It's worth noting that the proposed heavy lift design can put a hell of a lot more stuff (125 metric tons) into orbit than the Shuttle could.
That's very fine, except people have been talking for decades about clean-burning, re-usable, vertical take-off and landing, single-stage to orbit vehicle designs with much greater capacity. Also, space tugs, electric plasma drives, solar sails. Now, instead of taking a step forward, they're retreating to designs the oldsters grew up with. It's pathetic.
We don't need 125-ton space probes. None of the ones out there came close to maxing out the Shuttle's ~30 tons. We need more wee probes going lots of different places, and we need to listen to the ones that are already out there. (Voyager 1 and 2, remember? They mean to shut them down to help pay for this boondoggle.) We need a new Ulysses, and more SOHOs, and x-ray, infrared, and ultraviolet telescopes. We need ground support for them, and data analysis. (We'd have 'em already if not for those decrepit ISS tin cans.) Already planned missions, and still-working probes, will be shut down in favor of sending camping trips to the moon. Data already collected will molder on back-up disks for want of analysis.
For what? Low expectations and weak imagination. You can be sure that in a couple of years it will all become an excuse to cut the budget in half. The (popular!) space-probe stuff will already have been cut, leaving just the obvious boondoggle. Then they'll be able to hand the rest over to the Air Force, to do just the part it's interested in.
An underground city on the moon is probably the best way... to avoid micrometeorites and cosmic rays...
The best way to do that is to stay right here.
Since lunar gravity is about 1/10th what we are used to, that will mean that mining and digging out these tunnels will be 10x easier.
One sixth. But this evades the question. Having dug out a warren, what then? Live out your old age in comfort? For anything else you want to do (and probably that too), a hole in the moon is likely to be the worst place to do it in. If you're not willing to discuss what you want to do in your warren, it's probably not anything a reasonable person would want to pay Halliburton to have done for them.
You want building materials? Use asteroids. You want oxygen? Use asteroids. You want organics? Moon's not got 'em. You want solar power? Most of the moon hasn't got that for two weeks at a time. Want a garbage dump? Hmm, that works.
One of us isn't thinking things through, and it's not me.
I didn't find a tarball of source code, just the ISO image. When I loopback-mount the ISO image, I don't find anywhere near 80M of stuff. Is the source on the ISO image?
With all due respect to the engineers at NASA, this looks like the nastiest thing the Agency has ever been railroaded into. The solid rocket boosters were the worst feature of the Shuttle design; it was supposed to have a hydrogen first stage until NASA hit a budget crunch and strapped on the damned missiles. They're appalling polluters, unconscionably expensive, and fragile. (Why are they made in pieces and shipped to Florida? Jobs in Utah. If they had been built at Cape Canaveral they'd be in one piece, and the first boom wouldn't have happened.)
We can barely afford to keep a low-earth-orbit space station from burning up in the atmosphere, never mind actually doing anything useful. (The crew spends all its time on maintenance.) Now we're supposed to keep a lunar station going using super-sized Apollo designs that were abandoned decades ago because they were too wasteful. What are the crew supposed to do on the moon, anyway? Dig? What are they supposed to do on Mars? It's hard to imagine more useless lumps of dead rock.
Asteroid missions (manned or not) would be interesting. Space elevators would be very interesting. Even another Cassini (for Jupiter) would be interesting. Instead, they're gutting JPL. Anybody who says this is something other than a disaster for NASA and for space exploration is drinking Kool-aid.
Re:Watch a little more closely ...
on
Deep in the Core
·
· Score: 1
Some of Halton Arp's "associated" systems in particular have been quite strongly discredited.
Some? It would take more than isolated examples to weaken his statistical case. Have the selected cases been shown to be somehow representative of the rest?
If they really have something, they can make their case in a compelling way and people will listen.
Astronomers appear to listen until they establish that the evidence seems to contradict Big Bang cant. The process seems to involve rejecting evidence not accompanied by an acceptable theory to incorporate it into a mainstream model: without a theory, the evidence doesn't count. This is, of course, entirely counter to scrupulous scientific practice.
