Are the Gen IV water reactors fast reactors? I see absolutely zero point in putting non-fast reactors online. And I'm very partial to the Integral Fast Reactor. Exemplary safety, far better than anything else made, no PUREX (another kind of safety right there), burns existing waste (fast reactor), exemplary energy use (99% of available energy v. the 2% of the modern once through cycle).... and we had one working for quite a number of years. Till Clinton shut it down. Because it was a 'reprocessing' reactor, what idiots.
Your right though, pebble bed is worthless. I want someone to come up with a history for how these stupid things got proposed and integrated into US policy.
Anyways, Long Now just had a nuclear forum. Good discussion, please join.
Blah blah blah Bremsstrahlung losses, yada yada. Go read the Rider's paper, Todd Rider's, "Fundamental limitations on plasma fusion systems not in thermodynamic equilibrium". Here, have a link.
Aneutronic fusion is "impossible". 3He is silly. I just see little reason to go fusion at all if its going to be heavily neutronic, significantly more neutronic than fision. Just built some molten-salt fision reactors and start burning the nuclear waste we've already got. I dont think we'll run out anytime soon.
I'm obviously not a nuclear engineer, but I got my aneutronic fusion hopes basically dashed. I still havent seen anything to give me hope.
You sir have absolutely no empathy for people who do not pretend to understand every facet of their existance. There's plenty of people who live in really small parts of the world, who really never got a chance to see much; its really not their fault they didnt have the exposure to all the 21st century whoonanny any "properly civilized person" has. Or even smaller things, like say, literacy. You'd be astounded the unliteracy rate of some places in the US. Its not just desire to learn, its how our brains have been programmed. Some brains just dont have the patterns to digest what their seeing, and its not like sitting down and explaining something slowly and with good metaphors is going to suddenly make a light bulb go off. There's enormous interia to overcome. Many of these things are just non-integrable into people's existing conception of reality, there's simply no existing rules or patterns to start from, to begin to even understand the faintest thing about some of your obvious concepts. Knowledge is a pyramid with a very very wide base my young paduan. And with no one is volunteering to explain these things, to deconstruct the weird rules their grasping brain substitutes for objectively-seemingly-plausible ones, the situation borders hopeless. It just takes experience, it takes experience to build more. But without that core, if you're constantly rebuffed, if rules never make sense, its like a mice or pet cat that gets shocked for doing some objective task "wrong" but cant cognitize what & why it keeps happening... its just random negative reinforcement that builds up until the only solution is to turn off and turn out. Stick with what you know, with what doesnt hurt. Be glad you can eat ok. Worship your false icons. Life does what it must to survive, sometimes there's just not enough light left to flourish.
Computers especially are non-intuivite systems, criticizing people for not understanding is sub-human; being able to screen scrape and guestimate your way through menu-driven systems is like diving through a perpeutally changing rabit hole. A computer at 1024x768 updates 2000 words 75 times a second, a single menubar hides ninety different buttons which each spring up entirely different dialog boxes. Sometimes a couple mechanized processes to do basic tasks really is a stellar achievement. My dad, a wonderul and insightful craftsmen, always exploring new techniques and building styles, cant do much more than mechanized computer work. He gains familiarity with a couple facets of a couple systems and works with what he knows. He picked up quick, his mental model and experience allowed him to run mechanical task-based routines through computers reasonably well reasonaly quick with a little assistance and a little improvisation, but by and large his improvisational and navigational skills are really limited. But he doesnt do well with configuration or settings at all, he doesnt understand what to look for. I dare say, he'll be able to do what he wants with computers, but I doubt he'll ever really understand or intuit how computers work. Telling him to sit down in front of a computer and "play around" would do nothing. My mom on the other hand was much slower to learn, she could never automate the process of finding what she was looking for. So she's developed a strong ability to screen scrape and experiment. She's actually learning how it works, rather than how to do it. She's building the filters to digest what is in front of her. My dad still has to read every character on screen. Not everyone has a mental model to really understand. ------
There's a reason this world is filled with fundie right wing nut jobs. Its because they cant comprehend the world, and because there's these chest thumping self proclaimed liberally educated elite that demand everyone be as cultured and omni-knowing as them. Just do your cause a favor and shut the fuck up. I'm tired of loosing elections already.
Some day there'll be a world where the raison d'êtr is some wonderful
Let them think the tiny god could become angry with them if they browse the wrong folders, or tamper with the holy configurations.
Ok, once upon a time I had a Mage: The Awakening character (whitewolf pen and paper rpg) who was a technopagan. His watercooling ran on blood, he fed the spirits of electricity regularly, and had a wall of fire to keep the malicious Windows spirits far far away.
But I'm thinking I might have to port some of these jokes into the real world now. I never saw the real world applications; but man, that shit would work!
