Slashdot Mirror

From https://www.sno.phy.queensu.ca...

The stand-alone Windows executable does not require Perl.

Sounds to me like the Windows version is superior.

http://www.sno.phy.queensu.ca/...

exiftool is an opensource perl library. You can do everything Image Tagger does on the command line with exiftool. Image tagger however does all the hard work of building the commands for you.

It's another UI prototype about paper-thin hardware; in this case it bends when you touch it. It's not a finished product, though. "Prototypes" in this case means "we wired up a piece of cardboard and a projector to simulate what we think this one day might be like, maybe, just so we could explore interface design questions."

For the past several years HCI research has been pursuing various paper-thin interactions (another example, from my alma mater.) As a general rule they're very novel and creative ideas, but as far as I can tell there isn't a soul on the planet who would actually want to try and use such a horrible pseudo-skeumorphic mess. Paper's just too thin and delicate to use as a UI device.

But don't be too hard on the summary: they managed to avoid scarring your eyes forever with the prototypes' actual name, which is so horrible you'd think it was invented to market toys to six-year-olds.

Shameless plug: http://perk.cs.queensu.ca/software We do exactly this. Our software is open source for anyone to use/test/fix. We do use SVN to maintain some control over the code that is commited, but overall it works quite well. We have just launched some projects on github; it's a new experiment and we're interested to see how it turns out.

The newest camera / smartphones have GPS chips to geotag pictures so they can be overlaid on maps. For GPS-less cameras:

EXIFtool

GPSbabel

Have a GPS device turned on and logging tracks, take pictures, use the tools to add geotags to pictures.

... or use EXIFtool to strip identifying and geographic information before posting a picture.

"Even the Daily Mail thinks it's true" was not the basis of my argument. I gave a long list of scientific papers on the topic and then pointed out that the Daily Mail was unable to lie its way out of what was at hand. I discussed one of the claims made in the Daily Mail article, a presumably factual statement made by expert who works at Bastoy. At no point did I state:

(a) "it's true because the Daily Mail says so."

I did aim to communicate the following, however:

(b) "even the Daily Mail is convinced, because a prison guard said Bastoy has the lowest re-offending rate in Europe. That it does is remarkable, and supports the GP's claim."

(And technically this is incorrect since the prison guard didn't say it; that was research conducted by the author of the article. I believe this is the study used, but I don't have access to confirm it.)

(b) is semantically different from (a), and does not claim that the Daily Mail is an arbiter of correctness. I should also stress that I stated quite knowingly and clearly that the Daily Mail's opinion did not have "the integrity of a longitudinal study conducted by unbiased researchers."

I realise that the structuring and presentation may have suggested an attempt at deferring to the authority of the Daily Mail, and I initially considered writing the post as one, but the actual body of the post depends on the arguments made in the Daily Mail article, not the opinion of the journalist. If you still think I have presented something that amounts to deferring to the journalist's opinion, then I would counter by saying that I have justified and accounted for that opinion; my post then amounts to "the Daily Mail believes x because they did research and found y is true. Since we trust y, and assuming no wrongdoing on the part of the Daily Mail, we can also trust that when the Daily Mail says x, it is correct." This is not a fallacious argument either.

Probably a one-liner with exiftool!:

http://www.sno.phy.queensu.ca/~phil/exiftool/exiftool_pod.html#writing_examples

Unfortunately, doing this to Nikon raw files damages them for downstream processing, as the serial number is one of the keys for Nikon's lame encryption scheme that 'protects' white balance data.

I wonder if somebody could hack exiftool into a GIMP plugin.

Personally, I've just snapped a picture of the inside of my lens cap and then edited it in GIMP on the rare occasion when I wanted to have EXIF data in something I'd created.

Professional digital still cameras do indeed contain a large amount of metadata including photographer name, serial number of body and lens, GPS location and various other identifying information. Blobs like the thumbnail of the original unedited image can be included. And beyond metadata, the filename itself is from a timestamped sequence. Metadata can be viewed and edited with something like Phil Harvey's exiftool. I have not looked at video files myself though.