In Tifft's case, everyone breathed a sigh of relief when high redshifts didn't seem quantized the same way, but that didn't make the quantized low redshifts go away, however much everyone wished otherwise. The journal editors who apologized for publishing his papers are a disgrace to the whole field.
...[or] they die off and we don't worry about it anymore
Astonishing. Evidence doesn't go away just because the person reporting it does. It's every scientist's job to keep anomalous evidence on the table, because that's where all fundamental progress comes from.
I've got hundreds of examples of quasars with stars at exactly the same redshift.
What significance do you attach to that? Arp identifies (trails of) such stars himself.
Re:Watch a little more closely ...
on
Deep in the Core
·
· Score: 1
Are you talking about those galaxies that we see as they were 12 billion years ago...
I'm sorry, I should have been clearer. We see galaxies whose redshift suggests (according to the standard interpretation) that they're 12 billion light years away, and thus formed in the first billion years after presumed recombination, made of stars that had to be 4 billion years old at the time. I.e., 12+4 > 13.7.
As an astrophysicist, I'd like to be an apologetic for the standard cosmological model...
That's to your credit. Most astrophysicists would prefer to pretend that, e.g., quantized redshift as referenced to the CMB rest-frame in low-redshift galaxies, or to angularly-nearby low-redshift galaxies in the case of high-redshift galaxies and quasars, don't exist. Most would prefer to pretend that high-redshift quasars (e.g. z=2.11) physically in front of low-redshift opaque galaxies (e.g. NGC 7319, z=0.0225) don't exist.
Most seem to prefer to pretend that MHD conditions apply to the dynamics of interplanetary and interstellar plasmas, so they can pretend it's all just "hot gas". I have a private e-mail from a well-known astrophysicist asserting, without apparent embarrassment, "Plasmas behave as ideal gases on scales much larger than the Larmor radius of motion around the magnetic field." He admits most graduate students in astrophysics, still, never attend a laboratory class on plasma dynamics, and just work artificial problem sets using the MHD approximation.
Re:Watch a little more closely ...
on
Deep in the Core
·
· Score: 1
Apparently you haven't studied these things. The universe is 13.7 billion years old, [various true statements]...
Sorry, the "13.7 billions years" figure still counts as speculation, supermassive black holes or no supermassive black holes. We can see galaxies that are supposed to be over 12 billion years old made of stars that have to be 4 billion years old. "At least 10 billion years" is defensible without calling up spirits from the vasty deep.
Getting sucked in?
on
Deep in the Core
·
· Score: 4, Informative
does that mean it is slowly being sucked in
According to the original paper from 2002, the star is nowhere near close enough to be "tidally disrupted", so it's just orbiting. (What it says is that even at closest approach, it's still 70x too far way.)
With all those stars whipping around, though, it wouldn't be hard to get the occasional star either entirely ejected, or potted right in. More usually, an orbit would be changed so that it approaches closely enough on each orbit to have a bit of mass (say, a trillion tons) stripped off, and gets used up over the course of a few thousand years. Of course at some point we wouldn't be able to see it any more, so there could be a bunch of those happening right now.
Probably most of the mass moving near it is non-radiating low-density plasma whose motion is controlled less by gravitation than by unimaginably intense electromagnetic fields. We see stars, but there's lots else going on in there we can't see.
This certainly counts as positive evidence of a black hole or its moral equivalent. Note that the details date from 2002. Before 2002, we had a lot of conjecture. Now we have proof.
Everybody who was skeptical before 2002 (or who hadn't heard about this yet) was right to be skeptical. Given this, there seems no room left for skepticism about supermassive whatsits.
As they note, there remains now the mystery of how they got so much mass to concentrate in one place. Stars don't forget all about conventional orbital dynamics just because they've spotted a black hole somewhere not too far off.
The addition of the write-ahead log was a key factor in making PG consistently faster than MySQL.
If your application is limited by database writing speed, then chances are you shouldn't be using a relational database for that part of the job. (Eight hours is a damned long time to be banging a disk head around.) The right way is to get the data onto the disk in the quickest, un-relational- transactional-est way possible, and then (presuming you are actually going to run SQL queries against it) index it. Every practical database engine provides a way to do that because everybody runs into problems that don't fit neatly in an SQL-shaped box.