Yes, there is a unifying factor. Considering that most communities unify based on little more than geography, I'd consider the open source community's union to be simply astounding. I also think open source is actually a mis-characterization and misses the mark on what the community really is; its the hackers. Its the people who relish in manipulating complex systems and coercing their machines into operating in just the right patterns. Its the people who understand, Chromatic, that code is art.
Thats why no one else understands open source, as JM so demonstrated; they're looking at open source as an economic movement. Some kind of money driven system. Really its a technocratic movement, a hackers movement, a community built around people who need something to start from, not people who need free bear.
In my book, hack-a-day is the same blood as BSD is the same blood as chaos computer club. Sure we ally with those working on the same project, but really I think thats just a superficial grouping, a matter of convenience. Its the eugenics, the hacker gene that binds us. How many communities have that?
Re:Hear me True Gods of Interoperability
on
What is Perl 6?
·
· Score: 1
Thanks Chromatic, you always have a good reply for me.
This does make me feel a lot better. Evidently the SDL people are all over it, or at least a few active vocal users. Oh wait, in the words of a dear friend, "they is you!" *grin* I'll take your word for it.
I have a slightly harder time endorsing PDL now, but eh, waht the hey, it works.;-]
Hear me True Gods of Interoperability
on
What is Perl 6?
·
· Score: 1
Interoperbility is a great idea. I'm not sure if I buy unification though. Parrot might make a wonderful VM, but I simply cannot imagine it'll be able to stand alone. dotnet's virtual machine is fine, but it was clearly limited in scope and for all they the alternative dotnet languages just cannot overcome many of the static (v. dynamic) language limits built into the VM. Everything I read about parrot seems to indicate that it'll be a hundred times more dynamic, but I just cant picture anyone making a VM, what was the articles stupid word, oh yes, manipulexitious enough to be everything to everyone. Either the VM is abstract and the languages running on top of it have to start defining additional limitations to build functionality and we loose interopability, or the VM is overly rigid and people have to build something else.
Look at the modern langauges. Python is wonderful because its a wonderful scripting engine for C code, complete with eight hundred types of glue. Classpath is interoperable java. Dotnet can PInvoke normal libraries. Languages arent converging, they're interacting. Best of luck, please, oh please do your best to try, but I just cant imagine any being, god or man, making a VM so flawlessly unconstrained and yet utterly unifying that it can become everyone to everything.
More scripting languages built around an interoperable core would be delightful. But already there are far far too many libraries which have to be rewritten for each platform. How many SAX parsers need to be written, how many SOAP libraries, how many SAML parsers? Every language needs it own, and thats a crying shame.
What now? Do we go web services crazy and a SAML parser in Java as a web service for a dotnet application? Doesnt that seem a bit batty? Or do we break open IKVM and wholesave import the couple dozen of java libraries opensaml requires into dotnet? Every single protocol and xml-namespace faces this same barrier; what language do we build the interpreter in? how many times will we rebuild it?
My plea? Make Parrot truly interoperable. Not just by trying to bring the joyous miracle of parrot to every existing langauge concievable, but by interopting with the existing language ecosystems.
Normally you wait until a language is actually released before learning it. Traditionally, you let a couple early adopters build something with it first too. Most smart organizations wait to make sure the langauge actually is somewhat stable before buying into the list of benefits. They wait for books to be released.
Perl6 is not really here yet. Read the last page. Author doesnt come out and state it directly, but the current best implementation runs on Haskell.
I dunno, somehow I dont think the take-away was supposed to be "learn this or get fired, parrot is the one vm to rulezor them all!!11," I think it was more "perl6 is still coming and has some really cool new features, as well as being built around a much more solid core." There's mainstream technologies worth reading up on if you feel the heat to stay up to date for your job. Then there's things like Perl6/Parrot; cool technologies to read up on if you're actually fucking interested in computer languages or vm's. Forgive me, I realize you simply werent aware of the status of perl6, but perhaps you should see how many band members are actually on the bandwagon before hoping aboard yourself.
The article talks about some of the defining features for perl. Well, one of the defining features in my perl experience has been Perl Data Language, pdl. PDL _is_ whipitupitude. Its a wonderful wonderful matrix library. And it comes with the best perl shell I know.
I had to break down a equation into a sequence of linear equations. So I hacked up some PDL in like 2 hours to do that. Couldn't have been easier, even though I'd never used PDL or its perldl perl shell; I just started typing in the interactive shell until it worked as expected and until I knew what I was doing. Then I needed the results in interger, so I rounded everything down, built a permuter and sorted the permuted results for each individual segment. That took three hours, but only because I kept botching the matrix multiplication. Even with huge datasets, generating hundreds of thousands of linear equations, each spanning dozens of datapoints, permuting the linear equations, sorting them and selecting the optimal, PDL would run it all my slow arse 800mhz crusoe laptop in seconds. Matlab couldnt touch it.
