Back to previous post: Tropical scrud

Go to Making Light's front page.

Forward to next post: Power survey

Subscribe (via RSS) to this post's comment thread. (What does this mean? Here's a quick introduction.)

August 20, 2003

Kiss your glass goodbye
Posted by Teresa at 05:45 PM *

A lot of sites are blogging The New Diamond Age, a story in Wired, about a couple of companies that may have cracked the problem of synthesizing large top-notch diamonds. You should have a look; the technical details are fun.

There are a couple of interesting angles on this. One is, if one or both of these companies have cracked the synthetic diamond problem, they’ve also broken the De Beers diamond cartel. This is a good thing. De Beers, a monopoly run by the Oppenheimer family, is one of the great supervillain organizations. Consistently, over the last several generations, they have systematically lied to the retail customers about the permanent investment value of diamonds. They go to great and unscrupulous lengths to make sure they buy almost all the raw diamonds mined on this planet, stockpiling them to keep prices high. They support iniquitously evil regimes and insurgent groups by buying diamonds from them, knowing that by doing so they’re helping fund their activities. And it scarcely needs saying that De Beers’ labor policies are awful.

So hell with them. Off to the dusbin of history with De Beers.

Another interesting angle has nothing to do with jewelry. Diamond microchips could run much, much faster than silicon chips can. They have tremendous thermal conductivity, which means they won’t melt in the heat generated by all that processing. If Moore’s Law is going to continue to hold true, synthetic diamond microchips could be one of the means by which it happens.

Naturally, that would be a very cool thing.

But what I’m thinking about now is what happens when all those diamond chips get into circulation on the street. The current form of graffiti on NYC subway cars looks primitive compared to the polychrome spraypainted extravaganzas of bygone years. Trouble is, the new cars in the system have special surfaces and coatings that make pigmented graffiti hard to put on and easy to take off.

Kids in their teens have responded by using keys to scratch their initials onto the walls and windows. It’s ugly stuff—big squarish malformed letters made up of multiple scratches—and I doubt it’s anything like permanent, because I hardly ever see graffiti overlaid upon earlier graffiti. But apparently they have this irrepressible urge to leave their initials on subway cars, because they keep doing it.

(There’s another a subclass of vandal that likes to scratch stuff into the brushed-metal elevator doors in the Flatiron Building and other office buildings. I resent that one. It’s difficult and expensive to repair, the brushed metal doors are pretty, and the graffiti gouged into them is never interesting or creative. Eh, jerks, what can you do.)

Anyway, what occurs to me is that if diamond chips come into general use, there’s not a surface out there that’s safe—not glazed tile, or steel doors, or opaque architectural glass, or picture windows. A diamond will grave straight into the surface of any of those materials.

I’m wondering whether we maybe ought to loosen up about letting kids do spraypainted graffiti.

Comments on Kiss your glass goodbye:
#1 ::: Stefan Jones ::: (view all by) ::: August 21, 2003, 04:02 PM:

Teresa, I had the very same thoughts about the De Beers when a story about this piece ran on NPR the other day.

Except I think the Dustbin of History would be too kind a fate for them. The Compost Heap of History, perhaps, or The Medical Waste Bin of History, full of syringes and pus-laden bandages and latex gloves that have been in questionable places.

The NPR piece has a lovely bit about a De Beers wonk turning white when examining the synthetics.

Do you think they'll start a campaign decrying phony synthetic diamonds? Or plant weepy stories about miners being put out of work? Or maybe just bribe governments to ban the synthetics?

I think they'll try.

#2 ::: Yonmei ::: (view all by) ::: August 21, 2003, 04:30 PM:

If it becomes possible to produce diamonds cheaply, easily, and quickly... how about foiling scratched-graffiti by coating surfaces with diamond?

#3 ::: David Moles ::: (view all by) ::: August 21, 2003, 04:39 PM:

I’m still looking forward to the graffitti Gibsond describes in All Tomorrow’s Parties that crawls around the surface of the wall while the graffiti-eating nano tries to chase it down and eat it.

#4 ::: David Bilek ::: (view all by) ::: August 21, 2003, 04:41 PM:

Clearly, the only viable strategy for De Beers is to tell women "If he really loves you he'll buy you a true diamond, not an artificial one."

Whether or not this will work remains to be seen. Sadly, I think it probably will.

#5 ::: Stefan Jones ::: (view all by) ::: August 21, 2003, 04:49 PM:

Maybe there will be a marketing arms race of sorts.

De Beers: "Look for the logo!"

Synthetic diamond merchants: "Give her a diamond the size of an ostrich egg for less than that speck with a logo."

#6 ::: Chuck Nolan ::: (view all by) ::: August 21, 2003, 04:57 PM:

Being newly engaged, this is a VERY interesting story. Where can I buy some? And when?

#7 ::: Trinker ::: (view all by) ::: August 21, 2003, 05:15 PM:

But diamond-tipped gravers have been around for yonks, relatively the whole "what if diamond chips get easy to acquire" thing...or maybe it's just that the vandals haven't thought of doing that yet.

#8 ::: CEP ::: (view all by) ::: August 21, 2003, 05:28 PM:

Diamond chips aren't possible, at least not using the current paradigm of semiconductors. They're too pure (not to mention that electrical conductivity is wrong).

Then there is the problem of manufacturing a chip. Let's assume for a moment that the chemical problem of producing a synthetic diamondlike material with the correct electrical properties does get solved. OK, we've got a superhard surface that's impervious to acids in laboratory time. So how do we trace the electronic paths on the chip? If we burn them with a laser, we've also changed the electron conformations of the surrounding atoms. Irrevocably. Thus, it seems that the only way that synthetic diamond could be used in a chip is as part of a heat-dissipating sandwich. For a contemporary moderately complex four-layer chip, that would actually increase it to seven layers (two outside diamond layers, plus a diamond layer in each of the three spaces between electronic layers, plus the four electronic layers themselves). Now we begin to add registration and crosslinking problems...

I suspect that building organic and bioinorganic computing systems will prove substantially less complicated and costly than using synthetic diamond, particularly in mass-production quantities.

See what happens with a degree in biochemistry? One begins to realize that "better living through chemistry" is a curse as much as it is an ad slogan.

#9 ::: Jordin Kare ::: (view all by) ::: August 21, 2003, 05:40 PM:

For scratched-on graffiti, I doubt this changes anything -- not only are there already diamond-tipped engraving tools, but diamond abrasive materials (belts, wheels, sheets) are only moderately more expensive than ordinary abrasives, and even ordinary abrasives are harder than most anti-grafffiti coatings.

(I'm actually a bit surprised that this is new; I'd've sworn I'd seen articles years ago on a Japanese process for growing large gem-quality stones. Perhaps it proved impractically expensive, or perhaps it was bought up by DeBeers and suppressed. That would certainly be *my* approach to these inventions, if I were DeBeers.)

I've handled a polycrystalline diamond sheet. The thermal properties are very strange -- touch an ice cube to one corner of a 3" square and the opposite corner, where you're holding it, almost instantly feels like a piece of ice. It violates your common-sense expectations about heat in the same way a block of lead violates your expectations about weight.

I don't believe diamond semiconductors will revolutionize microprocessors -- although diamond heatsinks may start showing up. (Fannish physicist Howard Davidson at Sun has done a lot of work on them) Silicon has too many really nice properties for making complex IC's. But diamond semiconductors may make great high-power microwave transistors, thereby taking over the last remaining niche for vacuum tube technology.

#10 ::: PiscusFiche ::: (view all by) ::: August 21, 2003, 06:03 PM:

I loathe DeBeers. (Fuck you and your "a diamond is forever" bullshit. Pardon my French.) I've also never been big on diamonds to begin with--although I'm a big fan of industrial uses for luxury products like gold and diamonds.

I remember going into a jewelry store with a friend once and shocking the store clerk with my pronouncement about diamonds--after all, I still want to get my hand in my pockets once I get engaged. No diamonds for me!

An engagement computer sounds nice though. And a little titanium or platinum ring to go with it.

#11 ::: Stuart Dimond ::: (view all by) ::: August 21, 2003, 06:11 PM:

CEP says: Diamond chips aren't possible, at least not using the current paradigm of semiconductors.

If you read the Wired article you will find they address the issue of creating both N and P type semiconductors using Boron doping. Diamond semis are well within reach using standard processing technology.

Silicon on Insulator technology is old news, the most common form being silicon on saphire.

I think what we will see as the semiconductor industry matures are a number of generations of technology in existance in parallel rather than every product being made with the latest process technology. Moore's law is irrelevant to many kinds of semiconductor products. Chip sizes are in many cases limited by the size of the pads the lead wires bond to, not by the area of the active circuitry.

Chips that can endure high temperatures are useful for many different markets, one of the biggest being automotive electronics. Beyond that I have worked with with engineers doing down hole systems (down the bore of an oil well) that had to function above 200 degrees Centigrade. I don't see organic and bioinorganic technology replacing silicon or carbon semiconductors in those applications.

The other issue where diamond semis might shine is radiation hardened applications. The space program uses 15 year old chips because the more modern small geometry devices can't handle the radiation. The electronics of an interstellar probe will probably be made with diamond semiconductors.

I think Arthur Clarke wrote a story many years ago in which the protagonist has a wife who loves diamonds. He has spent a good part of his career trying to synthesize diamonds. While part of a moon expedition he finds a lode of diamonds. His wife divorces him for mental cruelty when the diamonds he brought back for her from the moon are surpassed by the synthetics from his finally successful process. Anybody remember the story?