It's true that I don't know anything about your requirements, datasets, or schemas, but I do know the difference between good and bad engineering.
... on what basis do you say that MySQL has the edge on development community size?
Mainly because the components of MySQL are worked on by different communities. Innobase is one, Berkeley DB is another. However, now that you mention it, the requirement to sign copyright over to MySQL has to reduce their core developer base. Back when I was on the PG list, though, there were only a half-dozen people doing serious work, but... that's been five years. How time flies.
If you "did a benchmark with the application my group is developing using MySQL and PostgreSQL and MySQL was much faster" recently, you did it wrong. PostgreSQL passed MySQL long ago, if you compare them apples-to-apples. To choose one over the other based on speed is usually a mistake.
Choose based on administration complexity for a minimal setup, if you like (favoring MySQL), or on license restrictions (favoring PostgreSQL), or on features (PostgreSQL, for now), or on development community size (MySQL), or on development community competence (PostgreSQL).
PostgreSQL is not, in fact, derived from Ingres. Commercial Ingres was derived from the same original (Free) codebase as was Postgres. There's probably very little if any of that code remaining in either. Commercial Ingres was opened by CA recently, but there's little reason to expect it to go far, given the modern alternatives it would have to compete with.
Postgres is a poster child for astonishing progress in a short time from a codebase of purely academic interest to a real, industrially useful product. Over the last ten years its code got radically cleaned up, then it got journaling, increasingly automated administration, and lately replication.
Given the rate of progress in both PostgreSQL and MySQL, Oracle has reason to be worried for its future. It's worried enough, for example, to have bought out Innobase, just to threaten (or tax) the proprietary arm of MySQL, Inc. Oracle certainly has no use for Innobase, so the only plausible reason is that they are worried. They should know, and we should know better than to second-guess them.
There are still things that can be done with Oracle but not PostgreSQL, but they're getting fewer all the time. At some point they will be too few to fund Oracle's corporate expenses, and the money Oracle charges for an annual license will be better spent beefing up PostgreSQL to be able to do the job indefinitely. We may reasonably expect national laboratories to commission such work; they have stringent needs, the budget to satisfy them, and plenty of favorable experience with Free Software projects.
It's no wonder you have trouble finding "OGG" support. There's no such thing. The post you replied to got it right: it's "Ogg".
Too, you really can get Ogg/Vorbis support on older iPods (now very cheap, on eBay), thanks to the excellent crew at http://ipodlinux.org/. Yes, it will dual-boot. Booting under Linux you can transfer files directly from yours to anybody else's, whatever they are running. I don't know whether you can install linux on them that way, but if you can it would be the polite thing to do when you get connected.
Ellen warned us of that when she said it stumped Donald Knuth. It was a far more useful hint than the one she linked to. It certainly is, though, a logic problem: once you establish that there is no purely logical solution, you canknow you must bring to bear what you know about switches, light bulbs, and time, and stir it in. We all have more than sufficient experience with all three. Since nobody who solves it actually experiences the switches or rooms, it remains a logic problem.
... the current administration has overspent so fucking much in the last half-decade that... you'll be paying interest out the asshole for decades."
And guess who those interest payments are going to? (Hint #1: Not you. Hint #2: Nobody you know. Hint #3: I wonder how China paid for its new space program...)
but I'd like to hear from people working on it. Will there be an OOo package with SQLite or something in there, and no Java? (E.g. on Debian and Ubuntu?)
I'd like to hear about Java-free builds. In particular, I wonder whether anyone has made progress plugging in SQLite in place of their Java-dependent database engine. Database access seems to be the only important feature in 2.0 that depends on Java.
While an OOo built with Gcj and Classpath is, apparently, legally unencumbered, the future of the language is uncertain. Some us would prefer, for a variety of reasons, to have OOo not dependent on Java for core features.
Looking at the site now, they want $105 more for the E510n, vs the E510. However, the E510 also comes with a full-year warranty, so the difference is even larger -- maybe $120?
Note well, this isn't a discount for buying without MS -- they're charging $105 extra, and shortening your warranty, for leaving the disk blank.