Oh and if my other reply isn't enough. You can read this(that's queens university in canada) it's rather dated, but covers the 20/80 and 10-10 on theft, and employee theft. My text books and notes are buried over yonder, and I still have no interest in digging through them. And they're more up to date based off of 99 though 2006.

Some friends in Cairo would like to bypass some of the online censorship measures. I've quickly suggested some things (below) to consider overnight. What have I missed?

Anonymous connection:
No:
https://www.eff.org/deeplinks/2010/01/help-eff-research-web-browser-tracking

But:
https://www.eff.org/https-everywhere/

Also:
http://www.hotspotshield.com/

And services like:
http://filesharefreak.com/2008/10/18/total-anonymity-a-list-of-vpn-service-providers/
but verify on the ground.

Only if they understand the tradeoffs:
http://www.privoxy.org/
https://techstdout.boum.org/TorDns/

Avoid random lists of anonymous proxies or DNS servers.

To secure the computer:
Use a popular boot disk that leaves nothing behind, e.g.:
http://www.ubuntu.com/desktop/get-ubuntu/download

Remove metadata:
http://owl.phy.queensu.ca/~phil/exiftool/
http://www.microsoft.com/downloads/en/details.aspx?FamilyId=144E54ED-D43E-42CA-BC7B-5446D34E5360&displaylang=en
and similar for other files they may deal with.

Delete/wipe files securely.

Many uses:
http://mailinator.com/
http://www.hushmail.com/

Consider:
http://www.disconnectere.com/
and its analogues

Software already available for that (has been since Instant TeX on a NeXT though that only did single characters):

InftyEditor - http://www.inftyproject.org/en/software.html

FFES - http://research.cs.queensu.ca/drl/ffes/

There's also

MathJournal - http://www.xthink.com/MathJournal.html

which is a commercial product, the new version supports LaTeX

William

I have found an interesting program called exiftool), a great command line tool for linux, mac osx and windows (there is apparently a windows gui too).

Yes, the latest professional cameras do indeed embed a lot of information including the photographer's name, multiple lens and body serial numbers, etc. This does not change the file name or creation date in the file system. Also the metadata I saw includes F-stop, exposure, etc. Perhaps if it had a sat chip there would be geographic coordinates on it. So the only way to be sure is to just erase all metadata not just a few tags.

It is important to understand that there is a lot (several pages) of metadata, some of which may be in binary format, which may not all be shown by the kinds of programs used by consumers ordinarily. Of course you can stare at the output of "strings" until you are blue in the face but ultimately you have to trust some tool that it has found all the metadata.

So far exiftool is what I trust but I know this is a best effort kind of thing. It seems that different manufacturers include different information, and the amount is far beyond what is needed to display an image. If the embedded data gets steganographed or otherwise hidden in the file it won't be easily found since people will not know it is there.

I started looking at these programs because a photographer asked me to sell his photos and was uncomfortable about personal data being left in files being handed to other people he didn't know. Not so unusual I would imagine.

Perhaps these cameras provide this data for the photographer's own records. But I think manufacturers should definitely include a template (a list of tags or byte offsets, not just "delete all") for an open tool like exiftool, which they guarantee, as part of their contract with the purchaser, will delete all metadata or embedded data. Then they are legally required to specify all tags. Otherwise I don't see how you can be sure you have anonymized your photos.

Speaking as someone who has worked with photojournalists I can assure you that such is to be desired. Perhaps it could be said that unless this comes to pass, no modern journalist can be said to be shielded from potential witch hunts, once his or her digital media is released to the Internet.

There are a couple of projects which OCR math properly:

http://inftyreader.org/

http://research.cs.queensu.ca/drl//ffes/

William

You don't ride taxicabs in Munich

Very interesting link. Thanks!