#12 ::: PiscusFiche ::: (view all by) ::: August 21, 2003, 06:12 PM:

Also, spraypainted graffiti, if done properly and not OVER other artwork like murals, can be a really cool art form which lends a certain je nais se quoi to urban areas. I think some forms should be encouraged a bit more overtly. Hate tagging though. Tagging bites. It's sloppy graffiti.

#13 ::: Lara Beaton ::: (view all by) ::: August 21, 2003, 06:13 PM:

Much as I agree with your desire for De Beers to start circling the drain, I can already see the slogans:

"Real love deserves real diamonds."

#14 ::: Jeremy Leader ::: (view all by) ::: August 21, 2003, 06:15 PM:

Seems to me that 20 years or so back, there was some attention paid to "Silicon-On-Sapphire" chips, where the sapphire substrate acted as a heat sink. I don't know if there's a better fit, chemically and crystallographically, between silicon and sapphire than between silicon and diamond.

#15 ::: Jeremy Leader ::: (view all by) ::: August 21, 2003, 06:17 PM:

D'oh, the appropriately named Stuart Dimond beat me to it, with a much more detailed discussion.

#16 ::: FranW ::: (view all by) ::: August 21, 2003, 06:21 PM:


Dunno about diamonds, but if you'd like an emerald engagement ring, you can get the lab grown ones here:

#17 ::: Isabeau ::: (view all by) ::: August 21, 2003, 06:37 PM:

I want a diamond chandelier. Cheap materials let you think big.

#18 ::: Carlos ::: (view all by) ::: August 21, 2003, 06:44 PM:

Teresa, you might be interested in Robert Hazen's _The Diamond Makers_ (if you haven't already read it), which covers the overt technical history up to 1995 or so.

Diamond manufacture has attracted some odd folks over the years ("let's make a diamond out of peanut butter!" IMS, it came out green because of the nitrogen) and I am glad to see that the current crop is no exception.


#19 ::: Jordin Kare ::: (view all by) ::: August 21, 2003, 07:09 PM:

>An engagement computer sounds nice though. And a little titanium or platinum ring to go with it.

So what you want is a computer with a token ring?

#20 ::: clew ::: (view all by) ::: August 21, 2003, 07:10 PM:

"Make love. Make diamonds."

#21 ::: Debra Doyle ::: (view all by) ::: August 21, 2003, 07:51 PM:

The truth of the matter is that diamonds are tacky. The small ones look like rhinestones and the big ones look like glass doorknobs.

#22 ::: Graydon ::: (view all by) ::: August 21, 2003, 08:02 PM:

Which is where cheap artificial diamond will really hurt DeBeers.

It's hard to keep something romantic if it's the same stuff as the lining of your refrigerator.

Even if it's just your PDA/smartphone screen and the lenses in your eyeglasses, it gets tricky; if the stuff gets everywhere, and it's used for architectural detailing, statuary, windows, and sewer pipe linings, I'd say it was probably hopeless.

I also think big carved diamond pieces -- broaches and napkin rings and the like -- will probably catch on, further harming the DeBeers brand.

#23 ::: Jonathan Edelstein ::: (view all by) ::: August 21, 2003, 08:05 PM:

The downside of DeBeers going into the tank, of course, is that the economies of Botswana (diamonds about 80 percent of export earnings) and Namibia (~40 percent) will go with it.

#24 ::: Xopher ::: (view all by) ::: August 21, 2003, 10:49 PM:

Hmm. Re DeBeers getting governments to ban the new artificial diamonds...isn't that what Patrick once called "the law protecting a business model"? As in, it won't?

I also hope DeBeers goes down, now that I know something about them.

#25 ::: Xopher ::: (view all by) ::: August 21, 2003, 10:53 PM:

Jonathan, (and I trust your last name isn't a nom-de-blog created for this discussion) a good part of Columbia's economy is coca leaf. Afghanistan's primary export used to be heroin IIRC, or maybe just raw opium paste. Botswana and Namibia can either a) get help converting their economies, or b) rot in hell, depending on whether they've been working the mines under humane conditions (a) or, at the other extreme, working children to death in forced labor (b).

#26 ::: Teresa Nielsen Hayden ::: (view all by) ::: August 21, 2003, 11:43 PM:

Jordin: Ow ow ow ow ow!

Chris, JE's real as far as I know.

Both of you: On the one hand, there'll be wrenching economic dislocations in Botswana and Namibia. That's a lot of suffering to wish on countries that already have a surfeit of it. On the other hand, the diamonds would eventually have run out anyway. I'm with Chris: start giving them help now to convert their economies to something else.

Debra, IMO a well-cut diamond is a pretty thing, especially if you shine a laser into the middle of it: sparklies everywhere!

Carlos, I have to say that making a diamond out of peanut butter sounds like something Jon Singer would do.

Fran, that is a cool site.

I figure that if someone thinks a less-expensive synthetic stone can't mean true love, their sweetie can make up the difference by setting fire to an equivalent amount of money while they watch.

#27 ::: Cassandra ::: (view all by) ::: August 22, 2003, 12:11 AM:

Even if this process doesn't catch on, you still have those "LifeGem" things.

Though I think it would be a bit scary to scratch graffitti with grandpa.

#28 ::: Madeline ::: (view all by) ::: August 22, 2003, 12:14 AM:

I remember the story about synthetic diamonds that Carlos remembers, the one with the peanut butter making a pretty shade of green diamond due to the nitrogen in the peanuts... I saw it perhaps a decade ago, in "Smithsonian"? It discussed the "get something that contains carbon and smash it into a diamond" method of synthetic diamond making.

My thought at the time was, "Aha! If they can smash a dollop of peanut butter into a diamond, how long until they can make a departed human into a diamond?"

There's the way to get around the problem that it's more expensive to make such a diamond than to mine a similar one. How many people would choose to become something valuable and enduring after their deaths, instead of just lying in a coffin, or in a pile of ash? One's descendents could take the mess of small diamonds resulting, cut and facet them, and set them into a necklace to become a family heirloom...

Would different organs become different colored diamonds? It would be a modern version of canopic jars! Perhaps for people going for the slightly less expensive route, just their heart could become a diamond, and the rest could be donated to medicine or science...

My apologies for getting macabre. I think the neatness of diamonds lies in the way they're so elemental.

#29 ::: Madeline ::: (view all by) ::: August 22, 2003, 12:27 AM:

Oh! Well, there ya go. Thanks, Cassandra! I guess that's less messy... LifeGem uses ashes to make diamonds. Miss out on the color, alas, but then again, perhaps humans turn boring colors.

The process-describing page above does have a fascinating graphic showing what percentage of each state opts for cremation. I wonder about the whys of the differences?

#30 ::: Paula Lieberman ::: (view all by) ::: August 22, 2003, 02:57 AM:


Diamond heatsinks for semiconductors have been around since at least the late 1970s. Companies [Kopin?] have been playing with diamonds as semiconductors for years, actually, but the last time I was seriously looking at such technology, it was going nowhere slowly, because of Installed Base issues.

Silicon on sapphire was a hot technology being pushed atthe end of the '70s, for its promise of higher speed, and its compatibility with silicon processing technology, which was/is what the chip industry is almost entirely invested in.

The last time I was paying serious attention a "wafer fab lab" -- large scale semiconductor plant -- cost in the vicinty of $1,000,000,000 to build. Intel spends soemthing like 20 or 25 percent of its entire revenues annually, on R&D and plant refurbishment -- the chips lines aren't than different from consumer electronics production lines, in terms of how long they're up before they get torn apart and rebuilt. In the case of consumer electronics, it's because the product cycles are very short, with product superceding occurring very rapidly. In the case of chips plants, there are a few different reasons -- one of them is the product cycles, the other is that some really nasty stuff and some really corrosive processes, get used, meaning that the equipment gets a lot of hard use and -has- to be replaces, if you want your yield rates and quality to stay at a profitable level and the devices resulting to not make the customers annoyed (quality drops and volume drops, make for very unhappy customers).

Getting back to that billion dollar plant that you have to spend a fortune on to keep running however -- the billion dollars includes all sorts of wafter processing equipment, robotic stuff made a few miles from where I;m sitting that moves cut wafers around for processing, X-ray lithography equipment, silicon processing and wafer growing equipment, big tanks for chemicals, doping equipment, etc. etc. etc., and all of it expensive and designed specifically for working with silicon. Change the materials, and you not only have to replicate that billion dollars of investment in different technologies, but you don't get the economies of scale and experience and installed base, that exist dealing with silicon.

There have been attempts over the past 40 years to use other materials -- twenty years ago a coworker commented to me, "Gallium arsenide has been the material [for semiconductors] for the future for the past 35 years." A few companies were making microwave and milimeter wave and other microcircuitry device chips out of GaAs, however, the problems includes arsenide gas being poisonous and people having died in industrial accidents in plants making GaAs microcircuits, and low volume making the equipment to produce GaAs chips very expensive.

Some companies have played around with Indium Phosphide for semiconductors -- nice performance, but the same installed base issues. IBM and Motorola doing copper interconnect, was not a huge change -- it still involved silicon semiconductor and procesing technology, tweaked a bit by what where at the time two of the biggest companies in the semiconductor industry, with their own pilot lines and IBM spending a lot of R&D at the cutting edge, which the likes of Taiwan-whatever (TMSC or TSMC, which I think is the largest chip foundry facilities in the world) weren't tending to do on their own.

SOI -- silicion on sapphire, CMOS-SOS -- Silicon on sapphire, same thing -- still involve- silicon. Then there were some exotics forms of silicon, and those were still -silicon-.