Thats the other really truly thing about PDL; the performance. If someone else would chime in and do it better justice, but my crude understanding is that it generates some kind of extremely optimized machine code on first use and runs whatever equations you've thrown at it like silk from that point on.
Little late and a little off topic, but PDL really is just a masterpiece of perl hackability. The PDL perl shell is truly spectacular; get some symbolic integrators and differential equation solving packages in there and I wouldn't need to break open Mathematica or Matlab ever again. Ok, long way away, pdl is really just about matricies, but it is really really sweet, and its shell is good for anyone who just wants to try something out really quickly in fully interactive perl.
That being said, I really cant wait to see where the perl6 VM is going.
i dunno, the notion of $25 shareware software always seemed like a quaint little windows phenomena to me. application or applet for every task. i guess thats not the point though, the point is the $80 industry standard mail clients. but then again, mail clients used to be something a couple people needed. when products start reaching commodity penetration, prices have to go towards commodity pricing. its really hard to justify $60 million buisnesses that do nothing but write mail clients. at the same, this shit is a whole lot of work.
we're at an intrem. service computing is going to come along and start making more people more money. but service computing is only possible because there's free infrastructure to build it on. eventually it'll swing back again when we have better information management tools and people realize they can build their own salesforce.com really easily, when people get tired of being chained to closed services just like users were chained to closed applications. push back and forth. thats the economics of whats happening to software. but its entirely besides the point from the open source movement, from the hacker movement.
for my money, i'd put the dollar down on open source community as being defined by the people who understand that the software ecosystem only exists because people are sharing, only exists when projects are hackable. open source is really very self reinforcing as a community; its all the people who want to be involved with technology, not merely users of it. i really think this is the key, that while commercial world is in spin cycle figuring out what to do with extreme-commoditization (answer: services), open source is still doing the same thing its always done, namely build its own hacker friendly alternate reality. money was never really part of this alternate reality, but sometimes the two can be pleasantly (or unpleasantly) connected.
in the end, i'd say open source is vital because it keeps the game moving. eudora had no reason to do anything different. ms was happily king of the pile. entrenched buisnesses have no want or need to innovate, the de facto standard by its nature isnt supposed to be moving anywhere. every application is some standalone item which will never meet with any other application, will never collaborate, will never remix. they're just growing towards deprecation, awaiting the day they'll be subsumed by some more general purpose application. the open source community, or more percisely really, the hacker community, is just a byproduct of people who want to keep the game moving. having access to a library of 50,000 programs you can install with 'apt-get', use, and-- most importantly-- remix, is absolutely vital to keeping the game moving. open source isnt about software now, its about the software that is not made yet, about building the blocks to build that future software. thats what unix was built on, small pieces loosely coupled, assembled into something greater. the commercial world does not fit that model, its inherently contrary to everything they strive for. its not about applications, its about pieces, about the aggregate. but industry isnt about creating frameworks of cooperation, not yet, its about dominance and WS-* specs so long tedious and boring they'll surely keep everyone else out. we need stuff we can hack. we cannot trust the goliaths to provide us with that. thats open source; hack. anyone in open source, linux, apache, bsd, whoevers colors they wear, i just cant stress this enough, if you cant see it you wont get it, its really just about the freedom to hack.
sry, tired from a long weekend. hope i made a couple ok points and didnt get too longwinded. thanks for the reply, i actually really would like to hear more. s'uming you're actually the author.
I frikkin quoted the line of contest; "With prices approaching zero, software developers have two choices when trying to win over users: (1) add features not available elsewhere, and (2) release the source code."
'make free or make valuable'? Tell me thats not an insult. He goes on about how ultimately the choice to go free brings you into the conversation, which makes your product better, &c &c, but he's basically saying-- especailly in my choice quote but more so the article at large-- open source is free trash people couldn't sell and exists because no one COULD make money on it.
"it becomes clear that Linux or its equivalent was bound to happen eventually, regardless of whether Linus decided to release a kernel in 1991. The same applies for Apache and any other project. Both of these are the natural result of massive price drops in their respective markets." A) Linux CAUSED that price drop, was because there WAS no cheap unix. Open source IS that price drop. Sure, its cyclical, sure its causing more people to have to resort to free software and people are starting to realize that the conversations free software allow you to have are extremely important, but OSS is the reason thats happening, its whats forcing the commercial world to change its game plan.
Now on to the actual problems with your anti-community fud. " The view that there is a core group of altruistic companies and true believers driving open source forward is simply false. The view that open source participants are idealistic Davids fighting against software Goliaths is also false." b) So there is no core philosophical underpinning. But all ecosystems have dynamics, philosophy is not the only thing that can pull together a community. b) the dynamic of open source, the thing that makes us a community, is that we like clean hackable solutions, we like using the best tools we can, and other people's free code is often the best tool available. Its a technical meritocracy, may the most hackable most hack-ensuing solutions win. More than anything else, we're united by the desire to play around with cool technologies
You want a philosophical underpinning? the purpose is not to overthrow Goliath. we dont care about the economics; we just want to be able to hack great stuff.