It's one thing to charge extra. It's quite another to hike up the premium after the Register has called them on it.
You have something against sociology? It's a science too.
Ha ha. Most sociologists remain Marxists.
All of those observations SUPPORT Big Bang cosmology, rather than contradict it.
Only if you ignore the evidence. Light-element isotope ratios were once trumpeted as supporting the Big Bang. Now that they don't match current models, they aren't discussed at polite conferences. Galactic gravitational lensing puts a strict upper limit on mass, leaving no place to put the dark matter. (Unless... yes! Dark energy at work? Or... maybe supermassive black holes?) You could put your dark matter out in those unaccountable megaparsec-wide voids, except you need it in galaxies to explain why their rotation doesn't match gravitational orbital dynamics (but does match laboratory plasma phenomena).
... there is no known preferred direction of the CMBR -- but even if there was, there are anisotropic Big Bang cosmologies with preferred directions.
This is a capsule example of the phenomenon. Big Bang can't be falsified, because all evidence can be reinterpreted, post hoc, in its favor. There have been other pseudo-scientific regimes like that: Marxist economics, Freudian and Skinnerian psychology, Lysenkoist genetics; lately, Intelligent Design has support in high places.
Theories of neutron stars and black holes make specific predictions of what you will see, and those predictions are supported by observations.
Two words: "post hoc"
Ultra high energy cosmic rays, for one, are within the capacity of jets from supermassive black holes.... Some possibilities: the experiments are miscalibrated... there is new physics or an unaccounted effect...
Another rich example. The "supermassive black hole", like dark matter, stands ready to rescue any failed model. But if necessary, observations can be discounted, or we can call up even more spirits from the vasty deep. Heaven forfend learning some of the physics of stuff we know exists, and applying it, but that's work. Worse, it's risky because you can make embarrassing mistakes. With Dark Matter (etc.) you never have to admit you're wrong, because it always has exactly the properties you need just when you need them (and not a moment before).
Oh, I get it, you're a plasma cosmology crank.
Ah, yes, and the obligatory ad hominem remark. I wondered when he would get around to that.
Most/. moderators would not recognize mainstream cosmology and astrophysics if it hit them in the face, thus I suspect that systematic discrimination against challenges to it are [ad hominem omitted]
The evidence, if you bother to examine it, suggests otherwise. Certainly it's not necessary actually to know anything to moderate down criticism of the referenced articles. The timing suggests that when certain individuals get mod points, they go back and moderate down whatever they have seen recently that they disagree with. It's a toadying sort of behavior, hence the systematic bias.
astrophysicists... are as uncomfortable with dark energy and inflation as the next guy... However, they have been unable to come up with any other theories...
Yes, that's what I said. Not being versed in plasma physics (beyond inapplicable MHD), they are not equipped to formulate or evaluate models that include it, however well such models predict the observed phenomena. Instead, they cling to models that have been repeatedly falsified, inventing new physics post hoc to cover up the failures, and carefully ignore other evidence. In the meantime, though, the ones who referee journals and sit on grants committees make sure not to allow in anything they aren't qualified to evaluate. "If we can't dance it, we can't teach it".
physicists will listen to challenges to almost any theory (and are proven wrong on a regular basis, science advances!)... Black holes have only been accepted for a short period of time, but if you challenge conservation of energy be prepared.... scientists make mistakes, too, but not usually for long in the face of strong evidence.
What this implies is that astrophysics, as practiced, is no more science than, say, sociology. Whenever current astrophysical theories are falsified by observation, a fundamental law gets tossed instead. Lately we have "dark matter" (6x as much of it as the visible universe), "dark energy" (18x as much!), "inflation", and distant galaxies producing hundreds of times more light than similar modern ones. All are futile attempts to rescue the Big Bang from the oblivion it earns by being, finally, irreconcilable with observation. (E.g. light-element ratios; gravitational lensing measurements of galactic mass; fractal, filamentary arrangement of galactic superclusters; preferred direction of cosmic microwave background anisotropy; shall I go on?)
