Slashdot Mirror

An anonymous reader quotes a report from Gizmodo: An international consortium involving over 50 institutions has announced an ambitious project to assemble high-quality genome sequences of all 66,000 vertebrate species on Earth, including all mammals, birds, reptiles, amphibians, and fish. With an estimated total cost of $600 million dollars, it's a project of biblical proportions. It's called the Vertebrate Genomes Project (VGP), and it's being organized by a consortium called Genome 10K, or G10K. As its name implies, this group had initially planned to sequence the genomes of at least 10,000 vertebrate species, but now, owing to tremendous advances and cost reductions in gene sequencing technologies, G10K has decided to up the ante, aiming to sequence both a male and female individual from each of the approximately 66,000 vertebrate species on Earth. Cofounders of the project announced the new goal yesterday at a press briefing held during the opening session of the 2018 Genome 10K conference, currently being held at Rockefeller University in New York City. The project will involve over 150 experts from 50 institutions in 12 countries.

A machine learning method which has been widely used in face recognition and other image- and speech-recognition applications, has shown promise in helping astronomers analyze images of galaxies and understand how they form and evolve. From a report: In a new study, accepted for publication in Astrophysical Journal and available online [PDF], researchers used computer simulations of galaxy formation to train a deep learning algorithm, which then proved surprisingly good at analyzing images of galaxies from the Hubble Space Telescope. The researchers used output from the simulations to generate mock images of simulated galaxies as they would look in observations by the Hubble Space Telescope. The mock images were used to train the deep learning system to recognize three key phases of galaxy evolution previously identified in the simulations. The researchers then gave the system a large set of actual Hubble images to classify.

The results showed a remarkable level of consistency in the neural network's classifications of simulated and real galaxies. "We were not expecting it to be all that successful. I'm amazed at how powerful this is," said coauthor Joel Primack, professor emeritus of physics and a member of the Santa Cruz Institute for Particle Physics (SCIPP) at UC Santa Cruz. "We know the simulations have limitations, so we don't want to make too strong a claim. But we don't think this is just a lucky fluke."

Tim Stephens reports via The University of California in Santa Cruz: A new study by an international team of astronomers reveals that four Earth-sized planets orbit the nearest sun-like star, tau Ceti, which is about 12 light years away and visible to the naked eye. These planets have masses as low as 1.7 Earth mass, making them among the smallest planets ever detected around nearby sun-like stars. Two of them are super-Earths located in the habitable zone of the star, meaning they could support liquid surface water. The planets were detected by observing the wobbles in the movement of tau Ceti. This required techniques sensitive enough to detect variations in the movement of the star as small as 30 centimeters per second. The outer two planets around tau Ceti are likely to be candidate habitable worlds, although a massive debris disc around the star probably reduces their habitability due to intensive bombardment by asteroids and comets.

Big Hairy Ian quotes the BBC: Engineer Harry Huskey, who helped build many of the first ever computers, has died aged 101. Dr. Huskey was a key member of the team that built the Electronic Numerical Integrator and Computer (ENIAC) which first ran in February 1946. ENIAC is widely considered to be one of the first electronic, general purpose, programmable computers. Dr. Huskey also helped complete work on the Ace -- the Automatic Computing Engine -- designed by Alan Turing.
U.C. Santa Cruz also remembers Huskey's work on the Bendix G-15 in 1954, "a 950-pound predecessor to today's laptops" which is sometimes hailed as the first personal computer (since it didn't require a separate technician to run) -- though each one cost over $50,000. The idea of an "electronic brain" was still so new, it led Huskey to an appearance on Groucho Marx's radio show You Bet Your Life, where Groucho warned him that "They're pretty tricky those machines! I wouldn't trust 'em... They'll turn on your like a mad dog, doctor!"

An anonymous reader quotes a report from Scientific American: For the first time ever, astronomers have found strong evidence of water clouds on a body outside the solar system. New observations of a frigid object called WISE 0855, which lies 7.2 light-years from Earth, suggest that the "failed star" has clouds of water, or water ice, in its atmosphere, the researchers said. "We would expect an object that cold to have water clouds, and this is the best evidence that it does," study lead author Andrew Skemer, an assistant professor of astronomy and astrophysics at the University of California, Santa Cruz, said in a statement released by the university. Scientists discovered WISE 0855 in 2014, using data from NASA's Wide-field Infrared Survey Explorer (WISE) spacecraft. A later paper in 2014 (co-authored by Skemer) uncovered some evidence of water clouds in the object's atmosphere, based on limited photometric data (how bright the object is in specific light wavelengths). In the new study, Skemer and his colleagues used the Gemini North telescope in Hawaii to study the brown dwarf for 13 nights. Gemini North is located on the highest Hawaiian mountain (Mauna Kea), at an altitude with little water vapor to interfere with telescopic observations. These observations allowed the astronomers to make the first spectroscopy (light fingerprint) measurements of WISE 0855. The team found water vapor and also confirmed the object's temperature, which is about minus 10 degrees Fahrenheit (minus 23 degrees Celsius, or 250 kelvins).

An anonymous reader writes: Researchers from UC Santa Cruz, the University of Washington, and the Benaroya Research Institute at Virginia Mason found shark jelly to have the highest proton conductivity ever seen in a biological material. The jelly's conductivity begins to approach that of leading proton-conducting polymers. Tiny organs in the skin of sharks, skates and rays, called the ampullae of Lorenzini, are key to the ability. Scientists believe that the jelly is what has been able to allow these animals to detect weak electric fields produced by their prey, as the organs, which are visible as pores in the skin, are connected to electrosensory cells via long, jelly-filled canals. Marco Rolandi, a co-author on a paper detailing the findings in Science Advances, sees potential use for the "shark jelly" in the development of new or enhanced materials or even the creation of new sensor technology. "The observation of high proton conductivity in the jelly is very exciting," Rolandi said. "We hope that our findings may contribute to future studies of the electrosensing function of the ampullae of Lorenzini and the organ overall, which is itself rather exceptional."

coondoggie writes: As the amount of newly discovered planets and systems outside our solar system grows, NASA is assembling a virtual team of scientific experts to search for signs of life. The program, Nexus for Exoplanet System Science (NExSS) will cull the collective expertise from each of NASA's science communities, including earth scientists, planetary scientists, heliophysicists, and astrophysicists. They'll work with key universities to better analyze all manner of exoplanets, as well as how the planet stars and neighbor planets interact to support life.