I've grown up very slightly - in large part due to taking an advanced driving course and having a subsequent driving ban due to speed, rather than simply because I'm older - but the point is that I definitely don't take risks just because I know my vehicle is "safer".

I suppose if someone doesn't actually know what ABS does then they may be more inclined to allow less time for braking etc, but ABS can actually increase stopping distance, for the sake of retaining directional control.

An interesting point from the article someone linked above is that ABS reduced the number of accidents "caused by the driver", but increased the number of accidents which "they had no control over". I'm presuming this to mean that they were better able to avoid obstacles and retain control of their vehicle, but in situations where there was nowhere safe to turn, the extra stopping distance caused them to crash.

Subsequent analysis of the rating scales showed that drivers of cabs with ABS made sharper turns in curves, were less accurate in their lane-holding behaviour, proceeded at a shorter forward sight distance, made more poorly adjusted merging manoeuvres and created more "traffic conflicts". This is a technical term for a situation in which one or more traffic participants have to take swift action to avoid a collision with another road user.[3] Finally, as compared with the non-ABS cabs, the ABS cabs were driven faster at one of the four measuring points along the route. All these differences were significant.

If you are looking for a tool to read and edit the exif (geolocation etc) information in files, exiftool http://www.sno.phy.queensu.ca/~phil/exiftool/ works on Windows, Mac and Linux and support many formats.

For adding exifdata to a number of photos at once, you can use exiftool http://www.sno.phy.queensu.ca/~phil/exiftool/ I also use NikonviewNX to add titles to my photos. This is free from Nikon.

However, I don't touch the originals and only update copies of the photos.

Once you have data in the photo's exifdata, you can extract for other purposes using exiftool or the underlying libraries.

Alan

I've been searching for the same feature set, a file centric image manager whose metadata is stored exclusively in the file.

One of the best ones I have found is Mapivi:

http://mapivi.sourceforge.net/mapivi.shtml

I still often use Digikam, but its metadata support is inconsistent at best. On the other hand the front end is more useable than Mapivi.

You should also look at ExifTool, because you can manipulate and query metadata with it on the command line.

http://www.sno.phy.queensu.ca/~phil/exiftool/

If you find a solution, please share!

An "ordinary" quantum computer is no more powerful than a Turing machine. It probably can't even solve NP-complete problems quickly because there's no way of directing the observation step (where you pull the answer out of the numerous universes) in such a way as to make the right answer show up often enough. This is, technically speaking, an open problem (just like P vs NP), but most believe the answer is a negative (just like P != NP).

I said "ordinary" quantum computer because there have been attempts at making quantum hypercomputers (i.e. computers more powerful than a Turing machine). The best known is Kieu's adiabatic quantum hypercomputation scheme, but that appears to have been refuted. Quantum adiabatic algorithms can be useful (they're a bit like genetic algorithms, only on the quantum level), but apparently they can't bring quantum computers above Turing power; a bit like the Brownian ratchet in this manner, in that it's not as powerful as thought, but still useful.

To sum all of that up: it's unknown whether it's possible to make a quantum computer solve NP-complete in polytime, but most think it's unlikely. It's unknown whether it's possible to make a quantum hypercomputer as well, but most think it's even less likely (although if consciousness would disprove this, that would be... interesting).

If metadata is such a concern, one simple command with ExifTool http://www.sno.phy.queensu.ca/~phil/exiftool/ can recursively delete all exif info in your whole image collection. And no, I will not post it can be just as destructive as rm -rf /

Exiftool. Perl, but with standalone packages for platforms where Perl is not available by default.

You should be able to look at the most interesting details in most up to date image software.

The buzzsaw is ExifTool:

http://www.sno.phy.queensu.ca/~phil/exiftool/

It doesn't sound like anything is proven, or else it would be "case closed".

Wikipedia is not authoritative. Neutrinos have been known to have mass for over five years now, and the physics community is now focused on refining the parameters that characterize massive neutrinos.