For all the claims of evidence for the role of neutron stars and black holes in galactic-scale events, it all amounts to negative evidence: those are the only way to concentrate enough energy when the only forces you are willing or equipped to work with are gravitation, fusion, and shock waves. Even so, multimillion-degree "hot gases" in free space and 10^14 eV cosmic rays remain beyond their capacity. Current flow in interstellar plasmas easily propagates and concentrates such energies, without reliance on untestable physical laws and ghosts. However, such work can, as a rule, only be published in Plasma Science journals not read (and perhaps not readable) by astrophysicists.
[p.s. read this quick;/. moderators prefer to prevent discussion of failures of mainstream cosmology and astrophysics.]
Your best use of magazine-reading time is not to read magazines. Instead, get on the mailing list of a Free Software project where you would actually want to use the result if it were further along. (Please, no IRC clients!) Then, start making little improvements and sending in patches. Develop a thick skin and a willingness to rewrite; your work will suck for a long time, but recoding under expert direction is more educational than coding without.
After you get a feeling for the project (or a second project, if the first doesn't click), pick some neglected corner and implement what people have wanted for for a long time. Don't announce it until you have something useful working. Don't try to get that integrated; instead, wait a couple of months until you see why it sucks, and redo it half as big, with cleaner features, and offer that. While you're rewriting you can think about what else to tackle once you're done. Rinse, repeat.
Whatever you have had integrated in a Free Software project you can refer to proudly on a resumé. That, and the collection of your postings on the mailing list, will count for way more than anything you can say in an interview at your next job.
Slashdot Mirror
That's very fine, except people have been talking for decades about clean-burning, re-usable, vertical take-off and landing, single-stage to orbit vehicle designs with much greater capacity. Also, space tugs, electric plasma drives, solar sails. Now, instead of taking a step forward, they're retreating to designs the oldsters grew up with. It's pathetic.
We don't need 125-ton space probes. None of the ones out there came close to maxing out the Shuttle's ~30 tons. We need more wee probes going lots of different places, and we need to listen to the ones that are already out there. (Voyager 1 and 2, remember? They mean to shut them down to help pay for this boondoggle.) We need a new Ulysses, and more SOHOs, and x-ray, infrared, and ultraviolet telescopes. We need ground support for them, and data analysis. (We'd have 'em already if not for those decrepit ISS tin cans.) Already planned missions, and still-working probes, will be shut down in favor of sending camping trips to the moon. Data already collected will molder on back-up disks for want of analysis.
For what? Low expectations and weak imagination. You can be sure that in a couple of years it will all become an excuse to cut the budget in half. The (popular!) space-probe stuff will already have been cut, leaving just the obvious boondoggle. Then they'll be able to hand the rest over to the Air Force, to do just the part it's interested in.
The best way to do that is to stay right here.
Since lunar gravity is about 1/10th what we are used to, that will mean that mining and digging out these tunnels will be 10x easier.
One sixth. But this evades the question. Having dug out a warren, what then? Live out your old age in comfort? For anything else you want to do (and probably that too), a hole in the moon is likely to be the worst place to do it in. If you're not willing to discuss what you want to do in your warren, it's probably not anything a reasonable person would want to pay Halliburton to have done for them.
You want building materials? Use asteroids. You want oxygen? Use asteroids. You want organics? Moon's not got 'em. You want solar power? Most of the moon hasn't got that for two weeks at a time. Want a garbage dump? Hmm, that works.
One of us isn't thinking things through, and it's not me.
I didn't find a tarball of source code, just the ISO image. When I loopback-mount the ISO image, I don't find anywhere near 80M of stuff. Is the source on the ISO image?
We can barely afford to keep a low-earth-orbit space station from burning up in the atmosphere, never mind actually doing anything useful. (The crew spends all its time on maintenance.) Now we're supposed to keep a lunar station going using super-sized Apollo designs that were abandoned decades ago because they were too wasteful. What are the crew supposed to do on the moon, anyway? Dig? What are they supposed to do on Mars? It's hard to imagine more useless lumps of dead rock.
Asteroid missions (manned or not) would be interesting. Space elevators would be very interesting. Even another Cassini (for Jupiter) would be interesting. Instead, they're gutting JPL. Anybody who says this is something other than a disaster for NASA and for space exploration is drinking Kool-aid.