An anonymous reader writes "An evolutionary race between rival elements within the genomes of primates drove the evolution of complex regulatory networks that orchestrate the activity of genes in every cell of our bodies, reveals new research. The race was between mobile DNA sequences known as 'retrotransposons' (jumping genes) and the genes that have evolved to control them. Scientists at the University of California Santa Cruz, identified genes in humans that make repressor proteins to shut down specific jumping genes. "We have basically the same 20,000 protein-coding genes as a frog, yet our genome is much more complicated, with more layers of gene regulation. This study helps explain how that came about," said Sofie Salama, a research associate at the UC Santa Cruz Genomics Institute who led the study."

Trepidity writes "AI systems can (sort of) paint and compose classical music, but can they design games? Slashdot looked at the question a few years ago, and several research groups now have experimental systems that design board games and platformers with varying levels of success. I've put together a survey of the AI game designers I know of, to round up what they can do so far (and what they can't). Are there any others out there? 'Pell's METAGAME is, to my knowledge, the first published game generator. He defines a generative space of games more general than chess, which he calls "symmetric, chess-like games." They're encoded in a representation specific to this genre, which is also symmetric by construction. By symmetric I mean that mechanics are specified only from the perspective of one player, with the starting positions and rules that apply to the other player always being the mirror of the first player's. The rules themselves are represented in a game grammar, and generation is done by stochastically sampling from that grammar, along with some checks for basic game playability, and generative-parameter knobs to tweak some aspects of what's likely to be generated.'"

derekmead writes "New evidence that the giant impact hypothesis is correct: A paper published today in Nature shares findings of a chemical analysis of Moon rocks that shows fractional differences between the makeup of the Earth and Moon that most likely were caused by the collision between Earth and a Mars-sized planet around 4.5 billion years ago. Although the two are quite similar, it's been previously shown that Moon rocks lack volatile elements, which suggests they may have evaporated during the incredibly intense heat and pressure created during an impact event. But if the hypothesis that light elements actually evaporated from Moon rocks during their formation is correct, you'd expect to find evidence of elements being layered by mass — heavier elements would condense first, and so on. That process is known as isotopic fractionation — a concept central to carbon dating — and the Washington University team's results suggest they found exactly that (abstract). They compared the blend of zinc isotopes in Moon rocks and Earth samples, and found that the Moon rocks held slightly higher proportions of heavier zinc isotopes. If the Moon was indeed once part of Earth — which has been shown by extensive modeling (PDF) — the difference in the balance of zinc profiles would most likely be explained by lighter zinc isotopes evaporating away following a collision."

the_newsbeagle writes "To study the mysterious phenomena of dark matter and dark energy, astronomers are turning to supercomputers that can simulate the entire evolution of the universe. One such simulation, the Bolshoi projection, recently did a complete run-through. It started with the state the universe was in around 13.7 billion years ago (not long after the Big Bang) and modeled the evolution of dark matter and energy up to the present day. The run used 14,000 CPUs on NASA's fastest supercomputer."

An anonymous reader writes "There are two types of office workers in the world — those who file their emails in folders, and those who use search. Well, it looks like the searchers are smarter. A 354-user study by IBM research found that users who just searched their inbox found emails slightly faster than users who had filed them by folder. Add the time spent filing and the searchers easily come out on top. Apparently the filers are using their inbox as a to-do list rather than wanting to categorize information to find it more easily."

The StarCraft 2 team spent most of Blizzcon talking about the map editor and custom games. We spoke with Alan Dabiri, a Lead Software Engineer for Wings of Liberty who worked on the user interface and helped out on the game's integration with Battle.net. He provided some more details about plans for making the map editor more approachable, the coming updates for Battle.net (including chat channels), and a bit about the development of Heart of the Swarm, the Zerg-themed expansion being worked on now. Read on for our conversation about StarCraft 2.

Slashdot: Can you tell us how development has been split up now that you're maintaining Wings of Liberty, starting on Heart of the Swarm, and providing updates through Battle.net?

Alan Dabiri: Since we did just ship Wings of Liberty, obviously we want to support that product and focus on issues we need to correct. Balance is always a focus we want to make sure we get right. We are kind of split right now in terms of our work. We're supporting the game, we've released a few patches, we have more coming, and at the same time we've got people working on Heart of the Swarm. Now is really a good time when the designers and the artists are jamming on all the design ideas for it, and the programmers can hit up any bugs we missed, things we can improve. We've added features in patches, and we're adding more coming up here. It is a bit of a balancing act, but it's one that we're used to, just because we've supported our games well after their initial launch.

Slashdot: Can you talk a bit about the upcoming 1.2 patch?

Alan Dabiri: 1.2 is our next feature patch. We've got a few things going in. Probably one that people have been asking for a lot and they'll be excited to hear about is chat channels. We've also added customizable hot-key support. We shipped with a few different profiles you could pick from, but now we've actually implemented an interface where people can completely configure the hotkeys and set them up the way they want. We're going to have more balance stuff as well, and that's dependent on the feedback we get with our most recent patches, although things have been looking pretty good in that department.

Another thing from the Battle.net competitive side is that we're adding a couple of new leagues. I don't know if both of them are going to get into 1.2, but I believe the first one, Masters, will be going in. Right now the Diamond league is a top league, and it's a pretty wide range. You get people who [barely] meet the cut all the way to the top. The Masters league is a desire to split up that top a little bit and make it the top couple percent of those people. Then, we're eventually going to come along with the Grandmasters League, and that is going to be the top 200 people, and literally it's by invitation. These are really the cream of the crop, on the level of the pro-gamers.

Slashdot: Are you working to publicize or promote these leagues so that the average player can more easily view them?

Alan Dabiri: Yeah, that's another thing that's coming along. There are already these tournaments going on, right? One thing we want to do is to start showcasing replays from these games, so people can watch and say, "Oh, there's that awesome match between FruitDealer and whoever." So, we're going to start putting them right on the front [page]. There's basically a news carousel where you get the latest news, and we're going to start throwing games on there where you literally click on that and it will launch the replay. So we're going to start showcasing those top-level games.