The thing about silicon, is that is a huge,massive industry, with huge fortunes -- I'd guess trillions of dollars -- of investment in installed base, equipment, knowledge, processing technology, tanks, distribution of raw materials,tools to make the tools to make the tools, facilities, chip packaging, performance characterizations, "integration" meaning that nearly all the circuit boards out there and other component, heat sinks, fans, etc. etc. etc., are all designed around theuse of silicon technology semiconductors and their electrical, thermal, etc., characteristics.

Change materials, and -all- that changes. you have to redesign all your circuit boards, redesign computer and other equipment for the changed electrical and thermal charactistics.... '
silicon chips have certain voltage ranges and requirements, I don;t know what the lastest computer chips use as supply voltage -- 3.3 volts? Less? It had been 5 volts, and before than 12 volts... but it's critical to board design and equipment desint, "what's the supply voltage?" And the power supplies - -more billions of dollars of investment in tools, equipment, production facilities, knowledge, familiarity, repair tools -- oh, I left out all the equipment that puts components on circuit boards. Talk about impressive equipment -- watching wave soldering systems and robotic equipment that grabs semiconductor chips, puts them on boards, and wave solders the whole thing, are extremely impressive to see operate. They're designed for working with the various types of "packaging" that silicon chips go into. Change from silicon and the packaging changes, and you have to start from group up to design/produce equipment for the packages the the diamond chips would go into, with mechanical, thermal, and electrical characteristics suitable for the voltages, charge, thermal considerations, and structural integrity of diamond as opposed to silicon, and then the interconnections -- if without with pinouts, the pins have to be of compatible materials electrically thermally mechanically etc to not come apart with heat changes, shock and vibration, etc., and have constant performance over the specificed environmental conditions of heat, humidity, pressure, power on;off, etc.

Then the equipment to place the chips on boards has to be optimized for diamond chips and the packages they go in, and the connections on the circuit board made for it....

current connector tecnolgoes include PGA -- Pin Grid Array -- that's where here are all those pins sticking out of e.g. a Pentium microprocessor, BGA -- Bump Grid Array (or Ball Grid Array) which instead of having the chip carrier be something that you stick all the pins in, it's optimized for permanent wave solder and there are little bumps instead of pins, and the wave soldering system permanently solders the chip to the circuit board - it;s less expensive to manufacture boards with, because there are fewer, smaller parts with less material -- bumps take up less space and materials than pins- -and flat packs where there are tiny connectors out the sides, or tiny flat metal areas on the bottom or top (I don't remember much about "flip chips"... IBM used that for some PowerPC chips) of the flat square at bordering the edges of the chip

Anyway, there are these trillions of dollars of investment in All About Silicon. Diamonds have really nice thermal characteristics. They have really nice electrical characteristics for being semiconductors went doped, too. But 40+ years of heavy investment in silicon semiconductor processing, decades of concentration of silicion for microprocessors and other chips, in processing, tools, handing equipment, incredibly high volume production (those billion dollar plants, when operating at volume production, create hundreds of millions of dollars of product a year -- they -have- to, because they cost most of that every year in operations and refurbishment costs!) means that anything else, just isn't economical to pursue in mass production chip making. Therehave been small compared to silicon commercial endeavors and investments in diamond and other non-silicon chip technologies, but the installed base not only just isn't there, but the economics mitigate against is -- if it's somethat the silicon -can't- do, for e.g. power/frequency reasons, then the other materials will get used, because if silicon can't do it, silicon can't do it, and you pay the price for using materials that there isn't the high volume economy of scale installed base and infrastructure for. And you put up with lousy yield rates, and custom equipment from specialty suppliers doing oddball customer work, and because you don't have the vast array of people and experience and researchers and literature and infrastructure that silicon has, you don't have the same chip "densities" in terms of the numberof transistors and such -- those come about because of the huge investment in silicon, and the mass demand for it.

A company that makes multiple lithographic systems, spread the research and development costs for making that type of system, across all those multiple systems, and that drops the cost. Plus, they can make larger orders of parts and raw materials, which further drops the cost, and invest in automated tools to build equipment, meaning greater precision and accuracy and repeatability, at lower cost.

If the Hubble had been something that ten or dozens were made, rather than a one-off, the mirror problem either wouldn't have occured, or the bad mirror could have been written off as a teething problem in development... if you're making ten of something, you can afford spoilage, and you can set up automated systems that once you have a good one, you replicate what your did to all the others. One-offs, though, have quirks, and are -extremely- expensive, because automation isn;t worth bothering with for a one-off, you're not going to make duplicates, so why invest in the time and effort for tooling up for replication?!

Diamond crystals have been grown in labs for decades -- but without an industry to consume large diamond wafers, no was going to invest the resources necessary for -large- diamond wafer processing. No one was going to invest the resources necessary for lithography tuned to high density production of transistors for diamond chips -- silicon's had some pretty sophisticated gometries of "stacked" transistors and "trench" geometries, all controlled by the lithography systems and processing technologies, at very very tiny sizes. For diamond processing, the lithography systems might have to use different electromagnetic frequencies, different dopants would be necessary, different chemical processing for treating, etc. etc. etc.

I've forgotten the number of wafers a single silicon crystal can be cut into -- and I forget the financial exposure than companies face if they had to do emergency shutdowns, which destroys the value of all the chips that could have been produced from the crystals and wafers that were in process at the time of the plant emergency shutdown -- but it;s a number of dollars which is extremely large.

Before a viable diamond semiconductor industry could emerge, diamond tech would have to provide some substantial, -required- improvement, in volume, over what silicon provides. And there would be a huge fortune in investment required, to produce the infrasture of tools, software, diamond equivalent of "silicon compilers" -- there is an array of software tools for designing semiconductor chips, used for designing microprocessors, microcontrollers, analog to digital converters, digital to analog converters, audio chips, etc. etc. etc. All of those tools, are for silicon. There are "standard cell design libraries" that each chip maker has, for customized design for their customers to pick and chose and use.... all that stuff represents a huge -sunk- investment, and for diamong tech to be mass volume available, all that stuff would have to -suddenly- appear for diamond, with people experienced in using in, with equipment for doing say 300 mm wafer processing....

Economically, the same factors that got the US steel industry bogged down apply to silicon -- only there's even -more- investment in silicon infrastruture. The fact that Europe and Japan's WWII and earlier era steel mills were all destroyed, meant they weren;'t invested in the old equipment and facilities and technology and could modernize without having to deal with Installed Base -- the installed base wasn't there anymore, so they could use the latest and greatest and exploit all the ideas and concepts and come up with new machinery.... it took years, but it blew the US steel industry away when it got running full-up.


Believe or not, I -like- the idea of diamond semi conductor tech -- it would be SO much better in terms of thermal characteristics, performance, and energy efficiency, that silicon. I've -wanted- it for years... but the reality is that "better is the enemy of good enough" and installed base and return on investment and sunk costs and experience and expertise and infrastructure, all keep the investment in silicon at the leading edge of -affordable- tech and infrastructure and performance. Diamond -could- way outperform silicon -- but the installed base, means that it;s just not feasible to convert to diamond. The costs and time delays are just too great, and the tools and knowledge and experience and learning curve aren't there, and the investment costs are too high, and the return on investment, too low.

It's like television -- NTSC and PAL are kludges come up with in the first half of the 20th century, they were what could be done with first half of the 20th century technology. We're still using them today, because several hundred million TV sets in people's homes and places of work, broadcast equipment and video cameras, and knowledge, all work to prevent change which requires junking the investments people hve in their TV sets, their camcorders, VCRs, videotapes, etc., and going to stuff that costs ten times as much.... it's coming, and painfully -- and with backward compatiblity mandated for some number of years -- but it's taken decades, and it going to take another decade -- or more -- to transitiion.

And TV factories are a lot less expensive to build than high volume semiconductor processing faciliites.

#31 ::: Paula Lieberman ::: (view all by) ::: August 22, 2003, 03:36 AM:

"General Electric managed to do this in 1954 by using a 400-ton press to crush the hell out of carbon. GE's machine economically produced diamond dust for industrial uses, and by the early 1970s the company had even managed to manufacture stones as large as 2 carats."

And was sued by Russell Seitz [who used to go to SF conventions] for patent infringement....

"His company, Spectrum Technology, pioneered the commercialization of gallium arsenide wafers, the microchip substrate that succeeded silicon and allowed cell phones to become smaller and handle more bandwidth."

Oh, puh-lease, GaAs has been around for decades, and was in use at those frequencies for decades, and some exotic forms of silicon chips are preferred because they're less expensive. There was a Japanese company that made GaAs chips and established a plant back in the 1980s on the US west coast, which started selling to the US DoD and finally sold the facility to [the old] HP, because the Japanese managers weren;t allowed into the facility due to the DoD contract work it was doing.... also, GE and TRW were among the other companies making GaAs chips for the US government for communications equipment use.

#32 ::: Larry Lurex ::: (view all by) ::: August 22, 2003, 06:17 AM:

As the author of a book on materials science once wrote, what would we be able to do differently if we had access to large quantities of diamond?

It isn't as though flint, with some similar properties of hardness and extreme cheapness, has set the modern world on fire. Mostly it is used for gravel on people's driveways.

Diamond has some interesting optical properties but it is probably the heat conductivity that will be most in demand - as clear as glass, yet as conductive as metal. There should be a few applications for that.

DeBeers - they aren't American are they? And cheap diamond wouldn't hurt the US economy, would it? No wonder this idea is not being killed at birth.

#33 ::: Carlos ::: (view all by) ::: August 22, 2003, 06:34 AM:

Xopher wrote:

"Jonathan, (and I trust your last name isn't a nom-de-blog created for this discussion)"

... Jonathan Edelstein is a hell of a guy, who happens to know quite a bit about African politics (among other things), thank you very much.

I am pretty sure he is rather proud of his last name, too.