Building what needs to be built is a community. Oftc and Freenode are communities. Really, what is activism besides building what needs to be built?
=== There's two ways to do open source. The apache way and the linux way. Apache uses committees and democratic processes to, as the article seems to want to stress, place community over individual developers. Linux, the epitome of monolithic, is built under a few chief architects who direct the project. Either one is valid; either one can foster community. Hackers will play with anyone who can play ball, anyone who lets them keep hacking, it doesnt get any simpler than that.
When one developer felt FreeBSD got overly committeed he went off and made DragonflyBSD (Whoo M. Dillon!). There's countless projects which have done the opposite; been lead by a leader until they got consumed into an Apache project. Ebb and flow; hackers follow whatever seems to be working, whether its individual run project or some democratic debian. Either one is capable of supporting community.
<b>Reducing open source to little more than "not vendor lock in" is fucking perposterous.</b> "With prices approaching zero, software developers have two choices when trying to win over users: (1) add features not available elsewhere, and (2) release the source code." Shame O'reilly, shame! To say that Linux does not innovate! Look at yesterday's ask slashdot about how to stream sounds from one system to another. Linux has countless solutions; esound, jack.udp, gstreamer, vlc. A hundred ways to dice it. Windows requires a $49 program to make a fake sound card. There used to be an open source program to do it, but the drive dev kit is now a couple hundred from MS, so the project died. Real feature added, eh?
Open source is built around the fundamental tenants of technocracy. The most elegant hackable solutions win. <b>Source code is simply our current modus operandi for ensuring our systems are maximally hackable.</b>
If the problem is no sound card, external USB sound cards are $20. Thats what we call "no excuses". I got mine at a geeks.com sale for $12. 24/96 and 8 channel output. What more do you want? Wait, your computer doesnt have USB? Try upgrading from a 486. Cant do cabling? Now we have an issue.
Look, there's no way in fucking hell you're going to get windows doing this. We all want our Pinto's to fly and do 180 on the highway, but its just not going to happen. There used to be some esound-based virtual sound card, it looked like a normal sound card and it streamed out to esound daemons across network, but it wasnt updated since the Win NT 4 days. The site was gone when I went looking for it again last year and the year before. Dead.
Since that time, MS has started charging for the device driver development kit. it used to be free. people used to be able to make cool hacks. those days are over now. there's a couple costly apps (Virtual Audio Cables is basically taht app) which create fake sound card devices, but harvesting these to make streaming connections to other machines is not simple. Shoutcast is your only bet, and good luck getting anything less than a 1s delay on that front.
This one question IS very interesting, it just so happens that the particulars of the question are quite boring bordering idiotic. We might not get flying pintos, but by-george slashdot will do its damnedest to whip up a flying car. "Where are my flying cars?" The "news for nerds" under the title is the hint that we're going to take your boring "give me software to do it" Suit-wearing question and mix it up to a "what CAN we do", "what IS possible" question. The answer's left wearing leftover's 90's grunge, but you know what, at least its a real answer. Which is something Windows cannot deliver. We've crushed the technical problem best as possible; mission accomplished.
I would've written your frakking software for you too, had windows continued giving away the device developers kit free.
Allow me to start off by being a bitch; "What sort of latency is there with this approach?" What sort of machine are you running? What sort of network cards & network speed? What sort of network congestion? What sort of tasks are you running on your systems?
Jack.udp itself is exceedingly low latency. I would be truly suprised if you found a lower latency streaming method. Take care to setup jack to use smaller buffers to reduce its latency.
On the other hand, I'm not sure how well jack.udp maintains synchronization. Synch is a very complex issue. Xine & all media players take great pains to keep the system clock in sync with the sound card's clock. How this is supposed to happen with a network thrown in between, I have no idea. I'd suggest running a very draconian NTP, but it probably will not help.
There's two main options. VLC will stream anything, simple as day, its fan-freaking-tastic. Really, VLC streaming is just superb.
On the other hand, there's icecasting/shoutcasting. you should be able to get a JACK stream from the FM tuner, in which case you can pickup Oddsock's Oddcast to stream it to an Oddcast/Icecast server; the server will require editing a couple lines of an xml config to get running properly but its not hard. There's plenty of other options for streaming to Icecast, I just happen to like Oddcast/JACK since its so low latency and can be configured while live. JACK is just a low latency patch-board, it can be useful. OTOH, it does take some actual CPU usage to run; liveice could potentially give you better results if you have a static setup and know you wont want to do anything else but listen to the FM stream.
Holographic will be the first new technology to present a worthy enough upgrade to prompt mass migration. Even 50GB is simply not enough, although it does at least become feasible for me to start doing backups again.
