is their fault because, as far as I know, CAP (Chip Authentication programme) was never publicly released. I have a great deal of respect for LightBlueTouchpaper.

The BBC Newsnight program on the issue (from last February) explains the issue pretty well. Watch it.

The funny/disturbing thing is why did it take 10 months! for some official at the UK banking industry association to have a revelation/panic and issue such a stupid letter. The professor's response to them is pretty effing on!

I think he should've said quite blunty: " listen, our students figured this weakness in your system during their free time, using our shoe string budget". Do you really think high tech criminals and criminal organizations with millions or even more at their disposal won't reproduce this? All you need to do is read the bloody manual! "

If I was a banker/bank/building society I would seriously consider funding research into this instead of whining about it. I mean those students don't have what the criminals can easily get with just money. At least buy them the latest oscilloscope/logic analyser for god sake! - its a miniscule fraction of the profits the banks make - or even what they stand to loose from such weaknesses...

You can kinda sorta be "skeptical" about how negative of an impact that will be

I think this is the real question that is rarely asked. Why is there so much focus on the change and so little investigation of its consequences? Is slow (e.g. over 50 years) environmental change anywhere near as significant as technological, political or economic change? If it happens slowly will it not just result in a change in the countries that produce different types of food? And possibly an increase in the number of dam builders?

Anyone know of any decent analysis? Even some of the more rigorous green advocates resort to rhetoric very quickly on this topic:
Sustainable Energy – without the hot air

Climate modelling is difcult and is dogged by uncertainties. But un-
certainty about exactly how the climate will respond to extra greenhouse
gases is no justication for inaction. If you were riding a fast-moving mo-
torcycle in fog near a cliff-edge, and you didn’t have a good map of the
cliff, would the lack of a map justify not slowing the bike down?

Why do so much analysis and then put in a nonsense scaremongering analogy like that?

This is bullshit. There are 4 causes of climate change - manufacturing, heating/cooling, cars, and planes. As long as people keep flying, driving and buying stuff they don't need we'll keep heading down the abyss. Blaming the oil industry is like blaming Hitler's chef for WWII, they may provide the chemical energy, but we are the ones that fuck up the environment with it. Biofuels won't make a significant difference to the environment - about 6% of our current energy use could be made up with them, according to this expert.

"Sustainable energy - without the hot air", available as a free PDF download.

I haven't read it yet, but I will given his credibility with the article and other book.

http://www.withouthotair.com/ or http://www.inference.phy.cam.ac.uk/withouthotair/

Here is a podcast of a lecture he gave at Cambridge on the topic of sustainable energy.

http://mediaplayer.group.cam.ac.uk/component/option,com_mediadb/task,play/idstr,CU-CSF-Lectures_2008-12_David_MacKay/vv,-1/Itemid,42

"Sustainable energy - without the hot air", available as a free PDF download.

I haven't read it yet, but I will given his credibility with the article and other book.

http://www.withouthotair.com/ or http://www.inference.phy.cam.ac.uk/withouthotair/

Here is a podcast of a lecture he gave at Cambridge on the topic of sustainable energy.

http://mediaplayer.group.cam.ac.uk/component/option,com_mediadb/task,play/idstr,CU-CSF-Lectures_2008-12_David_MacKay/vv,-1/Itemid,42

I just downloaded it.
http://www.inference.phy.cam.ac.uk/mackay/itila/book.html

Have they fixed the idiotic security issue with chip+PIN yet? You know, the one where the chip verifies the PIN? I remember a story where it turns out during PIN verification, the chip sends the reader an "OK" value (0x90, I believe?) if the PIN is OK and the transaction goes through. No, the bank's not checking your PIN at all - it's all done on the card you have. Which means anyone who can clone it doesn't need a PIN.

It is a feature that the card confirms the PIN. This allows offline-transactions, and is not per se insecure, if the protocol between terminal+card would have been designed correctly (which it unfortunately was not). The problem (link) is, that the current protocol allows a man-in-the-middle degradation attack: Ther terminal uses PIN+chip, but the man-in-the-middle tells the card not to use PIN+chip (i.e. to use chip+sign). The confirmation of the card is used to make the terminal think the PIN was accepted.

If the protocol is fixed (i.e. by properly authenticating the data exchange), everything would be perfectly fine. Additionally, they should get rid of the insecure payment methods (i.e. anything not involving a pin), to disable *all* degradation attacs (what use is chip and pin if any fraudster can still use all of the old payment methods with a forged card?)

The actual paper correctly says:

Non-locality can be exhibited when performing measurements on two or more distant quantum systems – the outcomes can be correlated in way that defies any local classical description. This is why we know that quantum theory will never by superceded by a local classical theory. Nevertheless, even quantum correlations are restricted to some extent – measurement results cannot be correlated so strongly that they would allow signalling between two distant systems.

Quantum entanglement (QE) provides a correlation not a communication. What this means is that not only can't you use QE to pass signals (or any information) between Alice and Bob, you actually need some other form of after-the-fact communication between them to detect the correlation in order to determine if QE happened at all. If QE was a method of communication then you could verify it by sending Bob a "cheat cheat" of what Alice was going to do or transmit. Instead, you need to look at the outcome of a series of measurements taken by Alice's and the outcome of a series of measurments taken by Bob just to see if QE actually happened.