"a good part of Columbia's economy is coca leaf. Afghanistan's primary export used to be heroin IIRC, or maybe just raw opium paste. Botswana and Namibia can either a) get help converting their economies, or b) rot in hell, depending on whether they've been working the mines under humane conditions (a) or, at the other extreme, working children to death in forced labor (b)."

Xopher (and I could make a snarky comment about noms de net here, I suppose), Botswana is one of the *nicer* places in sub-Saharan Africa. Good government, good labor practices; they've done well *despite* the slings and arrows that history, in the form of the British [1], the South Africans, and the AIDS virus, has thrown at them. They've invested the money from the diamond profits intelligently, and they're rapidly becoming one of Africa's few First World economies.

Unfortunately, the next industry down on the export list is ranching, and the area just doesn't have a very large internal market for the products of a transition economy. Hard to get export-led growth selling cans of bully beef.

So, yes, without their diamond exports, they will be screwed, and one of the few reasonably bright spots on that continent will become rather dimmer. With skill and luck (help? hollow laugh) Botswana will probably weather this development, but it's not inevitable.

Comparing them to current Afghanistan or Colombia shows a deep and rather arrogant ignorance about the facts of the situation, and your other comments display a rather unpleasant Schadenfreude.



[1] One of my favorite Botswana stories involves the time the Royal Navy decided to stage a punitive expedition against the native government for daring to sentence a white man to be flogged, the standard native punishment for extreme drunk and disorderly conduct; the silly folks had the notion of equal treatment under the law or something like that. Yes, Botswana is land-locked. They took the train.

#34 ::: Graydon ::: (view all by) ::: August 22, 2003, 06:40 AM:

Paula --
You're right about the investment, but the installed base has -- unlike steel plants -- very short lifetimes, somwhere around a decade before what was top of the line falls of embedded use, and there is a continuing performance drive. Once silicon starts to dead end for performance is when diamond is likely to get a try at things.

#35 ::: Kip W ::: (view all by) ::: August 22, 2003, 06:48 AM:

"Make me the happiest man in the world, darling. Would you... wear my grandpa?"

#36 ::: Robert L ::: (view all by) ::: August 22, 2003, 07:25 AM:

Teresa, next time youre walking around, take a look at graffiti tags on windows in your neighborhood (I'm sure there must be some). then look at them closely. I bet you'll see the same thing I see around here: a lot of graffiti on glass are now done with a glass-etching compund (hydrofluoric acid? don't know what's in it, but it etches right in). the only way to get rid of it is to get a new pane of glass.

#37 ::: CEP ::: (view all by) ::: August 22, 2003, 09:05 AM:

The Wired article's view of boron (or other row-two atom) doping plays awfully fast and loose with the physics of stable crystal structure. One of the things that it neglects to mention is that most of the publicly available information on "diamond-based" semiconductors remarks that fragility has turned out to be a problem as they approached electrical usability. Using atoms beyond row 2 would only increase that fragility.

Despite a diligent search, I have not yet found a satisfactory proposal for how circuits might be traced on a theoretically appropriate diamond substrate in industrial quantities.

#38 ::: Kate Nepveu ::: (view all by) ::: August 22, 2003, 09:36 AM:

IMO a well-cut diamond is a pretty thing, especially if you shine a laser into the middle of it: sparklies everywhere!

Or sunlight at the right angle.

Discussions like this tend to make me feel very guilty about my engagement ring, but then I go "oooh, shiny" and feel a little better. (Just a little.)

#39 ::: Kellie ::: (view all by) ::: August 22, 2003, 11:03 AM:

I remember the odd looks I got when I went hunting for a good quality .25 or .3 carat diamond for my engagement ring. It's like no one could understand that I didn't want some gargantuan rock weighting my hand down. It still sparkles in sunlight and laserlight.

DeBeers is Dutch or South African, I think. And I remember hearing somewhere that any of the top execs of the company would be arrested on sight should they set foot in the US, due to their "conflict diamond" buying and other tactics.

#40 ::: Greg Morrow ::: (view all by) ::: August 22, 2003, 11:42 AM:

My pal Jason Modisette figured out a while back why people prefer read gemstones to artificial ones:

The Love Pump Theory of Diamonds, or Why Size Doesn't Matter

#41 ::: LNHammer ::: (view all by) ::: August 22, 2003, 12:03 PM:

Speaking of tagging, last week I saw a tag labeled © 2003. I was bemused.


#42 ::: Paula Lieberman ::: (view all by) ::: August 22, 2003, 12:36 PM:

Graydon wrote:

"Paula --
"You're right about the investment, but the installed base has -- unlike steel plants -- very short lifetimes, somwhere around a decade before what was top of the line falls of embedded use, and there is a continuing performance drive. Once silicon starts to dead end for performance is when diamond is likely to get a try at things."

That's extremely misleading, though -- the companies making the the chemical vapor s[omething or other] deposition equipment, the ion implantation equipmment, the chemical processing systems, the wafer handling robotics, the chemical tanks, the reactors growing the crystals, the electronics controls systems, the sensor equipment and signaling and control systems, the lithography equipment, the packaging technology and equipment, the interconnect technology, the circuit boards, wavesoldering systems, all have sunken investements of large fortunes of technology, art, expertise, people training.

I'm blanking on company names right now beyond the likes of Nikon and Teradyne and KLsomething or other who make production and testing equipment, then there are the chipmakers IBM, Motorola, Hitachi, Sony, Mitsubishi, TMSC [or is that TSMC...], Analog Devices [which tends to use foundrys, but all its design expertise is in silicon], Intel, AMD, National Semiconductor, Fairchild,Philips, Samsung, Texas Instruments, Burr-Brown, etc. etc. etc. all of whom have decades of recipes, training, experience, process technology, business relationships, and facilities, and knowledge of working with silicon.

That stuff is -not- directly transferrable to working with diamond. All the handling systems and cutting and polishing, for example, have assumptions designed into them about the materials properties of silicon wafers and silicon chips, regards the level of force that can be used moving them around that won't cause warping, stress fractures, etc. Change the material from silicon wafers with their particular crystallographic characteristics, and you not only have to do a complete redesign, you have to do complete materials characteristics and testing and development of formal and commonsense knowledge of how the handle the new material.

Look at stuff like bread and houses -- take a crusty loaf of bread and cut slices from it. Take a soft load of bread and cut slices. You can't use the same style of bread cutting for both and get good results -- the crusty bread tends to generate lots of crusty crumbs if not carefully cut in the way best for cutting crusty bread, and you can apply more pressure holding in in place without squashing it. The soft bread, if you apply that level of pressure holding in place, you get a compressed squished loaf and "slices" which ridges and bumps on them. The soft bread has larger, software pieces that come off it as "crumbs" -- the don't tend to be little tiny flaky crummy pieces, they tend to be larger fibrous pieces torn and ripped and squishy. So, you have to use different cutting styles to get nice even slices off crusty, versus soft, bread -- even though both of them are made of the same base ingredients, flour and water and leavening and such!

Now think about houses and buildings -- different construction techniques get used depending on whether you're using wooden framing construction, wood post and beam, logs, brick construction, concrete, steel beams, steel whatever-it's-called-that-I-see-Barnes-and-Nobles-constructed with using metal like 2 X 4s and applying giant panels of some sort to the metal upright pieces, it goes up -very quickly]. Someone who's a skilled bricklayer isn't necessarily competent to put up steel framed buildings or pour concrete, the people who pour concrete, aren't necessarily skilled for doing wooden post and beam construction. The companies that make the concrete and deliver it, aren't competent to make prefab wooden panels usually, or make steel I-beams, or do plastering, etc. etc. etc.

The differences between silicon tech and diamdon tech for semiconductor processing, is somewhat analogous to the difference between building post and beam and brick giant old mill buildings, and making steel and glass and concrete skyscrapers -- the knowledge and supplies and tools are all so very different, and they involve totally different technology, skills, training, suppliers, tools, equipment, and knowledge.

And the equipment for construction, is a lot less expensive, a lot more rugged, and a lot less precision and needing that precision, that the equipment used for semicondustor processing. The equipment required to put up a skyscraper doesn't cost hundreds of millions of dollars and have hundreds of millions of dollars annual maintenance. The building might cost a hundred million to put up, but that's for the materials and labor mostly, with the equipment costs being a much smaller fraction of the costs -- and it doesn't cost hundreds of millions of dollars to keep operating every year.

The fact that semiconductor product cycles tend to be short, doesn't mean that the underlying technology and equipment changes substantially every year: 300 mm wafers are the result of scaling up from 200 mm, which was scaled up from smaller sizes. The technology changes are process improvements and incremental, the chemical processing mostly has been incremental and process improvement. It hasn't been wholesale tech base and infrastructure changeout, it's been refinement of processes, improved control systems, improved optics, sometimes changing the optical frequency -- but most of it has been incremental, gradual, planned tweaks and upgrades an enhancements to -existing- installed base equipment and processes. It's like replacing one internal combustion car or truck, with a bigger one. There may be a new engine, but they take the same types of gasoline, the same or similar types of oil, the parts are different, but they're made off the same types of car producton lines -- there are new machines used, but they're mostly tweaks and direct replacements and successors to the machines that were already in use. There are few revolutionary changes.... note how long it's taken to get much in the way of electric vehicles and hybrid gas-electric vehicles to commercial consumer availability. The auto industry's been -playing- with those technologies for decades, but the reality is that the installed base of cars, gas stations, repair garages, dealships, customer familiarity, lack of infrastructure for electric fleet vehicles -- where do you get a charge for you all-electric car on the Interstate system at 2 AM?? --all colluded to suppress commercial development and volume availability. Also, other factors included that the relatively low cost of cars is due to mass volume economy of scale production. A small company which was trying to make electric cars in Massachusetts, had had to buy entire new Geo Metros, and remove the internal combustion engine and replace it with batteries and the electric engine, the company coudn't purchase just car bodies, or have them made economically, in the low volume rate production they were involved with.