Reiser's stated goal is to increase throughput through additional CPU usage. If you're not using the CPU anything, since, say, you're blocking for I/O to go through, it makes sense to expense some CPU if you can get I/O done faster.
On the other hand, if you're running a database, yes, you need all the cpu you have. The question then becomes, how much I/O do you get per unit of CPU, but the situation is very complex; its not necessarily some easily reducable linear system. It could be that you get, for example, twice the I/O at ten times the CPU usage, but this doesnt imply you'll only get 1/5 the CPU at one times the cpu usage. Again, these systems dont have to be linear... scalability has many dimensions to it.
But yes, clearly the goal of ReiserFS is to serve a desktop platform where desktop apps are often going to be awaiting data to proceed.
Celeron 300A was Spring/Summer of `98. I cant imagine it took Intel a full 18 months to go from 450 mhz on the cheap end to 500 mhz on the fast end.
Those were the days. I still have a couple old Alpha heatsinks with what must've been an AMAZING 50 mm fan, weighing almost a whole third of a pound of solid aluminum. HA, fantastic. Two of the BP6's I built are still alive and kicking today. Dual powered celerons, whoo.
Compare that to my <a href="http://www.thermalright.com/a_page/main_prod uct_xp120.htm">XP-120</a>... hilarious. What a differnt age.
Checkout the CPU utilizations; reiserfs is pegged at 100% cpu utilization for ~8 tests. For a FS which describes itself as willing to use more CPU in order to achieve better I/O than the competition, running the benches on an antiquated 700 mhz machine is simply not fair.
OTOH, Untarring and tarring are notably NOT cpu limited, and still pretty lackluster for Reisers case. Disappointing, very disappointing. I was extremely impressed in the ext's; I simply had no idea how consistently well performing they were.
I'd also like to see FreeBSD's UFS/w and w/o softupdate benched.
I'll have to look for the demos. I wasnt aware of heirarchy. I thought it was just a size scaling replacement for toolbars. Helping people find stuff is defiantely important.
Anyways, we all know screen editing is the one true path, right?;-]
reiserfs is no fun without the meta... does that count as a double entranda?
i guess i didnt make my original post clear.
the kernel dev's still think of linux as a monolithic kernel. join the kernel if you want a working project; we will garuntee you no stable external facing API's... thats what they said to video drivers, and i'm amazed they made any consolation at all for fuse. but what happens when something like reiserfs comes along? it has to be shoehorned in and trimmed down to fit with the existing picture kernel-dev's have for the kernel. thats what i mean when i talk about stifling innovation. its not only a monolithic kernel, its a monolithic development process. its an entirely different form of monopoly, but it is one none the less, and its because linux is inherently monolithic. i'd love to see examples of how this has changed, but besides fuse, i'm not aware of many.
winfs is one of those big infinity projects, but it could be said that reiserfs is a ripoff off it (at least of some of the earlier different incarnations). conversely, singularity may be a rip off of HURD, another great infinity project, but it looks like some of singularity is actually going to make it to market. you're absolutely right, its a shame microsoft doesnt give credit where credit is due, but if your interested in isolation, interested in security, you know about EROS, HURD, & microkernel, and you've probably read some of the papers of people who did actual research.
i really dont see microsoft as using a more microkernel alike approach as means of suppressing others, i see it as the logical choice now that we have the performance to use microkernel systems. i'm not a shill for ms, i'm a shill for microkernel. i dont think i ever stated anything about ms doing goodwill, i just happen to be a technical purist; whether its money-sniffing ms or oss-touting linux following smart tracks, i think its important to note someone is doing something intelligent. i'm psyched to see someone actually delivering the one true microkernel path, even if i'm never going to frakking touch it.
Slashdot Mirror
Are the Gen IV water reactors fast reactors? I see absolutely zero point in putting non-fast reactors online. And I'm very partial to the Integral Fast Reactor. Exemplary safety, far better than anything else made, no PUREX (another kind of safety right there), burns existing waste (fast reactor), exemplary energy use (99% of available energy v. the 2% of the modern once through cycle).... and we had one working for quite a number of years. Till Clinton shut it down. Because it was a 'reprocessing' reactor, what idiots.
Your right though, pebble bed is worthless. I want someone to come up with a history for how these stupid things got proposed and integrated into US policy.
Anyways, Long Now just had a nuclear forum. Good discussion, please join.
Blah blah blah Bremsstrahlung losses, yada yada. Go read the Rider's paper, Todd Rider's, "Fundamental limitations on plasma fusion systems not in thermodynamic equilibrium". Here, have a link.
Aneutronic fusion is "impossible". 3He is silly. I just see little reason to go fusion at all if its going to be heavily neutronic, significantly more neutronic than fision. Just built some molten-salt fision reactors and start burning the nuclear waste we've already got. I dont think we'll run out anytime soon.