Slashdot: A lot of your focus for this Blizzcon has been the map editor, the custom games, and the new tools. Is it one of your goals to encourage map makers to think of themselves as software developers in their own right?

Alan Dabiri: Yeah, absolutely. Here's the thing. We're making StarCraft 2, right? So we're bound by some stuff, like lore, etc. These guys -- the sky's the limit. And they've got such creativity, such imagination, that we just want to give them tools to make stuff, and let them do whatever they want. So one thing that's nice, when you say software developers or software engineers -- with the editor, you don't need a programming background or anything like that. We've made it in a way that while it is a very involved tool with a lot of complexity to it, you can get someone who's not a programmer making these maps. So that's the cool thing about it. We absolutely want to encourage that, because from our previous games, people do amazing stuff. We go on Battle.net and we look at these custom game lists, and we get blown away. We're not even sure how they did half this stuff.

Slashdot: Have you thought about releasing some sort of guide or tutorial for map making?

Alan Dabiri: Yeah. We realize that there's quite a bit of knowledge that has to be built up, and we are working on some documentation, some tutorials. I don't know the timelines for all those rolling out, and at what level, but I know we were at one point talking about having a dedicated website, almost like a Wiki, where we would post all this information. We've already got a forum for this, and we've had people responding to those posts. But we want to take even a more active role than that, where people can start asking questions and we can answer and everyone can benefit from the answers. We don't want people just floundering around. We'd like to help them out any way we can.

Slashdot: At one of the panels, it was mentioned that you'd be releasing some new tools. Will those be released incrementally, or can we expect them alongside Heart of the Swarm?

Alan Dabiri: The four maps we've show here? These are 100% made in the editor. Anyone outside can make these right now. So, using the editor, you kind of have every tool you need. The one thing they're missing right now is being able to generate the art assets, because we have specific file formats for our models and textures and whatnot. And so, just like we did with Warcraft 3 -- shortly after we shipped, we released our exporter for Macs, and stuff like that -- we're going to do the same here, too. We're working on our tools. We don't have a timeline yet on when we'll be able to get them out, but we want to get them out as soon as possible.

The funny thing is that some guys have reverse-engineered our formats, and they're already putting assets into the game! But they're missing the bells and whistles. There are a lot of features in the engine. So we're like to get that out to help them. Another thing we've been talking about -- in our campaign, if you're seen our in-game cinematics, you've got these awesome-looking characters talking to each other, and they have facial animation. They're lip-syncing. We want to get that out to the end user too. So, literally, they can make their own movies in the engine. [They can have] a mini-campaign they've created, and use our story-mode space to make their guys talk as well. I'm super excited to see what happens when that gets out.

Slashdot: We saw how the Outbreak mission in the single-player campaign inspired the Left 2 Die custom game. Will we see the reverse happen? (Outbreak was a mission in which zombies rose and swarmed your base at night. The onslaught abated when the sun rose, and you had a brief window to rebuild and go on the offensive before you had to bunker in again. The Left 2 Die custom game expands on that concept and makes it a co-op fight.)

Alan Dabiri: I would say that it already has happened. Some of the maps in our campaign were kind of inspired by concepts from Warcraft 3 maps. If you've played through the campaign, you know it's not the same as the multiplayer game. Every mission has a unique twist on it, and some of those twists have come from either mods that we've made or other styles we've played in different game, and then also new stuff we've come up with.

Bob Colyaco: Specifically, the prison-break mission, where you play as Tosh — that's very DotA-like.

Alan Dabiri: I think that's definitely going to happen some more, too. That's the cool thing about the campaign space: we can play around. Multiplayer, we've got to make sure that it's balanced — we can't really screw around with just going crazy. On the campaign side, we still want it balanced, but we can flex our muscles a little more, and really play around with cool game types.

Slashdot: Is it getting tougher to come up with new twists for the campaign?

Alan Dabiri: I don't know. There are a lot of ideas out there. There are a lot of ideas that were brought up for Wings of Liberty that we didn't have a chance to get to. So, I don't think we'll run out of ideas soon.

Slashdot: Dustin Browder had mentioned that one of the downsides to splitting StarCraft 2 into three campaigns was making, for example, Zerg fans wait for a the Zerg campaign. Are you feeling the pressure to step it up for them in Heart of the Swarm?

Alan Dabiri: Well, since it is a Zerg-focused campaign, you're going to see a lot more spaces in the Zerg world and the characters of the Zerg, so it's definitely going to be a cool thing for Zerg players. But, definitely, for the Zerg campaign and then the Protoss campaign, the guys who play those races are going to be excited about what we have in store for them.

Slashdot: Now that you've had some time to see people play Wings of Liberty, and you've had some time to reflect on it, is there anything that made you think, "We did this well, but we can see a better way to do it for Heart of the Swarm?"

Alan Dabiri: Absolutely. Yeah. There are tons of things. We have these lists of everything want to do, and obviously there's just not enough time in the world. We try to get in as much as we can, but we acknowledge that there are certain areas where we think, "You know what, this is cool, it works in Wings of Liberty, but there's so much more we can do with this." Back at the office before Blizzcon we were talking about replays, and how we want to really expand that area. Everyone loves replays. They like watching their own games, pro games. We want to add a ton of features to flesh out that concept. On Battle.net, there are a lot of things we want to add. The custom game interface, chat channels that we didn't get in for shipping the game but now we're adding. And the game itself; the user interface, the units, the composition -- everything is fair game, and we're always trying to improve.

Slashdot: How will the multiplayer be affected by the launch of the subsequent games?

Alan Dabiri: Since we do have the split of the campaign versus the multiplayer side, we can go crazy on the campaign side without affecting the multiplayer side. But at the same time, if you look at our previous RTS expansions, we have pretty full-featured expansions in the sense that — we don't just ship a new campaign and that's it. We touch a lot of the units, we add new units, we come up with new game mechanics, new tile sets, all that stuff. I think the multiplayer space is open for change and will have change to it. Right now, in fact, we're asking ourselves, "what are the areas where there's maybe something missing from the multiplayer side?" Maybe the composition of units that exist on one side. "What hole is there that we might want to fill?" It's kind of like what happened with Brood War. Brood War identified areas where [we decided] we could add some cool units and make it better. I think we'll do the same for our expansions.