There's that whole integration thing-- you need -all- the parts. That old Poul Anderson story about the person thrown back in time, who doesn't have the tools to make the tools and the necessary knowledge and resource available to boostrap up, applies for diamond technology. Silicon is the baseline, and only if there is a mass demand and need for diamond, will the technology come in in any substantial quanity any time or decade soon. The resources haven't been spend on R&D for it, while the likes of IBM and Motorola and Hitachi and Siemens and Sony and Mitsubishi and Intel and AMD etc. etc. are each pouring large fortunes into R&D every year pushing to make silicon faster, lower power, more dense in the number of transistors.... GaAs for example, has 3" wafers, when silicon was at 6" wafers or larger, and at yields that may have approached 50%, when silicon yields were much above 80% -- the bottom line was that silicon cost a -lot- less to produce, and all the supporting stuff for -- boards, packaging, interconnects, etc., was also inexpensive. So, GaAs, continued lagging behind for anythign that could be done with silicon at the applicable performance level, and GaAs only got used where silicon couldnt' provide the performance necessary. It comes down to "how much can you afford and do you need it so much and have the budget, to foot the lead time and development costs and production costs?"

For 99.999 percent of semiconductor users, diamond isn't convenient, affordable, or applicable for them -- it doesn';t have the number of transistors, it doesn't have the infrastructure, and they can get most of the performance they want, with systems based on current state of the art silicon-based technologies.

Example of installed base stuff and incrementalism:

[government website, therefore public domain]

"This ATP project allowed Diamond Semiconductor Group (DSG), a two-person start-up company when it applied to the ATP, to develop a new and better way to implant dopants on large silicon crystal wafers measuring 300 mm or more in diameter, compared with the previous industry standard of 200 mm. Because the area of a 300-mm wafer is 2.25 times that of a 200-mm wafer and some waste always occurs at wafer edges, the new approach enables the production of about 2.5 times as many chips from a single wafer as the 200-mm technology can make. The use of DSG's new technology in production equipment makes it possible to lower the cost and improve the quality of computer chips and other integrated circuits.

"A key innovation in the new technology is passing the wafer under a 350-mm-wide ion beam for implantation, rather than scanning the ion beam across the wafer. The broad beam is very stable and therefore highly accurate. The new equipment incorporating this technology is also significantly simpler than earlier machines and so is cheaper to build and maintain and is more reliable. Use of the DSG technology has already improved fabrication quality substantially relative to the existing industrywide standard. It doubled the mean time between failures, which means that on average, failures occur only half as often as with current equipment.

"The DSG technology also lowers fabrication costs by allowing implant equipment to be designed to work on one wafer at a time. Although it seems counter-intuitive, single-wafer processing is actually an advantage. Fewer wafers are lost if equipment fails, compared with current technology. The latter involves clamping 13 to 17 wafers to a large wheel, which then rotates at about 1,200 rpm under the ion beam. One failure may result in 13 to 17 unacceptable wafers. With single-wafer processing, only one wafer would be lost. In addition, single-wafer processing enables ion implantation to be coordinated much better with other fabrication steps, most of which are also performed one wafer at a time."

Note that what's now is using a wide beam, and moving the wafer under the beam, rather than scanning the beam across the wafer. The light frequency hasn't changed, the wafers haven't changed -- yes, they got larger, but that was a planned process improvement for the entire industry--, the handling equipment loses a rotating table device, which is a win -- rotating joints are always likely sources of trouble,and the effects of failure are more restricted than the previous processes resulted in. Again, this is mostly process improvement, not a major tech change, and can be retrofitted into existing lines without huge disruptions, or relatively easily incorporated in new producton lines. And it's relatively low risk, and integrates into existing product lines of Varian [see below].

"Worldwide sales of ion implanters total $1 billion to $1.2 billion per year, and Varian has 40 percent to 50 percent of the market. Most of the equipment currently sold is for 200-mm wafers, and Varian was the first to market equipment that handles 300-mm wafers. Over the next five years, industry analysts say, the majority of implanters sold will be for 300-mm wafers. All 300-mm-wafer ion implanters currently manufactured by Varian include the DSG technology, and those produced in the future are expected to, as well."

So, what's going on there, is that Varian, going to a new generation of product line, to work on 300 mm wafers, make a tech change based on this new company's process improvement technology for ion implantation.

"IBM announces world's fastest silicon-based transistor
IBM has built the world's fastest silicon-based transistor, a basic building block used to make microchips.

"IBM expects the new transistor will result in communications chips at speeds of 100 GigaHertz (GHz) within two years -- five times faster and four years sooner than recently-announced competitive approaches.

"The transistor uses a modified design and IBM's silicon germanium (SiGe) technology to reach speeds of 210 GHz while drawing just a milliamp of electrical current. This represents an 80 percent performance improvement and a 50 percent reduction in power consumption over current designs...."

IBM's been playing with silicon germanium for a while. Note that there's silicon involved, which means evolution from existing IBM silicon processing technologies and expertise.... that's one of the reasons IBM chose to do R&D on SiGe, because it could exploit IBM's many billions of dollars of sunken investment in patents, processing facilities, knowledge, cognizant personel, relationships with the likes of Eaton, Varian, Nikon, etc, which make silicon semiconductor processing systems equipment.

has some gory details evidential of what I said about compatibility.... "The integration issues for low-k films are largely the result of trying to find a low-k dielectric film with electrical, thermal and mechanical properties comparable to those of SiO2 (k=3.9-4.3). [what silicon chips are make of.... high purity quartz, essentially. it;s actually silicon dioxide, not elemental silicon. That usually gets not mentioned and taken for granted.] Important metrics include film hardness, film strength (modulus), film stability throughout processing, glass transition temperature, and low keff. A low keff represents the "holy grail" of the intermetal dielectric stack, largely determining the line-to-line capacitive coupling and RC delay of the interconnect....

"A viable low-k material must be compatible with dual-damascene lithography, etching, stripping and cleaning processes, and especially polishing (CMP) and device packaging (including wire bonding) methods. Additional films, including etch stop layers, hard masks (for patterning assistance) and capping layers (to protect the low-k film during CMP) strongly influence keff. Early integrators of low-k films, which today offer keff values of about 3.0, encountered an insidious problem of photoresist poisoning, which required fundamental and empirical research to solve."

And that;'s just for a -coating- for the semiconductor, it's not even the semiconductor material!


[note -- why am I poking around doing a search at ? Answer -- IBM is the megaton gorilla of the semiconductor industry, anyone who's looking for more than a niche market in that world, is going to have to deal with IBM and its patent portfolio, which is the largest one in industry in the entire world.... don't bet against IBM having a relevant patent, or two, or three, or a few hundred, or thousand....

IBM patent awarded in USA (1998?)...

5,674,355 A. Grill, D. C. Edelstein, J. R.
Paraszczak, S. A. Cohen, and V. V. Patel Diamond-like carbon for use in VLSI and ULSI interconnect systems

"Diamond-Like Carbon Prepared by High-Density Plasma, J. C. Sanchez-Lopez (c9cole Central Lyon, UMR 5513, F-69131 c9cully, France) et al., Diamond and Related Materials 9, No. 3966, 63896642 (2000).

"Adsorption of Fluorine on Bare, Hydrogen-Covered and Hydrocarbon-Covered Diamond C(111) Surfaces, T. Yamada (Waseda University, Shinjuku Ku, 2-8-26 Nishiwaseda, Tokyo 1690051, Japan) et al., Japanese Journal of Applied Physics Part 1 39, No. 4A, 1826961834 (2000).

[this one smells like semiconductor processing technology considerations....]

#43 ::: Stuart Dimond ::: (view all by) ::: August 22, 2003, 12:49 PM:

Paula, you drastically underestimate the flexibility of the design tools and processes used in the electronics industry. Yes there would be some changes of specific chemical processes. The effort to get down to 90 nanometers in silicon has just as many thorny problems to be solved. State of the art wafer fabs are now in 2 to 3 billion dollar range.

CEP, photolithography and ion beam implantation, the same way they do it on silicon just different dopants. Starting with flawless crystal structure will make a lot of difference in final strength.

I'm trying to avoid doing a data dump. I worked for a semiconductor company and I assume the rest of the world doesn't find this as interesting as I do.

#44 ::: Graydon ::: (view all by) ::: August 22, 2003, 01:53 PM:

Paula -
Yes, there's a lot of invested skills base.

If, in order to stay in the game ten years from now, they have to go from silicon to diamond, they're going to do it anyway; it's not like there's ever been any sign of unwillingness to write technologies off in the chip industry!

They've already chucked several lithography processes and mask technologies to get here; the shift down to the next smaller nanometer count involves whole new fab lines currently; going down to 0.65 is a big deal that way.

It's still happening just as fast as it can be arranged.

#45 ::: Jesse ::: (view all by) ::: August 22, 2003, 02:11 PM:

Here's a question I've been kicking around -- if everything is made of diamond (or lined with diamond, etc.), how do we get rid of it? Biological building materials decompose over time. Plastics can be crushed or burned, and sometimes even recycled. Stone can be crushed. What do you do with a diamond refrigerator liner? It seems like a good idea to not develop and use materials technologies that can't be recycled, or at least easily disposed of.

Speaking as someone who has shopped for diamond jewelry, damn, those little sparkly rocks are expensive. I can't decide if my girlfriend would like more and bigger ones for the money, or if the money is part of the appeal. I'll have to ask her.