I'm obviously not a nuclear engineer, but I got my aneutronic fusion hopes basically dashed. I still havent seen anything to give me hope.
You sir have absolutely no empathy for people who do not pretend to understand every facet of their existance. There's plenty of people who live in really small parts of the world, who really never got a chance to see much; its really not their fault they didnt have the exposure to all the 21st century whoonanny any "properly civilized person" has. Or even smaller things, like say, literacy. You'd be astounded the unliteracy rate of some places in the US. Its not just desire to learn, its how our brains have been programmed. Some brains just dont have the patterns to digest what their seeing, and its not like sitting down and explaining something slowly and with good metaphors is going to suddenly make a light bulb go off. There's enormous interia to overcome. Many of these things are just non-integrable into people's existing conception of reality, there's simply no existing rules or patterns to start from, to begin to even understand the faintest thing about some of your obvious concepts. Knowledge is a pyramid with a very very wide base my young paduan. And with no one is volunteering to explain these things, to deconstruct the weird rules their grasping brain substitutes for objectively-seemingly-plausible ones, the situation borders hopeless. It just takes experience, it takes experience to build more. But without that core, if you're constantly rebuffed, if rules never make sense, its like a mice or pet cat that gets shocked for doing some objective task "wrong" but cant cognitize what & why it keeps happening... its just random negative reinforcement that builds up until the only solution is to turn off and turn out. Stick with what you know, with what doesnt hurt. Be glad you can eat ok. Worship your false icons. Life does what it must to survive, sometimes there's just not enough light left to flourish.
Computers especially are non-intuivite systems, criticizing people for not understanding is sub-human; being able to screen scrape and guestimate your way through menu-driven systems is like diving through a perpeutally changing rabit hole. A computer at 1024x768 updates 2000 words 75 times a second, a single menubar hides ninety different buttons which each spring up entirely different dialog boxes. Sometimes a couple mechanized processes to do basic tasks really is a stellar achievement. My dad, a wonderul and insightful craftsmen, always exploring new techniques and building styles, cant do much more than mechanized computer work. He gains familiarity with a couple facets of a couple systems and works with what he knows. He picked up quick, his mental model and experience allowed him to run mechanical task-based routines through computers reasonably well reasonaly quick with a little assistance and a little improvisation, but by and large his improvisational and navigational skills are really limited. But he doesnt do well with configuration or settings at all, he doesnt understand what to look for. I dare say, he'll be able to do what he wants with computers, but I doubt he'll ever really understand or intuit how computers work. Telling him to sit down in front of a computer and "play around" would do nothing. My mom on the other hand was much slower to learn, she could never automate the process of finding what she was looking for. So she's developed a strong ability to screen scrape and experiment. She's actually learning how it works, rather than how to do it. She's building the filters to digest what is in front of her. My dad still has to read every character on screen. Not everyone has a mental model to really understand.
------
There's a reason this world is filled with fundie right wing nut jobs. Its because they cant comprehend the world, and because there's these chest thumping self proclaimed liberally educated elite that demand everyone be as cultured and omni-knowing as them. Just do your cause a favor and shut the fuck up. I'm tired of loosing elections already.
Some day there'll be a world where the raison d'êtr is some wonderful
Let them think the tiny god could become angry with them if they browse the wrong folders, or tamper with the holy configurations.
Ok, once upon a time I had a Mage: The Awakening character (whitewolf pen and paper rpg) who was a technopagan. His watercooling ran on blood, he fed the spirits of electricity regularly, and had a wall of fire to keep the malicious Windows spirits far far away.
But I'm thinking I might have to port some of these jokes into the real world now. I never saw the real world applications; but man, that shit would work!
sincerely,
your newly paganed Technomagus,
Myren
Yes, there is a unifying factor. Considering that most communities unify based on little more than geography, I'd consider the open source community's union to be simply astounding. I also think open source is actually a mis-characterization and misses the mark on what the community really is; its the hackers. Its the people who relish in manipulating complex systems and coercing their machines into operating in just the right patterns. Its the people who understand, Chromatic, that code is art.
Thats why no one else understands open source, as JM so demonstrated; they're looking at open source as an economic movement. Some kind of money driven system. Really its a technocratic movement, a hackers movement, a community built around people who need something to start from, not people who need free bear.
In my book, hack-a-day is the same blood as BSD is the same blood as chaos computer club. Sure we ally with those working on the same project, but really I think thats just a superficial grouping, a matter of convenience. Its the eugenics, the hacker gene that binds us. How many communities have that?
Thanks Chromatic, you always have a good reply for me.
This does make me feel a lot better. Evidently the SDL people are all over it, or at least a few active vocal users. Oh wait, in the words of a dear friend, "they is you!" *grin* I'll take your word for it.