Although we know that neutrinos have mass, we don't know what the mass is because our current experiments are only sensitive to the square of the mass difference between different types of neutrino. However, we do know that all types of neutrino have mass, although the most plausible values are less than a millionth of the electron mass, making it tricky to detect by time-of-flight measurements because any detectable neutrino is going to be ultra-relativistic, travelling so close to the speed of light as to be indistinguishable from a massless particle under almost all circumstances, which is why it was so difficult to prove they do have mass.

Of course it might have... But if everything is staying about the same.. it can't be as "zomg scary" as the media makes it out to be. I think its highly likely people would just find other ways to drive less safely to achieve the same amount of risk on the road ways.

I think we have enough useless laws as it stands. Prohibition of communication should be low on the priority list, in my not so humble opinion.

Take a look at the Freehand Formula Entry System (GPL). Handwriting recognition for mathematics.

The poor Power Factor rating of CFLs means that the power grid must provide more energy to start them. The consumer may save money at home, but unless the power factor of CFLs can be improved, the electrical grid itself will have to be upgraded if we all switch to CFLs.

But there's no "if" because governments have already legislated the elimination of incandescent bulbs!

An explanation of the power factor (search for the heading "Power
Factor and Switching")
http://irc.nrc-cnrc.gc.ca/pubs/cp/lig3_e.html

More summaries of problems:
http://www.cours.polymtl.ca/inf1040/2008automne/Olivier_CanadianReviewDec2007.pdf
http://sound.westhost.com/articles/incandescent.htm

Current research:
http://qnc.queensu.ca/story_loader.php?id=49db90a6e3e3d

Search for "power factor":
http://www.bchydro.com/powersmart/technology_tips/buying_guides/lighting/compact_fluorescent_lamps.html

Here, you'll see that the "requirement" for "Energy Star" labelling is
a power factor of only 0.50!
http://oee.nrcan.gc.ca/residential/business/manufacturers/specifications/compact-flour.cfm?attr=12

If the image quality is sufficient, Nikon point and shoot cameras offer about 2/3 of what you are asking for (quiet, AA batteries):

http://www.nikonusa.com/Find-Your-Nikon/Digital-Camera/index.page

Good luck on getting them to change the filename standards (the arguments about what is correct could go on forever). Exiftool makes it straightforward to rename a bunch of pictures based on time and date and so forth (but probably not easy):

http://www.sno.phy.queensu.ca/~phil/exiftool/

I guess none of us are safe or have any real protections... we just need to always make sure popular opinion is on our side. In your twisted view of how government works, I mean.

Yes, that is actually true, unfortunately. Government in any case is just an extension of the will of the people, even in the worst case of a monarchy, if the people collectively decided not to do what the King wanted, then the king would have no power (although often a brutal dictator doesn't need the direct support of more than 30% of the people).

United States history is full of instances of smaller groups of people being oppressed by larger groups. The civil war was the North trying to force its will on the South; we had full on SLAVERY for many years, we oppressed the Native Americans, we destroyed the Hawaiian culture and language (and are still forcing our will on the native Hawaiians, who feel underrepresented in their attempts at sovereignty), persecuted pacifists during WWI with the Espionage Act (EE Cummings was arrested for declaring he didn't actually have a hatred of the Germans), persecution of communists, even before McCarthy era, open mob violence against Jehovah's Witnesses (for reasons such as their refusal to say the pledge of allegiance), open mob violence against the Mormon's along with political persecution against both, discrimination against gays, discrimination against polygamists............... I'm sure if you look you can find many other cases.