Correlation is not communication.

Okay, I get it, text shadow is the new 'thing' on the internet. But seriously, it makes your article harder to read. There is a time and a place people. And that time and place isn't everywhere all-the-time.

None of the links lead to anything that uses text-shadow. You may want to get your eyes checked.

Actually this one does: http://www.damtp.cam.ac.uk/user/jono/uncertainty-nonlocality.html - it's faint (depending on your monitor I guess, and obviously whether your browser supports it) and above the text, the style is inherited from the body thus:

body { text-shadow:0 -1px 1px rgba(0, 0, 0, 0.41); }

If anybody cares to read it, a preprint of the whole article can be found here.

Sure a car can have computers, even my 93 Dodge has some kind of controller.

Hey we could have another TCP/IP type revolution here... add layers and gain functionality.

What is needed is to connect existing automobiles to an autonomous vehicle interface. The autonomous vehicle interface would provide a connection point for an autonomous vehicle computer to be attached.

The autonomous vehicle computer would query the interface device and get a description response of all the controls and sensors available. The interface would organize and scale the data available from the car. The interface would convert autonomous vehicle computer data into the signals expected by the devices attached to the automobile.

Think of the autonomous vehicle interface as an Arduino Mega that is wired into whatever the vehicle has available. The interface would be like interaction with a python interpreter.

Now the autonomous vehicle computer, think of that as a Linux netbook running a variety of programs for the car. There could be add on sensor devices attached to the autonomous vehicle computer. Like a GPS, a data cell phone and a 802.g wireless connection to nearby vehicles and 802.g radio equipped traffic devices.

http://lessco2essay.blogspot.com/2010/11/proposal-for-autonomous-vehicle.html

So what to do with such a modification: Completely end drunk driving accidents. Reduce the kWh per 100 passenger-kilometres. Do aggressive dynamic insured and paid ride sharing.
Dramatically reduce distracted driving damage. Reduce direct fuel use by coasting up to stoplights. All of this with existing vehicles.

http://www.inference.phy.cam.ac.uk/withouthotair/c20/page_118.shtml

The exciting part of doing AVT autonomous vehicle technology like this is: It is not proprietary and locked up in islands of make and model specific functionality. Some aspects of AVT can be backported to most older cars and real energy and safety benefits accrued.

It is already amazing what can be done with an optical camera.

However using depth images Andrew Johnson did some impressive work on recognising objects in 3D depth maps. And Dan Munoz recently worked on applying this kind of algorithms to Willowgarage's PR2 robot. With your Kinect driver, depth sensors are getting within reach of hobby developers.

If he doesn't want to use the other hand or if he would like to explore other artistic entry methods, he should try Opengazer

Interesting to note that the "war machines" comment was not from a journalist but Sir Greg Winter, Deputy Director at the laboratory undertaking the research :-) Link to original article http://www2.mrc-lmb.cam.ac.uk/news-and-events/lmb-news/lmb-scientists-redefine-how-our-immune-system-responds-to-viruses Link to academic publication http://www.pnas.org/content/early/2010/11/01/1014074107.full.pdf+html

And, yes, me too: I wrote this in vi(1), which is why the article does not have all the fancy Unicode glyphs in the first place.

Excuse me - vim can handle utf-8 just fine. utf-8 file names and utf-8 content. on a vanilla slackware 13.1.
http://www.cl.cam.ac.uk/~mgk25/unicode.html#apps [cam.ac.uk]
# Vim (the popular clone of the classic vi editor) supports UTF-8 with wide characters and up to two combining characters starting from version 6.0.
# Emacs has quite good basic UTF-8 support starting from version 21.3. Emacs 23 changed the internal encoding to UTF-8.
And svn can handle utf-8 as well - http://svnbook.red-bean.com/en/1.4/svn.advanced.l10n.html [red-bean.com].

The repository stores all paths, filenames, and log messages in Unicode, encoded as UTF-8.

All it requires is ... set your locale and lang. "export LANG=en_DK.utf8" in "/etc/profile.d/lang.sh" (Slackware 13.1) and add some better fonts maybe.

I apologize for repeating myself. I've written the same thing further down already in reply to another user's post. But I just read tfa and felt the need to reply to the author of tfa.

the simplest programming tools for editing are straightfoward ASCII text editors: vi and (god help us) emacs. so by declaring that "Thou Shalt Use A Unicode Editor For This Language" you've just shot the chances of success of any such language stone dead: no self-respecting systems programmer is going to touch it.

Excuse me - vim can handle utf-8 just fine. utf-8 file names and utf-8 content. on a vanilla slackware 13.1.
http://www.cl.cam.ac.uk/~mgk25/unicode.html#apps
# Vim (the popular clone of the classic vi editor) supports UTF-8 with wide characters and up to two combining characters starting from version 6.0.
# Emacs has quite good basic UTF-8 support starting from version 21.3. Emacs 23 changed the internal encoding to UTF-8.
And svn can handle utf-8 as well - http://svnbook.red-bean.com/en/1.4/svn.advanced.l10n.html.

The repository stores all paths, filenames, and log messages in Unicode, encoded as UTF-8.

All it requires is ... set your locale and lang. "export LANG=en_DK.utf8" in "/etc/profile.d/lang.sh" (Slackware 13.1).