Slashdot: You mentioned earlier the constant need for balance updates. Can you take us through the internal process for identifying and fixing a balance issue?

Alan Dabiri: There are a lot of different ways we get feedback. It comes both from external sources and internally. We've got several balance designers on our team who are, literally, pro-level players. They're top-ranked on the ladder. So, they have an insight into how these things work. But, at the same time, we also view games that are played on the ladder from other top level players. We view replays, we get feedback from pro-gamers, and we get feedback from lower-level players, who maybe aren't so good. That, combined with direct community feedback from forums and other types of communications, we take all that, and the final piece we mix together is the real hard stats. We collect a ton of stats. We know who's winning, we know the race matchup, we know what units you're making. We have all this.

So we can take all this info and provide it to the designers. With all of it taken as a whole, we can sit down and make some intelligent decisions. You know, it's real easy when a guy on the forums goes and says, "Oh, Terran is this," or "Zerg is that," but we're not going to just change the whole race based off of one guy. But when you have all this info, now you can make an informed decision. I think we have a real advantage.. because, to be honest, we didn't have a lot of these things earlier on for StarCraft 1. We'd play the game ourselves and listen to what people were saying outside, and hope that we're doing the right thing. We've got a lot of hard data now that we can lean back on.

Slashdot: Have you thought about cutting out parts of that data that are relevant to a particular debate and working that up into a visualization for the community?

Bob Colyaco: We've done that already, to a degree. If you go to StarCraft2.com, Dustin [Browder] wrote a post showing the win percentages and ratios, compared against regions and compared against leagues. So, for example, Diamond-level PvZ or Gold-level TvT.

Alan Dabiri: Yeah, we've actually provided a lot of data already, and we plan on doing more. Even in Warcraft 3, we had a web page that broke down game matchups, what maps people liked to play on the most. So, I think we have plans to also add that stuff. It's just cool information for the end user. They want to kind of geek out on that. And it helps the debates also, because a lot of times on the forums they'll argue one way or the other, but they don't have really hard stats. They just got done playing against Protoss, and they got beat, so suddenly Protoss is the most overpowered race there is.

Slashdot: Have you heard about the StarCraft AI competition?

Alan Dabiri: I did, yeah! That was really cool. We were excited to see that. The funny things is, with StarCraft 2, we've actually built a system where now they can do that in the game. I know with that [competition], they kind of had to work around the game. So it'll be really cool to see what they do with StarCraft 2, because you can actually make your own AI in a map, using the scripting language, and do different takes on build orders and what the AI does defeat other types of players.

Slashdot: What would you, personally, consider the most underrated aspect of Wings of Liberty?

Alan Dabiri: I think everyone knows we have a full, epic campaign, and everyone has seen those and played those. And everyone knows that for RTS games the multiplayer side is huge — e-sports or just playing with your friends. I think this custom angle that we're pushing at Blizzcon now -- while we had that in Warcraft 3, and we had a bunch of maps, DotA being the most popular one — I think we're really seeing an explosion there, and still a lot of people don't realize it's a feature of StarCraft 2. And now, with Battle.net the way it is, it's so easy to get these maps. You don't have to trade them or find them in other ways, they're all just up there. And we're also going to improve that mechanism so it's even easier.

But, I think this custom game aspect is really cool because it makes it so there are infinite games within this game. You can play the single-player and multiplayer of StarCraft 2, which are both super fun, but now you have another thing that just goes forever. Myself, I love going and just jumping into a map I've never heard of and saying, "Wow. There's so much imagination here. We would never have thought to make something like that." I think that's one area people will play more and more. There are people who are open to the two different sides I've already talked about, but [custom games] are something where we can get even more players in.

Slashdot: Is that served by the large size of the community?

Alan Dabiri: Definitely. Without a lot of people creating these maps and playing them, it's not going to go anywhere. It's going to fall away. But the Battle.net community, the Blizzard community is enormous. We have that momentum, we have that group of people. Literally, every day there are maps on there I've never seen. So, these guys are just coming out with maps, and the awesome ones rise to the top. And obviously there's always going to be the ones that aren't so great. With our community, and how passionate they are -- they were making maps during the beta, before you could even publish maps to Battle.net. I think that's awesome.

Slashdot: Going back to the e-sport aspect, what's your goal for a brand new player loading up StarCraft 2 and looking over the multiplayer? Are you trying to nudge them gently into participating in e-sports?

Alan Dabiri: No, I don't think that's our initial goal. That's one of the cool things about how the matchmaking system works on Battle.net now. You will converge into your comfortable zone. So, if you're playing StarCraft 2 and you're not that great of a player — after those first few placement matches, you will actually be playing against people of your equivalent level, and you'll find really fun games. You are going to go back and forth. You are going to win some and you're going to lose some. I think that's what's cool. We also have other angles; if you don't want to play against other players to start with, because maybe you're nervous and don't want to jump right in, we have co-operative play, where you play with other players against the computer. It's a really cool way to practice strategies.

This came from StarCraft and Warcraft 3, where a lot of players were making these "comp-stomp" games, which were you and a bunch of other guys against a computer. So, we thought, "Hey, you know what, this is a cool thing, a lot of people like to do it, so let's build it into the system." So, the intention when you come to the multiplayer side is to have a fun experience, to have a fun game. And this applies to all levels; you can be horrible at games or horrible at RTS games and still have a fun experience in StarCraft 2. But, at the same time, we cater all the way up to the top, where you've got these extremely high level players, who are just out of control. I watch them play and I'm amazed. The matching system works beautifully where everyone will go to a fun experience.

Slashdot: I know you don't want to say anything about release dates. But can you give us a feel for how much work needs to be done yet on Heart of the Swarm?