More and more these days people are becoming of the opinion that persecuting others just because you don't like them is a bad idea, and, as you put it, protecting people from the tyranny of the majority, protecting fundamental civil rights is actually a good idea. These are things that can only happen if enough people support them. That ought to be the true American Way

To give you an idea of how things have been throughout history (paraphrased from Irving Stone) between Paul Revere and Abigail Adams, about a famous drawing he drew about the Boston Massacre:
Abigail: Why did you draw that picture? You know it is not what happened.
Paul: I know, but it helps people see the evils of monarchy.
Abigail: Those poor soldiers will be punished because of it, and it's not their fault! How would you feel if someone made such a picture of you, and you were punished because of it
Paul: That won't happen, I have too many friends.

John Adams apparently became very unpopular in Boston after his successful defense of those soldiers. But that is ok, he did what was right. Sometimes heros step up to the challenge, but it is not something we can depend on. Society only works when we agree to follow what makes it work.

I spent way too much time on this post. You better enjoy it. ;)

Hex editor? Really? Why use a hammer when you can use a buzz saw:

http://www.sno.phy.queensu.ca/~phil/exiftool/

more technical details:

http://research.cs.queensu.ca/~skill/uselection/index.html

A few reasons.

First, there actually are such organizations.

For example:
Comparative Cognition and Behavior Reviews
http://psyc.queensu.ca/ccbr/
a journal that is free, open web access, and still peer-reviewed by experts in the field.

I cannot find any such journal in physics or biophysics. I can either submit to a print journal where it will not be open (for at least some period of time) for free or an open access journal for a lot of money.

At my university, I can get away with going with a low profile publication but I know most people do not have that luxury.

A few reasons.

First, there actually are such organizations.

For example:
Comparative Cognition and Behavior Reviews
http://psyc.queensu.ca/ccbr/
a journal that is free, open web access, and still peer-reviewed by experts in the field.

But, there is still a large credibility gap. For example, if you put down on your resume that you were a senior writer for CNN, that has more weight when you go to find your next job than if you were a senior writer for a local newspaper. Also, large journals have a much broader readership and the ability to disseminate their work to other media outlets, which increases the probability of your work getting cited and referenced.

It's also not so easy to start a journal. Google scholar requires that you be a reputable source before they index you. OVID requires a much more stringent standard. ISI Web of Science requires an even more stringent standard. In a day where libraries are essentially not used and everyone finds out about your work through the internet, not having your work published in a major search engine is a serious publication barrier. There are also a lot of technological and technical (page setting, layout, typesetting, etc) barriers to creating professional looking manuscripts that are non-trivial.

Lastly, it is a difficult fight to win. You work through undergrad to get into a good grad program. You work during your grad studies to get a nice job. You get a nice job and you work hard to get tenure. You get tenure and you work hard to get a grant. You get a grant and you work hard to get it funded a second time. When in this process are you supposed to start a journal that competes with other publication outlets, and who is going to take you seriously when you do? It's very difficult for an individual scientist to buck the system, even though most are strongly pro-access.

It would be a little bit like being an independent musician, except there are no cheap bars where you can play small gigs to get by while you fight your crusade against the machine. Musicians always have that choice: 1) you can sign with a big label, 2) sign with a small independent label that preserves your rights but makes it very difficult to get widespread recognition, or 3) quit the music business. Scientists only get option 1 or option 3.

All right, clever people of Slashdot, answer me this:

If instead of making a balloon out of this atom-thick material you simply made a large sheet (say, the size of a sheet of A4 paper) could a person fold this super-thin sheet in half more than twelve times? That's the current record, shattering the previously accepted limit of eight folds.*

* Regarding the latter link: I know. (Ouch! My eyes!)

Not multi-platform, it stomps on makernotes, and even requires .net3, but a decent gui:

http://itagsoftware.com/download.php

The tags end up in the iptc and XMP headers, not a database. It isn't a big deal to get them out with something like Exiftool:

http://www.sno.phy.queensu.ca/~phil/exiftool/

and I think iPhoto even understands them.

(Exiftool has great support for makernotes, so something like iTag->IPTC-in-damaged-photo->Exiftool->Undamaged-photo should be possible)

We don't need central repositories like facebook to have social networks. We can do it ourselves in a distributed fashion. Here's how: Friends in Feed.