Some? It would take more than isolated examples to weaken his statistical case. Have the selected cases been shown to be somehow representative of the rest?
If they really have something, they can make their case in a compelling way and people will listen.
Astronomers appear to listen until they establish that the evidence seems to contradict Big Bang cant. The process seems to involve rejecting evidence not accompanied by an acceptable theory to incorporate it into a mainstream model: without a theory, the evidence doesn't count. This is, of course, entirely counter to scrupulous scientific practice.
In Tifft's case, everyone breathed a sigh of relief when high redshifts didn't seem quantized the same way, but that didn't make the quantized low redshifts go away, however much everyone wished otherwise. The journal editors who apologized for publishing his papers are a disgrace to the whole field.
Astonishing. Evidence doesn't go away just because the person reporting it does. It's every scientist's job to keep anomalous evidence on the table, because that's where all fundamental progress comes from.
I've got hundreds of examples of quasars with stars at exactly the same redshift.
What significance do you attach to that? Arp identifies (trails of) such stars himself.
I'm sorry, I should have been clearer. We see galaxies whose redshift suggests (according to the standard interpretation) that they're 12 billion light years away, and thus formed in the first billion years after presumed recombination, made of stars that had to be 4 billion years old at the time. I.e., 12+4 > 13.7.
As an astrophysicist, I'd like to be an apologetic for the standard cosmological model ...
That's to your credit. Most astrophysicists would prefer to pretend that, e.g., quantized redshift as referenced to the CMB rest-frame in low-redshift galaxies, or to angularly-nearby low-redshift galaxies in the case of high-redshift galaxies and quasars, don't exist. Most would prefer to pretend that high-redshift quasars (e.g. z=2.11) physically in front of low-redshift opaque galaxies (e.g. NGC 7319, z=0.0225) don't exist.
Most seem to prefer to pretend that MHD conditions apply to the dynamics of interplanetary and interstellar plasmas, so they can pretend it's all just "hot gas". I have a private e-mail from a well-known astrophysicist asserting, without apparent embarrassment, "Plasmas behave as ideal gases on scales much larger than the Larmor radius of motion around the magnetic field." He admits most graduate students in astrophysics, still, never attend a laboratory class on plasma dynamics, and just work artificial problem sets using the MHD approximation.
Sorry, the "13.7 billions years" figure still counts as speculation, supermassive black holes or no supermassive black holes. We can see galaxies that are supposed to be over 12 billion years old made of stars that have to be 4 billion years old. "At least 10 billion years" is defensible without calling up spirits from the vasty deep.
According to the original paper from 2002, the star is nowhere near close enough to be "tidally disrupted", so it's just orbiting. (What it says is that even at closest approach, it's still 70x too far way.)
With all those stars whipping around, though, it wouldn't be hard to get the occasional star either entirely ejected, or potted right in. More usually, an orbit would be changed so that it approaches closely enough on each orbit to have a bit of mass (say, a trillion tons) stripped off, and gets used up over the course of a few thousand years. Of course at some point we wouldn't be able to see it any more, so there could be a bunch of those happening right now.
Probably most of the mass moving near it is non-radiating low-density plasma whose motion is controlled less by gravitation than by unimaginably intense electromagnetic fields. We see stars, but there's lots else going on in there we can't see.
As they note, there remains now the mystery of how they got so much mass to concentrate in one place. Stars don't forget all about conventional orbital dynamics just because they've spotted a black hole somewhere not too far off.
If your application is limited by database writing speed, then chances are you shouldn't be using a relational database for that part of the job. (Eight hours is a damned long time to be banging a disk head around.) The right way is to get the data onto the disk in the quickest, un-relational- transactional-est way possible, and then (presuming you are actually going to run SQL queries against it) index it. Every practical database engine provides a way to do that because everybody runs into problems that don't fit neatly in an SQL-shaped box.
It's true that I don't know anything about your requirements, datasets, or schemas, but I do know the difference between good and bad engineering.
I did say apples-to-apples. If you ask MySQL to do the same work as PG, it's slower. If you ask it to do less, it's often faster.
