Slashdot Mirror

Remain silent? There was a BBC Horizon documentary on this very subject broadcast earlier this year. You can read more about the problem and the solution here:

http://www.spectrum.ieee.org/WEBONLY/publicfeature /oct04/1004titan.html

And here:

http://www.bbc.co.uk/sn/tvradio/programmes/horizon /saturn_prog_summary.shtml

Problem: Italian Company (Alenia Spazio) responsible for comms corrected for doppler shift on the carrier signal, but not on the data rate. Alenia Spazio's insistence on confidentiality may have played a role in this oversight. NASA reviewers were never given the specs of the receiver. As JPL's [Robert] Mitchell explained, "Alenia Spazio considered JPL to be a competitor and treated the radio design as proprietary data."

Solution: Altered the trajectory of Cassini / Huygens so that Huygens is moving parallel to Cassini during descent, sidestepping the doppler shift issue.

As someone in Engineering who has been involved with highschool recruiting/outreach, I've got to point out -- don't forget about the girls. Often, science and technology types (usually guys) assume that only boys are going to be interested in things like computer science, physics, engineering, etc. As a result many high school career councilors don't even point out computer science or engineering as career paths for women, even though there are many scholarships available to students who go that route. Studying in engineering/technical areas can be very lucrative for bright young women.

There's a pretty good reason there aren't a lot of women in computer science, engineering and other technical fields! The girls get turned away, or are not encouraged when they are in high school. If you're looking for a good resource on this topic I suggest you contact IEEE Women in Engineering (your local branch) since this is right up their alley.

The Chinese didn't lease the Soyuz design: at one point they tried to purchase a Soyuz for reverse-engineering purposes, but the only thing the Russians were willing to supply for the price was a capsule stripped of almost all its systems.

While obviously Soyuz inspired, the Chinese design is home grown and features a number of design improvements over the Soyuz: it's larger, and most notably, the orbital module is, unlike the Soyuz OM, capable of independant flight.

There's a good article by James Oberg about it in IEEE Spectrum.

There is another recent article on this topic here in the latest issue (Dec 2004) of IEEE Spectrum. From this article it looks like this technology will be commercialized within the next 2-3 years.

Interesting story here about how it almost went disasterously wrong...

First of all, the KLT (Karhunen-Loeve Transform) is theoretically most efficient at packing energy, but 1. the basis vectors depend on Markov statistics of the data, and 2. because the KLT isn't as factorable as the FFT or DCT, there is no "fast" algorithm like there are for FFT and DCT. Second, it appears that the KLT with Markov settings from a corpus of actual images closely resembles the DCT, as shown in the figures of these interesting slides.

Proofs exist, but IEEE non-members aren't allowed to view them, and no way am I spending three figures on an annual membership just to argue a point on Slashdot.

See Spectrum's article on the 50th anniversary of the transistor.

Here (at a Raytheon facility), we've got a display in the hallway with about a dozen vacuum tubes. Some dating back to 1922 (and patented...doh, there's that bad word)...the tube was originally named "Raytheon", then the company named after it.

Seems we had a bit to do with the microwave also.

The Fleming diode might have been the first electron tube (valve) but it was Lee De Forest who put in a control grid between the filament cathode and the anode plate and created a device that could amplify voltage, which made the whole world of useful electronics circuitry all possible. The 100th anniversity of De Forest's "Audion" triode is not until 2007.

The first meeting of the IEEE Computer Society "Local Network Standards Committee", Project 802, was held in February of 1980. (The project number, 802, was simply the next number in the sequence being issued by the IEEE for standards projects).

The story I've heard is that the '80' does relate to the year 1980 but the '2' is a sequence number saying 802 was the second committee formed in 1980. If the first meeting had been in March, IEEE802 would still be IEEE802 (and not IEEE803).

Can anyone provide clearer references to show that the '80' really does relate to 1980 (not just a coincidence) and whether the '2' is really just a sequence number (and not the month)?

Wouldn't you rather say 802.20 phones?

802.16 is the IEEE standard for broadband wireless access. Whereas 802.11 is the IEEE standard for wireless LANs.

802.16 is the IEEE standard for broadband wireless access. Whereas 802.11 is the IEEE standard for wireless LANs.