A couple of examples come to mind.

The Government of Canada marijuana farm is located in an old copper mine in Manitoba. You can't beat the security, which is something mentioned in tfa. http://www.cbc.ca/canada/story/2001/08/02/marijuana_010802.html

A solar neutrino observatory is installed in an old mine in Sudbury, Ontario, Canads. It has the advantage of being impervious to almost all kinds of radiation, except of course for neutrinos. http://www.sno.phy.queensu.ca/

As I look at the other posts, I see lots of naysayers. Well there are at least a couple of cases where old mines have been used successfully for other things.

Some photos from the teams this year:

http://engsoc.queensu.ca/sailboat/Photos.htm
http://www.roboat.at/en/fotos/
http://richard.shipman.me.uk/gallery/Roboat
http://web.ensica.fr/microtransat/slideshow/Challenge%202007/index.html

and a few from the 2006 event:

http://www.ensica.fr/microtransat/Challenge2006/Gallery.htm
http://users.aber.ac.uk/cjs06/homepage/index.php?display=content/Photos/Research/Microtransat%20Competition%202006
http://wiki.atrox.at/index.php/Microtransat-Fotos

Using a program he wrote (and provided on the conference CD-ROM) Krawetz could print out the quantization tables in a JPEG file (that indicate how the image was compressed) and determine the last tool that created the image -- that is, the make and model of the camera if the image is original or the version of Photoshop that was used to alter and re-save the image.

Comparing that data to the metadata embedded in the image he could determine if the photo was original or had been re-saved or altered. Then, using error level analysis of an image he could determine what were the last parts of an image that were added or modified.

If the space-saving feature of laptops is desired, look into thin client technology. The boxes are the size of chocolate/cigar boxes and can sometimes be bolted to the back of an LCD monitor or included inside the monitor. A thin client is way cheaper than a laptop, too. I do not see the value of a laptop being twice what a desktop device is. Why can I get a good desktop for a few hundred dollars but a laptop is starting at $600? If we were all salesmen or writing novels on the bus, I could see a laptop, but most of us are moving from one flat spot to another. We can afford a PC at both places for the cost of a laptop.

Here is an image of a thin client setup. Monitor and keyboard are way larger than a laptop but it takes up very little space/material/money/maintenance compared to a laptop. Combined with a network and a Linux terminal server, setups like this enable a whole department to be maintained by keeping one server going so we can compute instead of labour.

The one situation where I personally would use a laptop is if I were moving around and giving presentations and needed an application like OpenOffice running where I went. I am looking at one of those Dell laptops with Linux and in bright yellow... They do not sell them in Canada yet.

http://www.hml.queensu.ca/?q=node/126 Time to get me some tinfoil glasses.

Nope. A nuclear bunker won't do it. You need one of these: http://www.sno.phy.queensu.ca/images/mine.GIF.

Holy crap! When did they bury the CN tower? Those damn terrorists, now we have to bury our landmarks to guard them!

Nope. A nuclear bunker won't do it. You need one of these: http://www.sno.phy.queensu.ca/images/mine.GIF.

The box will give some protection, the metal of the freezer some, a lead box some, a 2km deep mine some... how much do you want? All radiation cannot be shielded because your shielding material will invariably be radioactive. Although, you can cut down that effect a bit by purchasing very expensive ancient lead from Roman shipwrecks. You'll be competing with a couple of observatories though.

Neutrino oscillations are a process by which different types of neutrino can turn into each other. The elementary particles (quarks, leptons and neutrinos) all come in three "families". We are made of the lightest family: up and down quarks (which are the constituents of protons and neutrons) and electrons. Members of the heavier families are unstable and decay rapidly into lighter particles.

However, it turns out that the weak nuclear interaction can mix quarks of different families. Down quarks turn out to be somewhat mixed with strange quarks of the next heaviest family due to this effect.