Alan Dabiri: Sure. I'll give you an example. For our previous RTS games, it took about a year for expansions to come out. The difference between StarCraft 2 and Warcraft 3 is that we're going to have a whole new campaign, and the campaign is going to be on the same level as with Wings of Liberty. Meaning the quality level, and [it'll be] enormous. So, obviously it's going to take a little more time than what we've previously done. But it's not going to take as long as the original game took. We were building the engine, we were building the infrastructure, so it's definitely going to be in a shorter time frame than that. But we do want to put a lot of features into the expansion. Historically, we've always done this. Our expansions have had tons of things -- it's not just a couple new maps. And we'll continue to do that.

bgweber writes "The StarCraft AI Competition announced last year has come to a conclusion. The competition received 28 bot submissions from universities and teams all over the world. The winner of the competition was UC Berkeley's submission, which executed a novel mutalisk micromanagement strategy. During the conference, a man versus machine exhibition match was held between the top ranking bot and a former World Cyber Games competitor. While the expert player was capable of defeating the best bot, less experienced players were not as successful. Complete results, bot releases, and replays are available at the competition website."

bgweber writes "The StarCraft AI Competition announced last year has come to a conclusion. The competition received 28 bot submissions from universities and teams all over the world. The winner of the competition was UC Berkeley's submission, which executed a novel mutalisk micromanagement strategy. During the conference, a man versus machine exhibition match was held between the top ranking bot and a former World Cyber Games competitor. While the expert player was capable of defeating the best bot, less experienced players were not as successful. Complete results, bot releases, and replays are available at the competition website."

bgweber writes "There's been a lot of discussion about whether games should adapt to the skills of players. However, most current techniques limit adaptation to parameter adjustment. But if the parameter adaptation is applied to procedural content generation, then new levels can be generated on-line in response to a player's skill. In this adaptation of Infinite Mario (with source [.JAR]), new levels are generated based on the performance of the player. What other gameplay mechanics are open for adaptation when games adapt to the skills of specific players?"

bgweber writes "There's been a lot of discussion about whether games should adapt to the skills of players. However, most current techniques limit adaptation to parameter adjustment. But if the parameter adaptation is applied to procedural content generation, then new levels can be generated on-line in response to a player's skill. In this adaptation of Infinite Mario (with source [.JAR]), new levels are generated based on the performance of the player. What other gameplay mechanics are open for adaptation when games adapt to the skills of specific players?"

bgweber writes "Last year, the IEEE conference on Computational Intelligence and Games hosted a competition to determine who could write the best AI for playing Mario levels (YouTube video). This year, the conference has expanded the competition to include a track on level generation as well, where the goal is to generate new levels online procedurally. Submitting an entry is as easy as implementing a Java interface that performs procedural content generation. The implications of this competition are techniques for greatly increasing the replayability of games, since each gameplay session could present new levels to the player."

bgweber writes "The 2010 conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE 2010) will be hosting a StarCraft AI competition as part of the conference program. This competition enables academic researchers to evaluate their AI systems in a robust, commercial RTS environment. The competition will be held in the weeks leading up to the conference. The final matches will be held live at the conference with commentary. Exhibition matches will also be held between skilled human players and the top-performing bots."

A team of researchers has simulated the gravitational interaction of dark matter particles over the course of a hypothetical 13.7 billion years. They found that the particles tended to form clumps large enough to assist in the formation of galaxies. The results contradicted observations from previous, smaller studies, but they lent support to an unrelated simulation of how the Milky Way formed. UCSC's press release is also available. Quoting ScienceNews: "The clumps of dark matter in the simulation have densities that are remarkably similar to densities that a University of California, Irvine research group found when simulating the formation of the Milky Way and its satellite dwarf galaxies, says James Bullock, the astrophysicist who leads the UC-Irvine group and was not involved in the new study. 'This is a remarkable success of the particular model simulated and adds strong support to the idea that the dark matter is made up of particles that are "cold." There are a number of planned experiments aimed at detecting the dark matter that are betting on it being cold, so this is generally good news for the community,' Bullock says. And, [study co-author Piero Madau] notes, larger simulations that might help constrain the nature of dark matter even more are already in the works."

paleshadows writes "Researchers at UCSC developed a tool that measures the trustworthiness of each Wikipedia page. Roughly speaking, the algorithm analyzes the entire 7-year user-editing-history and utilizes the longevity of the content to learn which contributors are the most reliable: If your contribution lasts, you gain 'reputation,' whereas if it's edited out, your reputation falls. The trustworthiness of a newly inserted text is a function of the reputation of all its authors, a heuristic that turned out to be successful in identifying poor content. The interested reader can take a look at this demonstration (random page with white/orange background marking trusted/untrusted text, respectively; note "random page" link at the left for more demo pages), this presentation (pdf), and this paper (pdf)."

paleshadows writes "Researchers at UCSC developed a tool that measures the trustworthiness of each Wikipedia page. Roughly speaking, the algorithm analyzes the entire 7-year user-editing-history and utilizes the longevity of the content to learn which contributors are the most reliable: If your contribution lasts, you gain 'reputation,' whereas if it's edited out, your reputation falls. The trustworthiness of a newly inserted text is a function of the reputation of all its authors, a heuristic that turned out to be successful in identifying poor content. The interested reader can take a look at this demonstration (random page with white/orange background marking trusted/untrusted text, respectively; note "random page" link at the left for more demo pages), this presentation (pdf), and this paper (pdf)."

paleshadows writes "Researchers at UCSC developed a tool that measures the trustworthiness of each Wikipedia page. Roughly speaking, the algorithm analyzes the entire 7-year user-editing-history and utilizes the longevity of the content to learn which contributors are the most reliable: If your contribution lasts, you gain 'reputation,' whereas if it's edited out, your reputation falls. The trustworthiness of a newly inserted text is a function of the reputation of all its authors, a heuristic that turned out to be successful in identifying poor content. The interested reader can take a look at this demonstration (random page with white/orange background marking trusted/untrusted text, respectively; note "random page" link at the left for more demo pages), this presentation (pdf), and this paper (pdf)."

paleshadows writes "Researchers at UCSC developed a tool that measures the trustworthiness of each Wikipedia page. Roughly speaking, the algorithm analyzes the entire 7-year user-editing-history and utilizes the longevity of the content to learn which contributors are the most reliable: If your contribution lasts, you gain 'reputation,' whereas if it's edited out, your reputation falls. The trustworthiness of a newly inserted text is a function of the reputation of all its authors, a heuristic that turned out to be successful in identifying poor content. The interested reader can take a look at this demonstration (random page with white/orange background marking trusted/untrusted text, respectively; note "random page" link at the left for more demo pages), this presentation (pdf), and this paper (pdf)."