#46 ::: Carlos ::: (view all by) ::: August 22, 2003, 02:43 PM:

Jesse wrote:

"Here's a question I've been kicking around -- if everything is made of diamond (or lined with diamond, etc.), how do we get rid of it? Biological building materials decompose over time. Plastics can be crushed or burned, and sometimes even recycled. Stone can be crushed. What do you do with a diamond refrigerator liner?"

Diamonds will burn, although it can be difficult to get a self-sustaining reaction. Faraday burnt one away using an early blowtorch. I'd think that an oxygen furnace as used in steelmaking would work just fine.

Diamonds would burn more cleanly than anthracite coal, though I think we're some years away from waste diamond as a fuel source.


#47 ::: Graydon ::: (view all by) ::: August 22, 2003, 03:07 PM:

If you need very pure carbon plasma to make the diamonds with in the first place, you can take the old diamond refrigerator liners and recycle them.

#48 ::: Jordin Kare ::: (view all by) ::: August 22, 2003, 04:16 PM:

>if everything is made of diamond (or lined with diamond, etc.), how do we get rid of it?

You don't *want* to get rid of it! All that non-biodegradable carbon reduces the atmospheric CO2 load and thus delays global warming. Just break it up into small pieces (diamond does fracture easily) and use it for landfill, or as aggregate in concrete...

(And now I'm reminded of one of my favorite Campbellian SF stories, though I can't at the moment place the title or author [someone here will promptly provide it, no doubt]: the spaceship crew that gets stranded on a planet covered with diamonds, and has to invent antigravity technology to get away before the planet falls into its sun; the last line is the memorable, "We brought the planet with us.")

#49 ::: Xopher ::: (view all by) ::: August 22, 2003, 05:58 PM:

Carlos: Comparing them to current Afghanistan or Colombia shows a deep and rather arrogant ignorance about the facts of the situation, and your other comments display a rather unpleasant Schadenfreude.

I admitted my ignorance. I don't see why you characterize it as arrogant. And I don't see why you think that people who are trying their best to get by are worse in Columbia or Afghanistan. Yes, those countries have given us a lot of trouble and maltreated their own people. But it's the maltreatED that we're talking about here. From your account (and thanks for the information) they seem like a good candidate for help converting. And yes, the US HAS done that in the past, notably with opium.

And schadenfreude? Explain where I display that nasty emotion. I don't think so. I might be being flip about the whole thing, but this isn't a crisis yet; I'm not giggling over pictures of starving babies or anything. If you're talking about de Beers, yep. Schadenfreude R Us there. They sound like the very nastiest kind of multinational, without which everyone, including the people of Botswana, will be better off.

And I certainly meant no slam on Mr. Edelstein; I was just noting, with pleasure, the appropriateness of his name to the present conversation.

And I'm Christopher Hatton. I just thought everyone knew that. Guess not. But I'm not hiding.

Carlos again: HTH. HAND.

Again, I am insufficiently leet. What's that mean?


#50 ::: Alan Bostick ::: (view all by) ::: August 22, 2003, 07:40 PM:

"Hope This Helps. Have A Nice Day."

#51 ::: Carlos ::: (view all by) ::: August 22, 2003, 09:56 PM:

Xopher wrote:

"I admitted my ignorance. I don't see why you characterize it as arrogant."

You want to dictate policy without knowing the situation, and even worse, knowing that you don't know the situation. That seems to me extremely arrogant.

"And I don't see why you think that people who are trying their best to get by are worse in Columbia or Afghanistan."

You're making a false moral equivalence between *anyone* who does business with De Beers, and people who manufacture opium paste or cocaine.

The government of Botswana is one of the cleanest in Africa. It's cleaner than many European governments. You don't have to believe me; you can check the various Internationals, Transparency, Amnesty, and so on. And the joint venture mines in Botswana are decently run (though not a dream job by any stretch of the imagination).

Snc y sm t nd mrl lssn wth ch fct, you might want to view the Botswana government's deal with De Beers as "gaming the system" or some such. Personally, I think this shows De Beers is _more_ culpable for their actions elsewhere, because it proves they can do better. But they choose not to.

"From your account (and thanks for the information) they seem like a good candidate for help converting. And yes, the US HAS done that in the past, notably with opium."

Xopher, Botswana does not have any product that will bring in the export money the way diamonds do, or employ as many people. Maybe it won't be a complete catastrophe, since they have a much more complex economy now, but Botswana's incredible growth, from being subsistence cattle herders at independence to eastern European levels of wealth today, has been almost completely dependent on diamond profits being invested back into the country.

This is not something an aid program can easily replace. The analogy isn't with opium, it's with old Pittsburgh or Lowell, MA, except the possible downward spiral is worse.

"And schadenfreude? Explain where I display that nasty emotion."

That 'rot in hell' comment. You know, it's still Schadenfreude when you enjoy *bad* people suffering.

"They sound like the very nastiest kind of multinational, without which everyone, including the people of Botswana, will be better off."

I just did a rough back-of-the-envelope calculation, to figure out what GDP per capita in Botswana would have been without the high diamond prices kept by the De Beers cartel over the years. It's about $8000 now. I get maybe $1500.

#52 ::: Elise ::: (view all by) ::: August 23, 2003, 02:18 AM:

The LifeGems sound interesting. Then again, I've always wanted them to make beads out of my leftovers, not gems. Then again again, if diamond beads were inexpensive, this could get really cool.

#53 ::: Xopher ::: (view all by) ::: August 23, 2003, 10:40 AM:

Carlos, I don't want to dictate policy, I wasn't drawing any moral equivalencies, and "rot in hell" was meant to mean I have no sympathy, not that I would rejoice in their suffering. Yes, there's a difference.

That last is irrelevant anyway, since you've explained that Botswana is humane by reasonable standards.

You took my whole post WAY too seriously. But switching into a more serious mode, it's going to be a while before these new diamonds become commercially available; this should give Botswana some time to cushion its fall. I hope they avail themselves of the opportunity. As you point out, it's not clear what they can do, but perhaps better economic minds than mine can come up with a plan; I certainly hope so.

As for de Beers' benefits to Botswana, you'd have a hard time coming up with an evil in history that benefitted no one. Nor could you come up with a technological advance that did not cost someone a livelihood or even a life. Sometimes it's bad overall, sometimes good overall, but there are two sides to everything.

Botswana's economic collapse will be a terrible tragedy. De Beers' consignment to the dustbin of history, as Teresa put it, will be the opposite. That these two things are linked together is the terrible irony of the real world. It's our job to work on minimizing the impact of the tragedies, while enjoying the benefits of the victories.

I worked in 1 WTC. When the towers came down, I lost 19 people who I knew at least by name and sight, some of whom I worked with daily. Emotionally, I became a wreck, and my life pretty well fell apart. I'm recovering now. BUT: as a direct result of that terrible tragedy, I now live and work in New Jersey, have an incredibly short commute, and pay much lower taxes. This is microscopically trivial compared to the tragedy, but it IS a benefit, and I'm going to enjoy it for all it's worth. It feels like I'm pulling something out of the ashes, if you understand me.

Perhaps Botswana, with its in-country expertise in handling diamonds, will become host to the new diamond-manufacturing industry? Or it could use its current prosperity to buy into the high-tech industry. Lowell, MA, which you mention, is where the last TWO big companies I've worked for have their data centers.

#54 ::: Xopher ::: (view all by) ::: August 23, 2003, 10:44 AM:

Elise, I want my long bones made into flutes, Tibetan style, and given to my close friends. Finding somebody to do it will be tough. Picking which friends get them will be tough, too!

#55 ::: Xopher ::: (view all by) ::: August 23, 2003, 10:46 AM:

Oh, wait, Carlos, I see your point about Botswana's GDP. Sorry. No, you're right, de Beers' fall will not benefit them. At least not in the short to medium term.

#56 ::: Paula Lieberman ::: (view all by) ::: August 23, 2003, 02:52 PM:

"Perhaps Botswana, with its in-country expertise in handling diamonds, will become host to the new diamond-manufacturing industry? Or it could use its current prosperity to buy into the high-tech industry. Lowell, MA, which you mention, is where the last TWO big companies I've worked for have their data centers."

Lowell, Massachusetts, the place where the former Wang Towers which cost $80 million to build, was handed over to the bank and sold by the bank for the munificent sum of $500,000 or so a few years ago -- which was a lot less that the value of the communications teleport in the buildings.... Lowell, the place that Readercon left because the one hotel in the center of Lowell, went backrupt from lack of business? Lowell, which has had a depressed economy since at least the end of World War II?! -some- companies used the place as a tech incubator, because it was full of big old empty mill buildings, the ones that hadn't burned down, left empty when the textile industry moved south. Some of the turned into museums -- the American Museum of Textile history moved into one at then end of the 1990s, for example -- that building had been empty for decades, and it's so old that it still has what appears to be an original waterwheel for producing mechanical power for textile machines. Other mill buildings got turned into a National Historic Park, and a state park, and varous museums. Lowell's not exactly what I'd call a center of high technology. It has -some-, but there's a lot more in Lexington, Waltham, Burlington, etc., closer into Boston-- thought there are lots of building for rent/lease/sales all over those place, from all the companies that have shrunk or disappeared completely -- the company that I was employed by two and a half years ago, pulled an Igli.... Basically, Lowell went floating down the canals decades ago, and some companies took advantage of big empty space, in mill buildings that have an unfortunate tendency to be overly combustible (all that lanolin soaked in from wool processing, and other such flammable stuff soaked into the old wooden post and beam construction mill buildings).