Yeah, I never really deemed perl XS "worthy".
And here I thought parent was being sarcastic.
;-]
I have a slightly harder time endorsing PDL now, but eh, waht the hey, it works.
Interoperbility is a great idea. I'm not sure if I buy unification though. Parrot might make a wonderful VM, but I simply cannot imagine it'll be able to stand alone. dotnet's virtual machine is fine, but it was clearly limited in scope and for all they the alternative dotnet languages just cannot overcome many of the static (v. dynamic) language limits built into the VM. Everything I read about parrot seems to indicate that it'll be a hundred times more dynamic, but I just cant picture anyone making a VM, what was the articles stupid word, oh yes, manipulexitious enough to be everything to everyone. Either the VM is abstract and the languages running on top of it have to start defining additional limitations to build functionality and we loose interopability, or the VM is overly rigid and people have to build something else.
Look at the modern langauges. Python is wonderful because its a wonderful scripting engine for C code, complete with eight hundred types of glue. Classpath is interoperable java. Dotnet can PInvoke normal libraries. Languages arent converging, they're interacting. Best of luck, please, oh please do your best to try, but I just cant imagine any being, god or man, making a VM so flawlessly unconstrained and yet utterly unifying that it can become everyone to everything.
More scripting languages built around an interoperable core would be delightful. But already there are far far too many libraries which have to be rewritten for each platform. How many SAX parsers need to be written, how many SOAP libraries, how many SAML parsers? Every language needs it own, and thats a crying shame.
What now? Do we go web services crazy and a SAML parser in Java as a web service for a dotnet application? Doesnt that seem a bit batty? Or do we break open IKVM and wholesave import the couple dozen of java libraries opensaml requires into dotnet? Every single protocol and xml-namespace faces this same barrier; what language do we build the interpreter in? how many times will we rebuild it?
My plea? Make Parrot truly interoperable. Not just by trying to bring the joyous miracle of parrot to every existing langauge concievable, but by interopting with the existing language ecosystems.
Myren
Normally you wait until a language is actually released before learning it. Traditionally, you let a couple early adopters build something with it first too. Most smart organizations wait to make sure the langauge actually is somewhat stable before buying into the list of benefits. They wait for books to be released.
Perl6 is not really here yet. Read the last page. Author doesnt come out and state it directly, but the current best implementation runs on Haskell.
I dunno, somehow I dont think the take-away was supposed to be "learn this or get fired, parrot is the one vm to rulezor them all!!11," I think it was more "perl6 is still coming and has some really cool new features, as well as being built around a much more solid core." There's mainstream technologies worth reading up on if you feel the heat to stay up to date for your job. Then there's things like Perl6/Parrot; cool technologies to read up on if you're actually fucking interested in computer languages or vm's. Forgive me, I realize you simply werent aware of the status of perl6, but perhaps you should see how many band members are actually on the bandwagon before hoping aboard yourself.
Myren
The article talks about some of the defining features for perl. Well, one of the defining features in my perl experience has been Perl Data Language, pdl. PDL _is_ whipitupitude. Its a wonderful wonderful matrix library. And it comes with the best perl shell I know.
I had to break down a equation into a sequence of linear equations. So I hacked up some PDL in like 2 hours to do that. Couldn't have been easier, even though I'd never used PDL or its perldl perl shell; I just started typing in the interactive shell until it worked as expected and until I knew what I was doing. Then I needed the results in interger, so I rounded everything down, built a permuter and sorted the permuted results for each individual segment. That took three hours, but only because I kept botching the matrix multiplication. Even with huge datasets, generating hundreds of thousands of linear equations, each spanning dozens of datapoints, permuting the linear equations, sorting them and selecting the optimal, PDL would run it all my slow arse 800mhz crusoe laptop in seconds. Matlab couldnt touch it.
Thats the other really truly thing about PDL; the performance. If someone else would chime in and do it better justice, but my crude understanding is that it generates some kind of extremely optimized machine code on first use and runs whatever equations you've thrown at it like silk from that point on.
Little late and a little off topic, but PDL really is just a masterpiece of perl hackability. The PDL perl shell is truly spectacular; get some symbolic integrators and differential equation solving packages in there and I wouldn't need to break open Mathematica or Matlab ever again. Ok, long way away, pdl is really just about matricies, but it is really really sweet, and its shell is good for anyone who just wants to try something out really quickly in fully interactive perl.
That being said, I really cant wait to see where the perl6 VM is going.
G'night!