Huygens has already "almost failed". Some clever guy was analyzing the design of the radios, and found out that the data decoder was incapable of dealing with the doppler shift the probe was to encounter. They've since changed the trajectory of the mother ship to minimize the problem. The problem was caused by using a closed source radio system, and not detected because of limited testing, especially using 'real world' doppler-distorted signals.

We will return to nuclear power, but to see what it will look like when we do, go to this site on Z-pinch and Wikipedia's article on Inertial fusion energy".

Cassini carries huygens, a land probe which will (hopefully) land on Titan on january 14th. There is an interesting story on ieee spectrum about an engineer who prevented the mission from certain failure.

Nuclear rockets on cars? probably not, but, as many people here have pointed out, nuclear powered cars are probably not too far off. Tritium is pretty DARN safe(it won't penetrate human skin) and hopefully it(or something like it) will someday be produced in enough quantity so we can have batteries powered by it. Imagine a car or laptop that would run for years on a "relatively" little amount of this crud. I don't doubt that developing a nuclear engine for space travel would have a seriously postive effect on getting radioactive batteries produced.

Anyhoo, NASA has produced a lot more than new rockets, how about catalytic converters that work when cold and cost 25% less.

Here is another great blurb on Tritium from the guy that made the Wooden Periodic Table Table.

In effect, the [Doppler] shift would push the signal out of synch with the timing scheme used to recover data from the phase-modulated carrier.

IEEE Spectrum has a nice article on this near-debacle. The problem was that the tests performed did not consider that the received data rate, in addition to the carrier frequency, would be affected by Doppler, and the receiver would lose bit synchronization.

The full system test, which would have detected the problem, was not performed because--wait for it--they wanted to save money.

They didn't skip any numbers (except for the unlucky 802.13).

802.11 Wireless LAN Working Group
802.12 Demand Priority Working Group
802.14 Cable Modem Working Group
802.15 Wireless Personal Area Network (WPAN) Working Group
802.16 Broadband Wireless Access Working Group
802.17 Resilient Packet Ring Working Group
802.18 Radio Regulatory TAG
802.19 Coexistence TAG
802.20 Mobile Broadband Wireless Access (MBWA) Working Group
802.21 Media Independent Handoff Working Group

Who designs the numbering systems?

These guys do.

Or you can wait 6 months and get it for free.

This work seems to be about two things (which I am not sure I see a strong connection between): lowering transport latency, and using available bandwidth better. The latter has been the subject of many papers in the last few years. There are now several serious proposals of how to fix TCP with respect to long fat pipes. They don't seem to support the idea that retransmissions are harmful. So I'm going to talk about the first issue, transport latency.

The idea of using error-correcting codes (ECC) to eliminate the need for retransmissions is an interesting one. The main benefit is to reduce transport latency (the total time it takes to send data from application A to B). Here is another paper proposes has a similar idea, applied at a different level of the network architecture.

The root problem here is that network loss leads to increases in the transport latency experienced by applications. In TCP, the latency increases because TCP will resend data that is lost. That means at least one extra round-trip-time per retransmission. This "Rateless TCP" approach uses ECC so that the lost data can be recovered from other packets that were not dropped. In this way, the time to retransmit packets may not be needed. I say may, because there will be a loss rate threshold which will exceed the capability of the ECC, and retransmission will become necessary to ensure reliability. But, as long as the loss rate is below the threshold, then retransmissions will not be necessary. Note that the more "resilient" you make the ECC (meaning supporting a higher loss threshold), the more work will be needed at the ends. So you are not eliminating latency due to packet loss, you are simply moving it away from packet retransmission into the process of ECC. However, if you've got good ECC, the total latency will go down.

The ECC approach may be a nice middle ground. But, it the ultimate solution to minimize latency is probably through a combination of active queue management (AQM) and early congestion notification (ECN). Unlike ECC, this approach really would aim to eliminate packet loss in the network due to congestion, and therefore completely eliminate the associated latency. Either ECC or regular TCP would benefit. In a controlled testbed using AQM and ECN, I've completely saturated a network with gigabits of traffic, consisting of thousands of flows, and had virtually no packet loss.

It should also be noted that retransmission is NOT the dominant source of transport latency in of TCP. I am a co-author on a paper that shows another way (other than eliminating retransmission) to greatly reduce the transport latency of TCP. The basic idea is that the send-side socket buffer turns out to be the dominant source of latency (data sits in the kernel socket buffer waiting for transmission). In the above paper, we show how a dynamic socket buffer (one that tracks the congestion window) can dramatically reduce the transport latency of TCP. We allow applications to select this behaviour through a TCP_MINBUF socket option.