NickFitz writes "Students at UCSC have recreated the first level of Donkey Kong using 6,400 Post-it notes stuck to the windows of the E2 building. It took a team of about 10 people five hours to complete the work, which will remain in place until May 1. There's a time-lapse video of the construction process."

QuantumCrypto writes "The All-wavelength Extended Groth strip International Survey (AEGIS), a collaborative effort involving nearly 100 scientists in half a dozen countries, revealed a new principle in the formation of all galaxies, from disk-like spirals, cloud-like ellipticals, and just irregulars. In essence the morphology of the galaxies depends on total mass involved and the internal speed it generates. 'By defining a new speed indicator, their analysis has managed to make sense out of very chaotic-looking objects,' said Sandra Faber, professor of astronomy and astrophysics at the University of California, Santa Cruz."

bigmammoth writes "C-SPAN's bid to "liberate" the House and Senate floor footage has re-emerged and been shot down. In an aim to build support a recent New York Times editorial called for reality TV for congress. But what is missing from this editorial is the issue of privatization and the subsequent restriction of meaningful access to these media assets. Currently the U.S. government produces this floor footage and it is public domain. This enables projects such as metavid to publicly archive these media assets in high-quality Ogg Theora using all open source software, guaranteeing freely reusable access to both the archive and all the media assets. In contrast C-SPAN's view-only online offerings disappear into their pay for access archive after two weeks and are then subject to many restrictions." (Continues) "If C-SPAN succeeds, reusable access to floor footage will be lost and sites such as metavid will be forced to stop archiving. Because of C-SPAN's zealous IP enforcement metavid has already been forced to take down all already 'liberated' committee hearings which are C-SPAN produced. Fortunately, the house leadership sees private cameras as a loss of 'dignity and decorum' and will be denying C-SPAN's request."

bigmammoth writes "C-SPAN's bid to "liberate" the House and Senate floor footage has re-emerged and been shot down. In an aim to build support a recent New York Times editorial called for reality TV for congress. But what is missing from this editorial is the issue of privatization and the subsequent restriction of meaningful access to these media assets. Currently the U.S. government produces this floor footage and it is public domain. This enables projects such as metavid to publicly archive these media assets in high-quality Ogg Theora using all open source software, guaranteeing freely reusable access to both the archive and all the media assets. In contrast C-SPAN's view-only online offerings disappear into their pay for access archive after two weeks and are then subject to many restrictions." (Continues) "If C-SPAN succeeds, reusable access to floor footage will be lost and sites such as metavid will be forced to stop archiving. Because of C-SPAN's zealous IP enforcement metavid has already been forced to take down all already 'liberated' committee hearings which are C-SPAN produced. Fortunately, the house leadership sees private cameras as a loss of 'dignity and decorum' and will be denying C-SPAN's request."

bigmammoth writes "C-SPAN's bid to "liberate" the House and Senate floor footage has re-emerged and been shot down. In an aim to build support a recent New York Times editorial called for reality TV for congress. But what is missing from this editorial is the issue of privatization and the subsequent restriction of meaningful access to these media assets. Currently the U.S. government produces this floor footage and it is public domain. This enables projects such as metavid to publicly archive these media assets in high-quality Ogg Theora using all open source software, guaranteeing freely reusable access to both the archive and all the media assets. In contrast C-SPAN's view-only online offerings disappear into their pay for access archive after two weeks and are then subject to many restrictions." (Continues) "If C-SPAN succeeds, reusable access to floor footage will be lost and sites such as metavid will be forced to stop archiving. Because of C-SPAN's zealous IP enforcement metavid has already been forced to take down all already 'liberated' committee hearings which are C-SPAN produced. Fortunately, the house leadership sees private cameras as a loss of 'dignity and decorum' and will be denying C-SPAN's request."

bvc writes "We Launched the Java Open Review Project today. We're reviewing open source Java code all the way from Tomcat down to PetStore looking for bugs and security vulnerabilities. We're using two static analysis tools to do the heavy lifting: the open source tool FindBugs, and the commercial tool Fortify SCA. We can use plenty of human eyes to help sort through the results. We're also soliciting ideas for which projects we should be reviewing next. Please help!"

Baldrson writes "Within the first week of the announcement of The Netflix Prize a team has already beaten Netflix's own movie recommendation algorithm. This is pretty impressive given the previously quoted researcher who said: 'You're competing with 15 years of really smart people banging away at the problem.' The team is WXYZConsulting.com apparently registered by a data mining professor named Yi Zhang. Congratulations are in order for Netflix and Prof. Zhang's team who are demonstrating, yet again, the power of prizes to accelerate progress."

MondoMor writes "Apparently, the atmosphere above Earth's strongest storms acts like a particle accelerator, according to a UC Santa Cruz paper. TGFs (Terrestrial Gamma ray Flashes) may occur as seldom as 50 times a day, 'but the rate could be up to 100 times higher if, as some models indicate, TGFs are emitted as narrowly focused beams that would only be detected when the satellite is directly in their path.' I'm glad the gamma-ray bursts are directed into space."

blamanj writes "Scientists at UC Santa Cruz have been using computational techniques to 'reverse engineer' the DNA of extinct species. David Haussler and colleagues created a hypothetical portion of ancestral mammalian DNA and let a computer model simulate the process of evolution. Then they made their algorithm work backward from these descendants, to see if it could recreate the original ancestor."