Mainly because the components of MySQL are worked on by different communities. Innobase is one, Berkeley DB is another. However, now that you mention it, the requirement to sign copyright over to MySQL has to reduce their core developer base. Back when I was on the PG list, though, there were only a half-dozen people doing serious work, but ... that's been five years. How time flies.
Choose based on administration complexity for a minimal setup, if you like (favoring MySQL), or on license restrictions (favoring PostgreSQL), or on features (PostgreSQL, for now), or on development community size (MySQL), or on development community competence (PostgreSQL).
Postgres is a poster child for astonishing progress in a short time from a codebase of purely academic interest to a real, industrially useful product. Over the last ten years its code got radically cleaned up, then it got journaling, increasingly automated administration, and lately replication.
Given the rate of progress in both PostgreSQL and MySQL, Oracle has reason to be worried for its future. It's worried enough, for example, to have bought out Innobase, just to threaten (or tax) the proprietary arm of MySQL, Inc. Oracle certainly has no use for Innobase, so the only plausible reason is that they are worried. They should know, and we should know better than to second-guess them.
There are still things that can be done with Oracle but not PostgreSQL, but they're getting fewer all the time. At some point they will be too few to fund Oracle's corporate expenses, and the money Oracle charges for an annual license will be better spent beefing up PostgreSQL to be able to do the job indefinitely. We may reasonably expect national laboratories to commission such work; they have stringent needs, the budget to satisfy them, and plenty of favorable experience with Free Software projects.
It's no wonder you have trouble finding "OGG" support. There's no such thing. The post you replied to got it right: it's "Ogg".
Too, you really can get Ogg/Vorbis support on older iPods (now very cheap, on eBay), thanks to the excellent crew at http://ipodlinux.org/. Yes, it will dual-boot. Booting under Linux you can transfer files directly from yours to anybody else's, whatever they are running. I don't know whether you can install linux on them that way, but if you can it would be the polite thing to do when you get connected.
Ellen warned us of that when she said it stumped Donald Knuth. It was a far more useful hint than the one she linked to. It certainly is, though, a logic problem: once you establish that there is no purely logical solution, you canknow you must bring to bear what you know about switches, light bulbs, and time, and stir it in. We all have more than sufficient experience with all three. Since nobody who solves it actually experiences the switches or rooms, it remains a logic problem.
One I enjoyed a great deal. Thank you, Ellen.
Lots of better (non-registration-required) links here:
+ News
http://news.google.com/news?q=resilin&btnG=Search
And guess who those interest payments are going to? (Hint #1: Not you. Hint #2: Nobody you know. Hint #3: I wonder how China paid for its new space program...)
Answering my own post:
/ 20041114b
http://dba.openoffice.org/drivers/sqlite/
http://oooauthors.org/en/FAQs/Database/connectors
http://www.ch-werner.de/sqliteodbc/html/
but I'd like to hear from people working on it. Will there be an OOo package with SQLite or something in there, and no Java? (E.g. on Debian and Ubuntu?)
I'd like to hear about Java-free builds. In particular, I wonder whether anyone has made progress plugging in SQLite in place of their Java-dependent database engine. Database access seems to be the only important feature in 2.0 that depends on Java.
While an OOo built with Gcj and Classpath is, apparently, legally unencumbered, the future of the language is uncertain. Some us would prefer, for a variety of reasons, to have OOo not dependent on Java for core features.
Note well, this isn't a discount for buying without MS -- they're charging $105 extra, and shortening your warranty, for leaving the disk blank.
It's one thing to charge extra. It's quite another to hike up the premium after the Register has called them on it.
Ha ha. Most sociologists remain Marxists.
All of those observations SUPPORT Big Bang cosmology, rather than contradict it.
Only if you ignore the evidence. Light-element isotope ratios were once trumpeted as supporting the Big Bang. Now that they don't match current models, they aren't discussed at polite conferences. Galactic gravitational lensing puts a strict upper limit on mass, leaving no place to put the dark matter. (Unless... yes! Dark energy at work? Or ... maybe supermassive black holes?) You could put your dark matter out in those unaccountable megaparsec-wide voids, except you need it in galaxies to explain why their rotation doesn't match gravitational orbital dynamics (but does match laboratory plasma phenomena).