-- Buck

We just spent $10K+ ... to mitigate the costs of having an outside lab...

What would have been the cost to have an outside lab do the testing?

it's a big burden for small manufacturers.

Agreed - I know how small business can be, but it's a bigger burden on the community at large to do anything else.

Current FCC/CE regs ... [are] a bit onerous IMHO.

What makes them onerous it the cost of complying. A better solution than a "this probably passes" statement is better access to testing labs for small businesses. Something like a Business Assistance Grant for small businesses in the electronics sector. Maybe take it up with your local governing authorities. Or see if (as someone else suggested) you can use the fact that you've grown beyond the infancy stage, you can persuade your local industry association to set up some sort of program.

Or there's an opportunity for some enterprising engineers to set up a "FCC regs compliance testing" company, priced below what it costs internally, and sweep up all the small-business testing jobs in your state...

There is an interesting article (http://www.spectrum.ieee.org/WEBONLY/publicfeatur e/oct04/1004titan.html) in the current issue of IEEE spectrum. They discuss how a disaster was avoided by Boris Smeds who pushed for stringent tests of the communication between Cassini and Huygens. It turns out that the Italian manufacturers of the radio didn't take into account the significant doppler shift between both craft. As the firmware of the radio could not be remotely upgraded, Cassini's trajectory was altered (further away from Titan) to lower these doppler shifts.

Let's hope no other misfortunes turn up.

You wouldn't bend the pins in the upper left corner, like they have. Note the pin located at (7,-1). That's going to be a B#*%^ to straighten without snapping it off.

Something tells me that's a spent battery though, since there's minor damage to many of the pins.

Where are you getting your definition of Ethernet? You're not allowed to redefine it to "the definition that makes me right" The actual standard includes the physical layer specifications.

Here's the abstract for 802.3 aka, Ethernet (if you care to bother, you can download the full standard for free, and I've added emphasis here):

IEEE Std 802.3: CSMA/CD Access Method and Physical Layer Specifications. Abstract: The media access control characteristics for the Carrier Sense Multiple Access with Collision Detection (CSMA/CD) access method for shared medium local area networks are described. The control characteristics for full duplex dedicated channel use are also described. Specifications are provided for MAU types 1BASE5 at 1 Mb/s; Attachment Unit Interface (AUI) and MAU types 10BASE5, 10BASE2, FOIRL (fiber optic inter-repeater link), 10BROAD36, 10BASE-T, 10BASE-FL, 10BASE-FB, and 10BASE-FP at 10 Mb/s; Media Independent Interface (MII) and PHY types 100BASE-T4, 100BASE-TX, 100BASE-FX, and 100BASE-T2 at 100 Mb/s; and the Gigabit MII (GMII) and 1000BASE-X PHY types, 1000BASE-SX, 1000BASE-LX, and 1000BASE-CX, which operate at 1000 Mb/s (Gigabit Ethernet) as well as PHY type 1000BASE-T. Repeater specifications are provided at each speed. Full duplex specifications are provided at the Physical Layer for 10BASE-T, 10BASE-FL, 100BASE-TX, 100BASE-FX, 100BASE-T2, and Gigabit Ethernet. System considerations for multisegment networks at each speed and management information base (MIB) specifications and additions to support Virtual Bridged Local Area Networks (VLANs) as specified in IEEE P802.1Q are also provided. Also specified is an optional Link Aggregation sublayer which multiple physical links to be aggregated together to form a single logical link.

Thus, just as a protocol which fits the specifications in 802.3 is known as an "Ethernet protocol", a physical cable which also meets the given specs is correctly known as an "Ethernet cable." Ethernet can not run on "any type of cable" and still be Ethernet. To quote the standard: communication by way of the ISO/IEC 8802-3 [IEEE Std 802.3] Local Area Network requires complete compatibility at the Physical Medium interface (that is, the physical cable interface). The standard describes a number of ways of physical cabling a network together (co-ax, twisted pair, fibre optic), but these must all meet the specs and so be "Ethernet cables."

Now, if you can quote something more authorative than the standard, I love to see it.