An anonymous reader writes "Scientists have long probed Earth's interior by monitoring seismic waves (if earthquakes don't make them, they can be induced with explosives, and one nuke test actually triggered an earthquake!), which reveal the inner structure of the planet. But what if the method is wrong? LiveScience reports on a new study suggesting Earth is like a pinball machine, with sound waves careening around before they get to the surface. What is interpreted as a broad layer change could be nothing more than a localized density variation."

prof_peabody writes "The IODP (Intergrated Ocean Drilling Program) is about to get rolling in a couple of days. If you live in one of these countries then your tax dollars have contributed to the construction of the giant drillship Chikyu, which was launched a little while back (project timeline). The American contigent website is loaded with info and obligatory acronyms. The first leg of the IODP will investigate how water flows through rock formations beneath the seafloor during an eight-week expedition this summer to the eastern flank of the Juan de Fuca Ridge off the coast of British Columbia. Some of you geeks with beards may remember the DSDP (Deep Sea Drilling Project) or the recently completed ODP (Ocean Drilling Program). The real advance in the new program that will cost well over a billion dollars is the IODP riser drill ship that 'will provide a way to drill into continental margins where oil and gas deposits can cause drilling safety concerns and into regions with thick sediment sections, fault zones, and unstable formations.' A good overview of the IODP can be found here, and the necessary references to Megalodon and none other than The Core."

Thanks to B3ta for its interview with Wes Cherry, creator of Solitaire for Windows, as installed on "hundreds of millions of machines worldwide." Cherry discusses an 'Easter egg' left out of the final version ("There was a 'boss-key' which when pressed would display some random .C code. Microsoft made me remove that"), the all-important card back designs ("My fave is the dealer with the Ace crawling up and down his sleeve, which is a reference to a Grateful Dead song, 'Doin' that Rag'"), and bizarre benchmarking concepts using Solitaire ("At one point, a computer magazine proposed a SolMark computer speed test: The faster the cascade, the faster your computer.")

Slashback tonight brings you more information on homemade railguns, the future history of SCO (seen from the past), one website's response to alleged RIAA mendacity, a legal victory for famous web jerk Tucker Max, and more -- read on for the details.

Please don't point that thing at me. BoomZilla writes "If you thought that the recent Gauss Gun article on Slashdot was spiffy, check out Jengel & Fatro's Rail Gun Page. Everything you need to know about the physics behind building your very own rail gun. Ever used the Quake rail gun and lusted after the real thing? Here's your opportunity."

Telepathic telegrams work as well as ever, though. markgo2k writes "After the web site experienced 'an unusually high number of visitors,' the White House modified the contact page and added a prominent link to president@whitehouse.gov. Here's the latest NYT story (once again, reprinted here in the non-subscription Seattle PI). Of course, the White House is still confused on the difference between the New York Times effect and the Slashdot effect. It's not mentioned in the story, but there is also now a feedback link to submit comments to the 'White House Web Development Team,' if you want to drop them a line..."

It's half-Greek to me. In response to the recent story on perpendicular data storage in next-generation hard drives, Anonymous Coward writes "Here is a better overview of Perpendicular HDD technology. Here is a real detailed scientific article that seems to be written in Greek."

They're off my Christmas card list, too. Techfocus ran an interview with Fred von Lohmann earlier this year. Now, an anonymous reader points to an update on their site: "Effective immediately, the RIAA and MPAA will need to find another way to get to Techfocus. In response to their legal targeting of individual file-swappers, access from their known networks to this site has now been blocked. While it may still be possible for them to access Techfocus via address ranges which we're not aware of, they'll otherwise have to use non-RIAA and non-MPAA networks to view the site."

Techfocus cites three reasons for the denial, the top one being that the RIAA took advantage of the interview with von Lohmann, "quoting him out of context in a manner which could lead readers of their materials to believe that we supported their efforts. This could not be further from the truth."

The secret is to predict enough things. An anonymous reader points out this article from early 2000 citing Gartner analyst Al Hilwa's prediction that Linux is "probably going to kill SCO UnixWare," writing "As you can read, SCO's end was predicted near perfectly." I think "hinted at" is more accurate, since SCO is still alive and at least making a good show of kicking, but it's interesting to revisit a story about SCO which mentions that "industry observers thought that the company would be Linux's first victim," back when Project Monterey was a going concern.

A victory for discourteous boors everywhere. aeaas writes "The beauty queen Katy Johnson dropped her suit against Tucker Max over the posting of stories from their relationship on his website. This story was first brought up in the context that he was forced to take down stories relating to her without holding a hearing or notifying Max prior to it. This is unusual in American law."

A quarter mil is a lot of suffering, even in Canadian money. Skippy321 writes "Justified or not, Ghyslain Raza--better known as the Star Wars Kid--is suing the four students who posted his homemade video of himself doing acrobatic "sword-fighting stunts" on the Internet for $250,000. He claims that he has suffered harassment and persecution. It's also interesting how the article states that he quit high school due to this video, at only 15 years of age. Although things aren't so bad for him -- here's a petition for him to get a role in Episode III."

JoeRobe writes "Researchers at UCSC have simulated a possible outcome of an impact by asteroid 1950DA when it passes near us in the year 2880. Note that there is a 0.3% chance of impact during that encounter. In the event that it impacts in the Atlantic, they predict that the '60,000 megaton blast' would create 400 foot waves along the east coast. In addition to an assessment of the danger, their studies point out the resulting geologic features that we should be looking for now, which would indicate where and when such impacts have occured in the past."

learithe writes "A new Middle Earth MMORPG, Middle Earth Online, has just been announced by Turbine, who produced Asheron's Call 1 and 2 with Microsoft. It looks to be just as pretty and cpu/graphics card intensive as AC2. More (flash-free) information can be found at IGN and Gamespy."

jason writes "tangle is a system for what we call "web page entanglement". tangle creates links between pages automatically based on how users move from one page to another. tangle proxies connect together in a peer-to-peer network for scalability: as users surf the entangled web, they are passed from proxy to proxy. Each proxy serves as an expert for a particular subset of web pages. For example, you can take a look at the entangled version of the GNU homepage as seen through a tangle proxy. tangle alpha2, the first public version, has just been released. See http://tangle.sourceforge.net for more information, or read on..."

jason continues:

"By viewing the web through a tangle proxy, you can see the connections and associations left by those who surfed the web before you. By surfing the web using tangle, you also leave behind connections and associations for others who will surf in the future.

When you exit one page and enter another (by clicking a link or performing a search), a two-way link is created between the pages. As users surf through a particular page over time, tangle keeps track of popular ways to get to the page and popular places to go next. These entry and exit links are displayed at the top of each page, sorted by popularity.