#57 ::: Teresa Nielsen Hayden ::: (view all by) ::: August 23, 2003, 03:21 PM:

Less heat more light, Carlos.

#58 ::: Kevin J. Maroney ::: (view all by) ::: August 24, 2003, 02:00 AM:

Current asphalt mixes use a fair amount of glass. It makes the asphalt more reflective, which reduces the heat-absorbtion problem, but it makes the asphalt weaker. I wonder how well diamonds would work instead?

#59 ::: Alan Bostick ::: (view all by) ::: August 24, 2003, 12:17 PM:

Well, that's one way to cure those walking blues.

#61 ::: Adina Adler ::: (view all by) ::: August 24, 2003, 04:08 PM:

Paula, Readercon did not leave Lowell because the hotel went out of business. The hotel was doing just fine when we left.

Kevin, if asphalt became more reflective, I'd worry about what would happen to people walking (and driving!) on it on a sunny day.

#62 ::: Carlos ::: (view all by) ::: August 24, 2003, 10:21 PM:

Apologies, Xopher, Teresa. That came out more harshly than I intended.

(Though next time let me put in a word like 'Ouagadougou' or 'eerie', just to keep future readers guessing.)

Xopher, the phrase I think you want might be 'creative destruction'. The Viennese-Harvard economist Schumpeter came up with it to describe a process of entrepreneurial capitalism that (he thought) drives growth. It has occasionally struck me that it could also work as a spiritual metaphor.

In general, I wonder how many economic transactions it takes to launder a sense of culpability. I suppose an extreme reductionist would say only one, _non olet_. In practice it seems to be around two; most people don't feel OK about buying stuff from a fence, but also don't inquire too deeply about those batteries labelled in Hindi down at the local drug store. By my count it's at least four away from a De Beers agent buying blood diamonds to getting an engagement ring.

Thought experiment that just came to me: consider if De Beers partook more of Hank Scorpio nature than Ernst Stavro Blofeld's. Gullible Western consumers buy diamonds from a shady cartel... and fund happy steady shiny African development.

#63 ::: Steve ::: (view all by) ::: August 25, 2003, 10:01 PM:

For what it's worth, a 1982 Atlantic article that gives a history of the engagement ring and beats the hell out of the idea that diamonds are a good investment. I dug it up when Anil Dash was analyzing the semiotics of De Beers ads back in January.

#64 ::: Teresa Nielsen Hayden ::: (view all by) ::: August 26, 2003, 12:32 AM:

Steve, thank you for linking to that article in the Atlantic. I remembered it and went looking for it when I was writing the original post, and couldn't find it.

Carlos, no problem. This is fascinating.

My problem with "creative destruction" is that when I hear the phrase, I think of what happened when Ron Perelman got hold of a large segment of the comics industry. He very nearly destroyed both Marvel Comics and the existing comics distribution system.

Comics is full of guys who've put years and years of hard work into learning demanding and highly specialized skills. Ron Perelman never studied under Joe Kubert. I doubt he knew more than one-point-five nanosquats about the Marvel or DC continuities. But he leveraged his leverage into enough leverage to grab hold of Marvel, and proceeded to wreck the hopes and livelihoods of half the people in the industry. Maybe more than half.

I also think of SF writer James White's politely expressed but profound dismay when a system he'd accepted and worked with all his professional life (modest advances on individual books, but steadily increasing cumulative royalties from all the titles in one's backlist) collapsed, mutated, collapsed some more, and left him with no publisher and a long string of out-of-print titles in his old age.

As I later said to James in one of our exchanges of editorial correspondence, the invisible hand of the marketplace isn't to human scale. "Creative destruction" would be one way to describe what happened to the mass market paperback book distribution system, and no doubt it'll eventually shake itself out into some new form; but just the initial repercussions of that shakeout troubled the later years of an old man I was very fond of.

(The story improves. When I heard that his previous publisher of many years had dropped him, I picked up James and his ongoing series, and we happily published him for the rest of his life and beyond. But for a while there, he was staring into the gap.)

The more I watch how people build their lives, the less I like large upheavals. All of us are forever trying to spin out some modest little web of opportunity and possibility in the gaps and angles formed by the much larger economic entities around us. And when the lords of this earth bring their creative destruction down upon us, we weep for the wreck of our small schemes.

Some years back I came upon a little heap of old possessions that had been left next to a streetcorner garbage basket on the Upper West Side. It definitely had the air of someone cleaning out a long-inhabited apartment.

Naturally, I rummaged through it. What I found, bundled up together as they'd been stored, were an old newspaper, a language instruction book for teaching yourself Dutch, and a complete set of 78 rpm records to go with the book. They were all from 1929. The book was an artifact from another age, full of conversations about lifestyles involving cooks, nursemaids, and first-class accommodations on transatlantic liners. The newspaper -- one of the old second-string New York papers, I forget which one -- was from the day before the great stock market crash of 1929. Like the instruction book, it was an artifact from a vanished world. The 78s looked like they'd never been used.

I've always wondered what the deal was. It seemed to me that something had come to an end there, and that it had been important enough to someone that they'd preserved its remains intact for the next fifty-odd years.

Any time you have chaos and destruction, you get stories like that. They don't get told much, because there's not a lot of punch in a story about things not happening; and also because one of the hardest things to remember is all the possiblities and contingencies you had spinning in the air before large events precluded their existence.

Wired sees the possibility of faster chips and the wreck of De Beers. Paula sees the difficulties of such a basic changeover in the extensive and much-invested-in silicon chip industry. Carlos sees the potential damage to the fragile complex hopeful economy of Botswana. It's all true, it's all real, and it all might never happen.

I want to believe that the stories we tell about material culture matter as much (or almost as much) as the material culture itself. But then, I would. My lifelihood involves the stories we tell about material culture. Of course I think they're important.

#65 ::: Stefan Jones ::: (view all by) ::: August 26, 2003, 01:26 AM:

"a little heap of old possessions"

Imagine bundling it up again, and giving it to the parents of a child born next year, with instructions to open it up on his or her 25th birthday.

#66 ::: Christina Schulman ::: (view all by) ::: August 26, 2003, 03:35 AM:

What I found, bundled up together as they'd been stored, were an old newspaper, a language instruction book for teaching yourself Dutch, and a complete set of 78 rpm records to go with the book.

Obviously the Dumpster Gods want you to learn Dutch.

Best not to annoy the Dumpster Gods. Besides, it astonishes the Dutch when anyone speaks their language, let alone Americans. And you have an excuse to eat chocolate sprinkle sandwiches for breakfast. (No, really.)

#67 ::: John M. Ford ::: (view all by) ::: August 26, 2003, 03:52 AM:

Edward Jay Epstein's article from the ATLANTIC (referenced above) was later expanded into the book THE RISE AND FALL OF DIAMONDS. As I recall, the book covers considerably more ground, including an account of the murder of a 47th Street dealer (one of the Hasidim with stacks of flat white boxes one sees waiting for buses to Crown Heights), which absolutely shocked the business, and (more relevantly to the immediate discussion) mention of the Soviets (remember them?) selling small, highly uniform diamonds known as "silver bears," which they claimed were natural stones from a new mine, cut with an automatic device, but the trade was terrified were synthetics. (At the time the book appeared, this situation hadn't sorted out; I assume it must have done since then.)

Two carats' worth: While I would not be surprised to see some kind of slogan to the effect that "natural diamonds are preferable/more romantic/lower in fat/more 802.11x compatible" than synthetics, they have a different problem than just selling more romance. It was never really true that "diamonds are forever," but it didn't really matter, since the stones were likely to outlive generations of owners (being reset as fashions changed), but it -was- true that diamond is a unique material -- it isn't white sapphire or cubic zirconia or Austrian crystal. While I suspect that the difference will be detectable with the right equipment (which may be anything from a loupe to a mass spectrometer), for display purposes a natural gemstone will be no different from a synthetic, and the synthetics will doubtless be available in exotic shapes and rare pure colors (the Kimberly Octahedron, a huge uncut yellow in the Smithsonian, comes to mind, as does the Black Orloff, subject of innumerable pulp yarns) for a small extra charge.

I don't know if the cartel can survive without major change (though I doubt it), but if I were the retail jewelry trade, I would be working very hard on the idea that craft and design are a lot more important to the value of jewelry than the hypothetical value of the materials. This has the virtue of being true (and demonstrably true, as anyone who watches Antiques Roadshow knows). It would also have a positive effect for good jewelry designers, but this leads me into . . . hello, Lioness.

#68 ::: --kip ::: (view all by) ::: August 26, 2003, 12:15 PM:

The trouble with creative destruction is that it isn't a process one can set out to put in place (or undergo--but one rarely sets out to undergo it oneself; funny, that). It's a perspective: one can look back and say, oh, hey, it was good that situation got bolluxed up; it forced me to pick up and move and look how much better things are now than they were then, or even before.

But it's after the fact: you've got to be in a position to look back and figure it out. Was it creative? --The trick of it is, of course, that with enough of the proper perspective, any act of destruction is creative; this is one of the first lessons you're taught about that nasty ol' Death card in the Major Arcana. After all, one could imagine a point sometime soon when one could say the short sharp shock of Perelman's stupid paper games forced a number of smaller publishers to reevaluate their dependency on the direct comics market and get serious about cracking book distributors, and look how much better off the industry is (on this one hopes not-too-distant day. Simplistic, but). And I can also say that the destruction of Robert Moses's plans for a second ring-road highway system here in Portland, Oregon was a creative destruction, indeed; if those had not been voted down at the last possible minute, whole neighborhoods of gorgeous old houses with decades of life still in them would have been wiped out, and Portland's downtown would have suffered the same walking death as other American urban centers. I still smile every time I drive by one of the aborted onramps to a highway system that almost got built: creative destruction, oh yes.