The Daintiest Dynamos

Alternate Post:
By the time they release this, these batteries will be for sale at your local gas station.

Actually, the point of this article is batteries that scale *down* rather than up. One of the stumbling blocks for miniturized mechanics (MEMS,e tc.) has been the lack of a comparably sized power source. Sure, you can have MEMS accelerometers powered off of your car battery (to sense when to deploy your airbags). But if you want to sever the tether and keep things at a micro scale, you must scale the power to that scale.

Also worth noting, the batteries mentioned in the article actually operate on a different principle than RTGs. The T in RTG stands for "thermoelectric." The article talks about generating power using peizioelectrics. See the figure (http://www.spectrum.ieee.org/WEBONLY/publicfeatur e/sep04/0904nucf1.html).

There also is an interesting sidebar comparing the amount of radioisotope needed for such batteries to current commercial applications in which radioisotopes are used (http://www.spectrum.ieee.org/WEBONLY/publicfeatur e/sep04/0904nucsb1.html). Individual devices sound tame enough, but I think the real problem will be disposal - especially when everyone has one in their cell phone.

Actually, the point of this article is batteries that scale *down* rather than up. One of the stumbling blocks for miniturized mechanics (MEMS,e tc.) has been the lack of a comparably sized power source. Sure, you can have MEMS accelerometers powered off of your car battery (to sense when to deploy your airbags). But if you want to sever the tether and keep things at a micro scale, you must scale the power to that scale.

Also worth noting, the batteries mentioned in the article actually operate on a different principle than RTGs. The T in RTG stands for "thermoelectric." The article talks about generating power using peizioelectrics. See the figure (http://www.spectrum.ieee.org/WEBONLY/publicfeatur e/sep04/0904nucf1.html).

There also is an interesting sidebar comparing the amount of radioisotope needed for such batteries to current commercial applications in which radioisotopes are used (http://www.spectrum.ieee.org/WEBONLY/publicfeatur e/sep04/0904nucsb1.html). Individual devices sound tame enough, but I think the real problem will be disposal - especially when everyone has one in their cell phone.

It takes 4 years for all those 57000 whatevers to come out. I doubt that's very powerful for anything. I believe they are aiming at long term stuff like airbags and such.

Here's hopefully a clarification of all the below, from someone who works on net metering and interconnection a solid number of hours per week...

Equipment for interconnection and backfeed into the grid are governed by UL and, more relevantly, IEEE standards. (742 and 1547, respectively.)

These have fairly elaborate specifications for, as you said, not hitting the ground protection equipment, going into non-export mode (within a certain number of milliseconds) if the grid trips off, or if the current frequency or reactive power or a variety of other things vary outside certain parameters. Most utilities also require you to have an external, lockable disconnect (for instance, for if linemen are working outside your house on a segemnt of line that's still giving you good enough power that the inverter hasn't tripped you off. (they sensibly treat power lines the same way you treat working on guns - be 100% sure it's unloaded, and then be 100% sure you treat it as though it's loaded.)

These technical standards are fairly well established by the community of power engineers to be more than sufficiently safe, and we're seeing progress in their implementation: the National Association of Regulatory Utility Commissioners, for instance, has issued a model standard for their use.

However, it's far from uniform nationwide; utilities are used to implementing their own idiosyncratic rules for every part of their grid, and this (testing and retesting the same inverters in enery state, making them programmable to comply with Podunk Utility Co's individual requirements, etc.,) has become a major cost obstacle for the use of what's termed "distributed generation."

It does vary from state to state and utility to utility; while they *are* required under the Public Utility Holding Company Act or the Public Utilities Regulatory Policies Act (don't remember which it is) to buy your power back from you, that's only at wholesale (ca. 2 cents per kWH,) whereas if you're truly net metering (and single-meter, spin-it-backwards net metering is something literally thousands of solar and small wind users legally do every day), you can get 7 - 14 cents. An easy way to check all of these, state by state, is the DSIRE website.

As for the solar energy payback timeand toxicity,which I'm frustratingly sure someone will bring up in this thread, those are essentially canards that "seem right" to those who haven't really looked into it but maybe heard it ten years ago, but don't hold up to empirical scrutiny.