Clicking on one of these entry/exit links tells tangle that you think the link is relevant and useful (like a vote for the link) and increases the link's popularity. In other words, if a user thinks of something relevant while reading a page and performs a search for it from that page, tangle gauges how others react to that association over time.

tangle is similar in some ways to the closed-loop hypertext system Everything2, though tangle works for the web at large.

We have several tangle proxies up and running. The tangle proxy software is also available for download.

A note for the paranoid:
Though tangle keeps track of web usage patterns, the focus is not on tracking the habits of individual users, but on tracking the trends of an entire community of users. tangle is GPL'd open source [source here], so you can see for yourself: clicking a link through a tangle proxy simply bumps up the links popularity---user IP addresses are completely ignored."

wezelboy asks: "I need independent verification of a data corruption problem with the Tyan Thunder LE mobo during DMA writes to an IDE drive. The problem manifests itself under heavy disk load in large files, does not change file size, and only corrupts 4 byte chunks. Thus, it is not easily spotted. We are running Red Hat 7.1, but this corruption may be reproducible on other operating systems." There's more detail of this problem in the article. Any of you Tyan Thunder motherboard owners care to help track this problem down?

"A simple way to verify this problem follows:

Using a Red Hat 7.1 install (2.4.2 kernel, but other 2.4 kernels exhibit this problem) and a file /tmp/x that is over 500MB in size, "cp /tmp/x /tmp/y", and compare the md5 checksums using md5sum. Keep doing this, deleting /tmp/y each time before you do.

/tmp/x and /tmp/y will continuously show the same size, but will sometimes have different md5 checksums. A diff between the files usually shows that 4 bytes are moved to another part of the file. You may have to go through a number of iterations before you see it. We have a more sophisticated test script available, just e-mail me at here and I'll send it to you.

Tyan claims that this problem is a limitation of the Serverworks LE chipset, however I have not been able to reproduce this problem on any of the Asus CUR-DLS motherboards (which use the same chipset!).

If anyone can reproduce this problem on a different Serverworks LE mobo, I'd like to hear about it. Thanks!"

wezelboy asks: "I need independent verification of a data corruption problem with the Tyan Thunder LE mobo during DMA writes to an IDE drive. The problem manifests itself under heavy disk load in large files, does not change file size, and only corrupts 4 byte chunks. Thus, it is not easily spotted. We are running Red Hat 7.1, but this corruption may be reproducible on other operating systems." There's more detail of this problem in the article. Any of you Tyan Thunder motherboard owners care to help track this problem down?

"A simple way to verify this problem follows:

Using a Red Hat 7.1 install (2.4.2 kernel, but other 2.4 kernels exhibit this problem) and a file /tmp/x that is over 500MB in size, "cp /tmp/x /tmp/y", and compare the md5 checksums using md5sum. Keep doing this, deleting /tmp/y each time before you do.

/tmp/x and /tmp/y will continuously show the same size, but will sometimes have different md5 checksums. A diff between the files usually shows that 4 bytes are moved to another part of the file. You may have to go through a number of iterations before you see it. We have a more sophisticated test script available, just e-mail me at here and I'll send it to you.

Tyan claims that this problem is a limitation of the Serverworks LE chipset, however I have not been able to reproduce this problem on any of the Asus CUR-DLS motherboards (which use the same chipset!).

If anyone can reproduce this problem on a different Serverworks LE mobo, I'd like to hear about it. Thanks!"

It's the scientific achievment of our generation; what can you say about the mapping of the human genome? But here's a story behind the story. parvati turned us on to this NYT article about James Kent, who wrote the gene assembly program GigAssembler last June. It turns out that, thanks to his code, the public Human Genome Project had actually finished its work three days before the private effort by Celera Genomics -- a feather in their cap and a boon to public science. The head of Celera was "astonished" to learn of this grad student's genius -- ten thousand lines of C in a month, and why? -- "because of his concern that the genome would be locked up by commercial patents if an assembled sequence was not made publicly available for all scientists to work on." (The debate over public vs. private science continues to rage; see this Seattle P-I article, which discusses among other things the ethics of NDA'ing scientific data produced for profit.)

Update: 02/13 02:26 PM by J : Thanks to tlunde for finding the link to GigAssembler and thus clarifying which language it was written in.

webword writes: "There is an article in the July 2000 issue of Crosstalk (The Journal of Defense Software Engineering) about gathering requirements in Open Source projects. This is especially interesting because most commercial projects are driven by the requirements gathering processes (or the 'SDLC') whereas most Open Source projects are written to 'scratch an itch'. Let's face it, Open Source requirements are almost always in the mind(s) of the programmer(s), not on paper. However, if the requirements are captured, and shared, the quality and speed of development will improve." How many articles like this have we seen? How many open source applications have ideas like these spawned? I remain skeptical.

CMU_Nort writes: "The San Francisco Gate has a story about the completion of the human genome project. Apparently the University of California at Santa Cruz has put the Genome online for downloading here. I don't know about you, but I think this sort of sharing is very cool. We finally have the source for human beings. Now if only they'd GPL it."

v@mp writes "I noticed a few posts today about space, which reminded me that there will be a total lunar eclipse in North America and Western Europe on Jan. 20. The moon will turn a deep red color for little over an hour around 8 p.m. on the west coast and 11 p.m. on the east coast. I'll see you all "under a blood red sky"--U2. "

etphone writes "One of the Gods of information theory, David Huffman, has passed on: here's the official the press release. Damn, he was a good professor too..."

etphone writes "One of the Gods of information theory, David Huffman, has passed on: here's the official the press release. Damn, he was a good professor too..."

Druppy writes "I just got an email about this article in our local paper here in Santa Cruz. It's basically about illegal copies of movies like The Matrix being moved over campus lines. Needless to say since my school was mentioned 4-5 times in the article the administration is starting to crack down. " So are people sneaking cam corders in to movie theaters now? I get nervous just trying to sneak in a candy bar.

Pont writes "Sony failed to get an injunction against Bleem, a Playstation emulator for the PC. " This is a nice step in the right direction for an industry struggling to come to grips with annoying little problems like emulation. Way to go Bleem. Now how about a Linux Port?