For a variety of reasons, it was natural to ask if neutrinos were mixed in the same way. In particular, this could account for the unexpected deficit of electron-type neutrinos from the sun. Various terrestrial experiments were done in the 80's and 90's to try to detect this effect, including LSND.

Neutrino experiments are extremely difficult and subject to all kinds of backgrounds, making them highly susceptible to errors in calibration and calculation. The LSND results were at odds with everything else that had been seen, but the stakes were high and no one wanted to give up on a result that might be right although it was not widely believed by people outside the LSND collaboration itself.

The experiment described in TFA has tried to independently reproduce the LSND results. This is somewhat easier to do than the original experiment because you can design things so that you are most sensitive to the most interesting region. They have failed to find the effect that the LSND result would predict if it was due to neutrino oscillations, and it is likely that this is the end of it.

The article never says so, but the most likely cause of the LSND result is some error in analysis, particularly in accounting for backgrounds and instrument effects. This kind of thing happens, particularly in neutrino physics, where the background processes are fundamentally many orders of magnitude stronger than the effects you are looking for, and have to be designed out with the most excruciating care.

like this or like this or like this or with this if you want to go low tech (light has no charge and is smaller than a proton).

Okay sorry for the flippant answer, basically in particle physics, protons are huge, very large (but not massive) objects. Finding something smaller than them is pretty easy because size doesnt matter. What matters is the strength of its interaction with the rest of the universe. So we find small objects via their interactions with other objects which we can detect in our detectors. No charge makes things a little more tricky but objects can also carry colour charge and weak isospin and thats how we would find an electrically neutral object. Neutrinos, the hardest particles to detector only interact via the weak force and they are almost impossible to see but we do detect them. Also we can detect things like neutrinos by the absence of things, they carry away energy from the collision and we can detect that theres not all the energy there should be.

Taking notes in math classes is actually possible!

Have you ever tried this?

I'm getting one of these X60 tablets, and I plan to run Linux on it. Do you have any links to information that would be useful to me? How much effort did it take for you to get all this working?

What isn't really available, and I'm sure this is what the parent post was talking about, is handwriting recognition software. AFAIK there isn't really any available for GNU/Linux (please reply if I'm wrong here).

Well, there's stuff like XStroke, but this post doesn't sound very encouraging. One thing I have found that seems cool, though, is this equation-entry system. Imagine how cool it would be if that were integrated into a general-purpose utility...

Anecdote related to heavy water and its cost:

I had the chance to visit the Sudbury Neurtrino Observatory (SNO) and their system uses 1000 tons of heavy water. The detector sits in a big (ten-story) chamber filled with heavy water, which also gets pumped around to purify and test for contaminants (a large part of the facility is dedicated to this). The water has a value of $300 million, but they do not own it, rather they loan it from Atomic Energy of Canada Limited. So, they have to make sure it stays super clean, and that it does not leak much, since they have to give it back at the end of the experiment.

It's a pretty cool place. All these big facilities seem like out of science fiction. The detector (photo) is permentantly walled off (except for a hatch they can put a miniature submarine into), so you can't see it, but the huge masses of pipes and wires there is impressive

.

Anecdote related to heavy water and its cost:

I had the chance to visit the Sudbury Neurtrino Observatory (SNO) and their system uses 1000 tons of heavy water. The detector sits in a big (ten-story) chamber filled with heavy water, which also gets pumped around to purify and test for contaminants (a large part of the facility is dedicated to this). The water has a value of $300 million, but they do not own it, rather they loan it from Atomic Energy of Canada Limited. So, they have to make sure it stays super clean, and that it does not leak much, since they have to give it back at the end of the experiment.

It's a pretty cool place. All these big facilities seem like out of science fiction. The detector (photo) is permentantly walled off (except for a hatch they can put a miniature submarine into), so you can't see it, but the huge masses of pipes and wires there is impressive

.