But the law of unintended consequences is itself enough to give anyone pause before crying out, "Hey! Let's go creatively fuck shit up!"

Punk notwithstanding.

#69 ::: Josh ::: (view all by) ::: August 26, 2003, 01:28 PM:

Kip: That's not what I understand the term "creative destruction" to mean. The destruction is not an end in itself, it's a side effect of improvements in the ways of doing things. Here's how Schumpeter himself defined it:

As we have seen in the preceding chapter, the contents of the laborer's budget, say from 1760 to 1940, did not simply grow on unchanging lines but they underwent a process of qualitative change. Similarly, the history of the productive apparatus of a typical farm, from the beginnings of the rationalization of crop rotation, plowing and fattening to the mechanized thing of today96linking up with elevators and railroads96is a history of revolutions. So is the history of the productive apparatus of the iron and steel industry from the charcoal furnace to our own type of furnace, or the history of the apparatus of power production from the overshot water wheel to the modern power plant, or the history of transportation from the mailcoach to the airplane. The opening up of new markets, foreign or domestic, and the organizational development from the craft shop and factory to such concerns as U.S. Steel illustrate the same process of industrial mutation96if I may use that biological term96that incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one. This process of Creative Destruction is the essential fact about capitalism. It is what capitalism consists in and what every capitalist concern has got to live in. . . .

#70 ::: Tom Whitmore ::: (view all by) ::: August 26, 2003, 01:46 PM:

TNH --

Dealing properly with the dumpster gods is what led me to find the original ms of THE BOOK OF THE LAW, as you well know. Be careful what you look through....

Those who don't know the story may google on "Raiders of the Lost Basement". Others here were Present At The Event, and don't need to.


#71 ::: Steve ::: (view all by) ::: August 26, 2003, 01:54 PM:

Edward Jay Epstein's article from the ATLANTIC (referenced above) was later expanded into the book THE RISE AND FALL OF DIAMONDS.

Despite having posted the link, I didn't realize that the author was Edward Epstein -- whom I know from his excellent book on network newscasts, News From Nowhere -- until I saw that sentence. Huh. A trip to the library to try to find the diamond book may be in order. He's got a website, too.

Besides, it astonishes the Dutch when anyone speaks their language, let alone Americans.

One of the funnier bits in Galatea 2.2 is about this very point.

#72 ::: --kip ::: (view all by) ::: August 26, 2003, 02:35 PM:

And from Schumpeter's dispassionate, objective perspective, all that roiling doubtless seems quite creative; Schumpeter's perspective, though, is chilly comfort to a farmer or steelworker left holding the bag when an old business plan mutates (incessantly or not) into a new one. (One could cheekily reimagine Schumpeter's perspective on creative destruction were he just starting out as a graduate student or adjunct faculty today, musing on the creative destruction that the capitalist enterprise has wrought on the not-so-stately groves of academe.)

I'm not denying the truth of the insight, at all; the Death card is an old friend--if not entirely trustworthy. I'm just noting the importance of perspective. It's easy to look back on the destruction of something you didn't have a stake in and say that it was creative. Or look ahead: after all, I'm cheering the possible creative destruction of Big Content's business plans, with their monstrous intellectual property schemes and digital rights management madness--then, I don't own any stock in Time Warner.

Labelling it a "process," though, makes it all too easy to say, oh, never mind those out-of-work buggy-whip crafters. Something good will come of it all. It's creative destruction!

Any act of destruction offers the opportunity for something to be created. This is a very basic observation, which one can pick up from economics--or natural science, or the various occult traditions, or dumpster diving. But it's also an act of destruction, and whether it's buggy-whip crafters, or small farmers, or cartoonists, or slimy Time Warner A&R reps, their perspectives must not be discounted in the rush toward a shiny happy new market model.

#73 ::: Emmet ::: (view all by) ::: August 26, 2003, 04:03 PM:

Stepping back up to the issue of protecting brushed-metal doors and likewise appealing bits of human surroundings from vandals with diamond styli; given mass production of diamond, how feasible if at all is it to clad or coat such things with diamond enough to protect them without affecting their appearance enough to ruin them ?

#74 ::: Jakob Whitfield ::: (view all by) ::: August 26, 2003, 04:55 PM:

Cladding doors, etc. with diamond would be much like coating them with glass. Like glass, diamond has high strength and hardness, but such a low fracture toughness that you could shatter it with a hammer.

What *would* be a really cool application for diamond would be in diamond fibres for use in composite materials, where the low toughness can be tempered by the matrix material, giving good properties overall.

What might work would be embedding small diamond particles in the door surface, but I don't know what kind of scratch protection it might give.

#75 ::: Carlos ::: (view all by) ::: August 26, 2003, 06:31 PM:

Kip, Schumpeter was hardly a dispassionate or an objective observer with regards to 'creative destruction'. He was born in old Austria-Hungary, and after WWI (he was against it) his hometown became part of Czechoslovakia, and his first university gig got finagled into the Soviet Union. He spent a year as a lord of this earth (subgenus Austrian finance minister) before his career came crashing down: he was trying to stop the wheelbarrow hyperinflation, but was too abrasive and pompous for his political rivals. In Austria yet. Then he started a bank. That crashed too. Eventually he made it to Harvard.

Economists, like most of us, write what they feel in their bones, and Schumpeter felt the disconnect between the Habsburg twilight and the vroom of early 20th century America in his. I'm not going to fault him for that.


#76 ::: --kip ::: (view all by) ::: August 26, 2003, 08:10 PM:

I don't think my point rests on the notion that Schumpeter is dispassionate and objective with regards to the overall concept of creative destruction; merely as compared with the somewhat less dispassionate and objective perspectives of small farmers and steelworkers disjointed and dispossessed by the specific creative destructions in the passage cited.

I don't deny that creative destruction is a factor; I don't deny that it's a key part of what drives growth in this capitalist era--look at how the malls which replaced the downtowns are being replaced by the big box retail outlets for a whole epic of creative destruction. But your opinion of that epic is going to differ--depending (in part, yes) on whether you're looking at sales figures and business plans, a paycheck where maybe there hadn't been one before and the cheapest tennis shoes in town, property values and traffic congestion, or dwindling money coming into your store's till and the choice of which employee to fire. --Labelling it "creative" destruction is from some perspectives a luxury; that those perspectives might be wed to a business plan as outmoded as buggy-whip crafting doesn't mean they should be discounted out of hand. And I'm not saying we should stop all Progress so that file clerks used to green eyeshades and sleeve garters can go about their daily round without ever being discomfitted, and heck, I suppose I shouldn't say Schumpeter discounted the perspectives of those dispossessed by capital's revolutions--how awful to judge him by one brief passage in a debate ostensibly about diamonds. --So maybe I should just recuse my gut and jerking knee, already.

The tree of capitalism must be watered with the blood of the also-ran, fine--but all of us who benefit from that tree sometimes do a lousy job of helping pick up the casualties. Is that all I'm trying to say? Maybe.

So! These diamonds, they glitter?

#77 ::: Randolph Fritz ::: (view all by) ::: August 27, 2003, 08:58 PM:

"how feasible if at all is it to clad or coat such things with diamond enough to protect them without affecting their appearance enough to ruin them ?"

Why not use it in places where glass or plastic are already used? Lenses, laptop windows. Hey! Vacuum tubes envelopes!

#78 ::: Robert L ::: (view all by) ::: August 28, 2003, 02:13 AM:

Speaking of the Dumpster gods...I just found an unopened bottle of Veuve Clicquot on the street on my way home from my local bar...

#79 ::: PDM ::: (view all by) ::: August 28, 2003, 09:53 AM:

This kinda reminds me of Napster and the record industry (which, in its own way, has been as evil a bunch o' capatalist bastards as the folks who run deBeers).

The problem with Napster was, IMHO, that it was no better than the record co.'s in ripping off artists/musicans.

OTOH, I'd have no qualms with this. MASH THE DIRTY DeBEERS SCUM! KILL! KILL!!!

#80 ::: Christina Schulman ::: (view all by) ::: August 28, 2003, 08:40 PM:

Speaking of the Dumpster gods...I just found an unopened bottle of Veuve Clicquot on the street on my way home from my local bar...

Nifty. You must be associating with the right sort of dumpster. Ironically, the last thing I found next to a trash can (while vacuuming dog hair out of my car) looks a lot like a diamond earring.

(No attendant; no signage; I left my business card with the clerk at the gas station next door in case someone came looking.)

#81 ::: Jon H ::: (view all by) ::: August 30, 2003, 01:03 PM:

Coming to this late, so I haven't read the comments so I don't know if this has been brought up.

De Beers could have a major problem if they try the "natural is better" defense.

The artificial diamond manufacturers may have a distinct advantage, if they can add impurities to the diamonds during manufacture, to create custom colors, or even patterns.

If the artificial diamond firms can figure out how to do that, and generate demand for the customized-color diamonds, De Beers will be in a rough spot, being at the mercy of nature for their selection of gems.

#82 ::: adamsj ::: (view all by) ::: September 02, 2003, 09:56 AM:

The problem with creative destruction is who decides what gets created and what--or who--gets destroyed.

#83 ::: adamsj ::: (view all by) ::: September 02, 2003, 09:58 AM:

The problem with creative destruction isn't the creation and it isn't the destruction--it's the question of who decides what gets created and what--or who--gets destroyed.

Smaller type (our default)
Larger type
Even larger type, with serifs

Dire legal notice
Making Light copyright 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017 by Patrick & Teresa Nielsen Hayden. All rights reserved.