Yeah, which a diesel-hybrid would be able to do. Daimler-Chrysler is apparently investigating a 'plug-in' diesel-electric hybrid version of their 'Sprinter' cargo van (Think UPS truck,) that could get an astonishing 50 mpg in diesel/hybrid driving, and could drive up to 30 miles in electric-only mode. (You can CHOOSE to plug it in at night if you want, to get the extended range, or you can just let it work as a 'normal' hybrid, where it uses the diesel engine to recharge the hybrid battery pack.) The idea is that it could be used as a delivery truck in neighborhoods where delivery trucks are banned due to noise and pollution. It would drive in hybrid mode on the highway, then switch to electric mode in the neighborhoods.

You can seen an article on it here.

The IEEE has a nice overview of the plug-in hybrid concept. The fleet turnover to plug-in hybrids could parallel the build out of economic wind energy generation for a truly zero emission vehicle.

This is the problem with having "standards" put out by people other than the IETF and IEEE

Here is an article about the problem on IEEE Spectrum. I have had to replace many motherboards because of this problem, all MSI or Epox.

Nanotube based displays will enable the use of CRT-like technology. But instead of a whole box-sized vacuum tube, we'll have a flat grid (why didn't I think of that?). The advantage is that the voltage requirements will be minimal. Like 20 volts instead of the 200 that today's CRT-based displays require.

Hey look at that. This article is weekly - i wonder why didn't I submit it as a story in the first place... ;-)

...except WEP actually stands for Wired Equivalent Privacy

The link for those interested:

IEEE Spectrum

Airports suck.
But now what am I referring to? The Apple Airport? A generic access point? Wifi in general? Or am I referring to an actual airport? Nobody knows because of your fanboyism.

Actually the best thing to do is keep stum, start documenting everything and then leak the information. There's a good article, The Whistelblower's Dilemman written by one of my colleagues about dealing with this kind of siutation. Some excerpts:

Among the other mistakes Martin cites are that people don't collect enough evidence of the problem they're trying to expose, don't build support among colleagues and others, and don't wait for the right opportunity to come forward. "My advice to most people is, 'Don't do it--until you're done investigating, preparing an escape route, and weighing your options,'" he says.

That last piece of advice is especially important. "People think the right thing to do is just speaking out. But there are many different ways to do the right thing. It may be best to wait and collect more information. You also have to look at the consequences, for yourself, your family, your colleagues."

...

Because of the many bad things that happen to whistle-blowers, Dina Rasor likens the act to "setting your hair on fire for one glorious minute." She has two words of advice for would-be whistle-blowers: remain anonymous. "If there's any way to get the information out--through a nonprofit, or a trusted reporter, or a friend--without identifying yourself and having your fingerprints all over it, that's preferable to going public. Then the fraud becomes the issue, and not you."

There's also contact details for organizations that help whistleblowers too.

Putting RFID chips on your shoes is nothing new.

• All Dockers khaki pants
• All Colgate Shave Cream packages
• All Trojan Ultra Ribbed condom boxes
• Some Gilette razors

While I'm sure that nobody is tracking you right now, RFID tags can be read by several meters away and contain unique identifiers. If you thought the Pentium chip unique IDs were bad, this should (rightly so) worry you considerably more.

http://www.spectrum.ieee.org/WEBONLY/resource/aug0 4/0804ntvf1.html

link to a chromacity diagram from the Spectrum article. No triangle in the diagram can cover the whole diagram. The RGB phosphors are far inferior to pure RGB, and the 5-color system is a significant improvement. Note that just using their green in a 3-color system would provide half of the improvement they claim.

I'll wait for HDR display and feeds, thanks.

Judging from the gamut chart for this RGBCMY, the boost in color range is primarily in yellows and cyans. Gold, as they note, would be a good application. Cyan.. well, that's mostly skies - and those already appear just fine on TV. A fairly decent increase in magentas/purples as well (when taking the assymetric lobe into account), but again.. not seeing its application much.
Unless following the British royal family (lots of golds and purples) a lot, it doesn't appear to offer all that much. Especially considering movie people butcher things anyway (DVD gives a more stable picture, sure.. at the compromise of mpeg artifacting and even encoding issues.. twitches ever 25 frames are annoying - luckily only a few suffer from this).

On the other hand, a higher dynamic range would be immediately noticeable anywhere.
A sequence with the sun glaring into the camera ?
A car's headlights shining at the camera ?
Highlights on objects ?
Blown-out surfaces from bright lighting ?