This is a capsule example of the phenomenon. Big Bang can't be falsified, because all evidence can be reinterpreted, post hoc, in its favor. There have been other pseudo-scientific regimes like that: Marxist economics, Freudian and Skinnerian psychology, Lysenkoist genetics; lately, Intelligent Design has support in high places.
Theories of neutron stars and black holes make specific predictions of what you will see, and those predictions are supported by observations.
Two words: "post hoc"
Ultra high energy cosmic rays, for one, are within the capacity of jets from supermassive black holes. ... Some possibilities: the experiments are miscalibrated ... there is new physics or an unaccounted effect ...
Another rich example. The "supermassive black hole", like dark matter, stands ready to rescue any failed model. But if necessary, observations can be discounted, or we can call up even more spirits from the vasty deep. Heaven forfend learning some of the physics of stuff we know exists, and applying it, but that's work. Worse, it's risky because you can make embarrassing mistakes. With Dark Matter (etc.) you never have to admit you're wrong, because it always has exactly the properties you need just when you need them (and not a moment before).
Oh, I get it, you're a plasma cosmology crank.
Ah, yes, and the obligatory ad hominem remark. I wondered when he would get around to that.
The evidence, if you bother to examine it, suggests otherwise. Certainly it's not necessary actually to know anything to moderate down criticism of the referenced articles. The timing suggests that when certain individuals get mod points, they go back and moderate down whatever they have seen recently that they disagree with. It's a toadying sort of behavior, hence the systematic bias.
astrophysicists ... are as uncomfortable with dark energy and inflation as the next guy ... However, they have been unable to come up with any other theories ...
Yes, that's what I said. Not being versed in plasma physics (beyond inapplicable MHD), they are not equipped to formulate or evaluate models that include it, however well such models predict the observed phenomena. Instead, they cling to models that have been repeatedly falsified, inventing new physics post hoc to cover up the failures, and carefully ignore other evidence. In the meantime, though, the ones who referee journals and sit on grants committees make sure not to allow in anything they aren't qualified to evaluate. "If we can't dance it, we can't teach it".
It may be human nature, but it ain't science.
What this implies is that astrophysics, as practiced, is no more science than, say, sociology. Whenever current astrophysical theories are falsified by observation, a fundamental law gets tossed instead. Lately we have "dark matter" (6x as much of it as the visible universe), "dark energy" (18x as much!), "inflation", and distant galaxies producing hundreds of times more light than similar modern ones. All are futile attempts to rescue the Big Bang from the oblivion it earns by being, finally, irreconcilable with observation. (E.g. light-element ratios; gravitational lensing measurements of galactic mass; fractal, filamentary arrangement of galactic superclusters; preferred direction of cosmic microwave background anisotropy; shall I go on?)
For all the claims of evidence for the role of neutron stars and black holes in galactic-scale events, it all amounts to negative evidence: those are the only way to concentrate enough energy when the only forces you are willing or equipped to work with are gravitation, fusion, and shock waves. Even so, multimillion-degree "hot gases" in free space and 10^14 eV cosmic rays remain beyond their capacity. Current flow in interstellar plasmas easily propagates and concentrates such energies, without reliance on untestable physical laws and ghosts. However, such work can, as a rule, only be published in Plasma Science journals not read (and perhaps not readable) by astrophysicists.
[p.s. read this quick; /. moderators prefer to prevent discussion of failures of mainstream cosmology and astrophysics.]
After you get a feeling for the project (or a second project, if the first doesn't click), pick some neglected corner and implement what people have wanted for for a long time. Don't announce it until you have something useful working. Don't try to get that integrated; instead, wait a couple of months until you see why it sucks, and redo it half as big, with cleaner features, and offer that. While you're rewriting you can think about what else to tackle once you're done. Rinse, repeat.
Whatever you have had integrated in a Free Software project you can refer to proudly on a resumé. That, and the collection of your postings on the mailing list, will count for way more than anything you can say in an interview at your next job.