All that could then more accurately be represented. And thanks to most things still being shot on film, or already on 10bit CCDs with, formally, underexposure but a gain for the operator, a good bit of extra range is already available in previous and current productions.
Whilst RGBCMY would only really be of use for film (as in, actual film) productions, as digital cameras are in much the same RGB limbo that current displays are.

The link provided by the poster explains quite thoroughly the idea behind the story: that fundamental changes in the way that the spectrum is used today, and will be used in the future, claims for a revised regulation.

This is definetly not new nor novel, this article on IEEE's Spectrum (pretty convenient huh?:) magazine had stated this trend months ago.

This issue is very interesting due to the potential economic consequences of a shift from a scarcity of spectrum regulatory policy, to an abundance one, potentially rendering some established telecom business models (and companies) worthless.

In particular, the thrust of Shirky's argument is that we should change how we do things (i.e., the regulatory environment) because we can make use of the spectrum as a public commons without interfering with one another. The gaping hole in this argument is that, absent FCC regulation (or something equivalent), there is nothing to guarantee that everyone will operate this way. And it only takes one bad actor to ruin everyone's fun.

As a specific example, imagine that a large telecommunications company decides to market their new "bulletproof" phone service in the (currently unrestricted) 2.4GHz band by spending a huge chunk of cash to set up megawatt-level transmitters all over the place. Sure, their service will work great... but given enough power, it will drown out many/most other devices in the band, whether they are spread-spectrum or not. Shirky also mentions people in adjacent homes using wireless routers without interfering with one another, but there is nothing fundamental about that, either -- I could build a jammer for less than $100 that would disrupt every wireless 802.11g router within a city block.

Nor is this phenomenon limited to the WiFi band -- my lab has done quite a bit of research into the potentially disruptive effects of the proposed ultrawideband (UWB) allocation on GPS, which is in wide use worldwide, including some safety-critical applications.

As for the Economist piece, many other posters in this thread have noted a multitude of problems originating from the journalist-writing-as-engineer nature of the article; here's another big one: The article suggests (in the first two paragraphs of the section entitled "The sweet and low down") that simply repurposing the lower (i.e. currently licensed) swaths of spectrum is something of a panacea. What the author doesn't seem to understand is that there is an attendant difficulty in designing efficient antennas at these lower frequencies. There's a reason, for example, that commercial radio broadcasts aren't done in the 100KHz band -- the antennas on both ends would have to be hundreds of meters long (on the order of a quarter-wavelength) to be even marginally efficient. And if the antennas have to be a manageable size (and therefore inefficient), the transmitter power has to go way up to make the link work -- and we're right back where we started.

There is certainly promise in spread-spectrum radios, mesh networks, and other cool new technology. But it's not nearly as much of a no-brainer as these two pieces make it sound. I hate to be on the side of the "old school," but there is considerable merit to that line of thinking here.

-HJ

Almost correct. You need AES-128 authentication and message integrity checking, not encryption, but ZigBee has both. Encryption merely makes the message private, but authentication ensures that the source address wasn't spoofed, and integrity checking ensures that it hasn't been corrupted or modified before reaching the recipient.

BTW, the IEEE 802.15.4 spec is available for free download.

The high cost of access is also why I gave up my membership in IEEE. Of all the organizations, one would think IEEE would allow open access; but they don't. And want to charge an arm and a leg for everything. Screw them. I urge others to drop their IEEE membership too. Only when people start leaving them in droves will they change their policies.

Some sort of deposit had appeared on top of the electrolytic capacitors on the motherboard, and the machine became unbearably unstable.

Green Hills Software, Inc. has registered INTEGRITY 5.0 PPC as conforming to the 1003.1-2003 System Interfaces Product Standard.

Coincidence?

If he were Martha Stewart, he would have gotten a slap on the wrist (and would have still appealed). As it is, I bet he was on some FBI shit list.

By the way, this same sort of thing happened recently with the IEEE and other professional organizations with respect to embargoed axis of evil (TM) countries. They reasoned that if you edit a paper submitted from Iran, you are providing a service to that country. A couple professional societies gave the Treasury Department the finger. In April they finally recently fixed that part of the law so that the organizations are in the clear again.

Government shouldn't block chess or science... Or crypto while I'm at it. :)