Myren
i dunno, the notion of $25 shareware software always seemed like a quaint little windows phenomena to me. application or applet for every task. i guess thats not the point though, the point is the $80 industry standard mail clients. but then again, mail clients used to be something a couple people needed. when products start reaching commodity penetration, prices have to go towards commodity pricing. its really hard to justify $60 million buisnesses that do nothing but write mail clients. at the same, this shit is a whole lot of work.
we're at an intrem. service computing is going to come along and start making more people more money. but service computing is only possible because there's free infrastructure to build it on. eventually it'll swing back again when we have better information management tools and people realize they can build their own salesforce.com really easily, when people get tired of being chained to closed services just like users were chained to closed applications. push back and forth. thats the economics of whats happening to software. but its entirely besides the point from the open source movement, from the hacker movement.
for my money, i'd put the dollar down on open source community as being defined by the people who understand that the software ecosystem only exists because people are sharing, only exists when projects are hackable. open source is really very self reinforcing as a community; its all the people who want to be involved with technology, not merely users of it. i really think this is the key, that while commercial world is in spin cycle figuring out what to do with extreme-commoditization (answer: services), open source is still doing the same thing its always done, namely build its own hacker friendly alternate reality. money was never really part of this alternate reality, but sometimes the two can be pleasantly (or unpleasantly) connected.
in the end, i'd say open source is vital because it keeps the game moving. eudora had no reason to do anything different. ms was happily king of the pile. entrenched buisnesses have no want or need to innovate, the de facto standard by its nature isnt supposed to be moving anywhere. every application is some standalone item which will never meet with any other application, will never collaborate, will never remix. they're just growing towards deprecation, awaiting the day they'll be subsumed by some more general purpose application. the open source community, or more percisely really, the hacker community, is just a byproduct of people who want to keep the game moving. having access to a library of 50,000 programs you can install with 'apt-get', use, and-- most importantly-- remix, is absolutely vital to keeping the game moving. open source isnt about software now, its about the software that is not made yet, about building the blocks to build that future software. thats what unix was built on, small pieces loosely coupled, assembled into something greater. the commercial world does not fit that model, its inherently contrary to everything they strive for. its not about applications, its about pieces, about the aggregate. but industry isnt about creating frameworks of cooperation, not yet, its about dominance and WS-* specs so long tedious and boring they'll surely keep everyone else out. we need stuff we can hack. we cannot trust the goliaths to provide us with that. thats open source; hack. anyone in open source, linux, apache, bsd, whoevers colors they wear, i just cant stress this enough, if you cant see it you wont get it, its really just about the freedom to hack.
sry, tired from a long weekend. hope i made a couple ok points and didnt get too longwinded. thanks for the reply, i actually really would like to hear more. s'uming you're actually the author.
I frikkin quoted the line of contest;
"With prices approaching zero, software developers have two choices when trying to win over users: (1) add features not available elsewhere, and (2) release the source code."
'make free or make valuable'? Tell me thats not an insult. He goes on about how ultimately the choice to go free brings you into the conversation, which makes your product better, &c &c, but he's basically saying-- especailly in my choice quote but more so the article at large-- open source is free trash people couldn't sell and exists because no one COULD make money on it.
How many communities have clear divine purpose?
#1, above all; CAUSE & EFFECT
"it becomes clear that Linux or its equivalent was bound to happen eventually, regardless of whether Linus decided to release a kernel in 1991. The same applies for Apache and any other project. Both of these are the natural result of massive price drops in their respective markets."
A) Linux CAUSED that price drop, was because there WAS no cheap unix. Open source IS that price drop. Sure, its cyclical, sure its causing more people to have to resort to free software and people are starting to realize that the conversations free software allow you to have are extremely important, but OSS is the reason thats happening, its whats forcing the commercial world to change its game plan.
Now on to the actual problems with your anti-community fud.
" The view that there is a core group of altruistic companies and true believers driving open source forward is simply false. The view that open source participants are idealistic Davids fighting against software Goliaths is also false."
b) So there is no core philosophical underpinning. But all ecosystems have dynamics, philosophy is not the only thing that can pull together a community.
b) the dynamic of open source, the thing that makes us a community, is that we like clean hackable solutions, we like using the best tools we can, and other people's free code is often the best tool available. Its a technical meritocracy, may the most hackable most hack-ensuing solutions win. More than anything else, we're united by the desire to play around with cool technologies
You want a philosophical underpinning? the purpose is not to overthrow Goliath. we dont care about the economics; we just want to be able to hack great stuff.
Luv,
Myren
Building what needs to be built is a community. Oftc and Freenode are communities.
Really, what is activism besides building what needs to be built?
===
There's two ways to do open source. The apache way and the linux way. Apache uses committees and democratic processes to, as the article seems to want to stress, place community over individual developers. Linux, the epitome of monolithic, is built under a few chief architects who direct the project. Either one is valid; either one can foster community. Hackers will play with anyone who can play ball, anyone who lets them keep hacking, it doesnt get any simpler than that.
When one developer felt FreeBSD got overly committeed he went off and made DragonflyBSD (Whoo M. Dillon!). There's countless projects which have done the opposite; been lead by a leader until they got consumed into an Apache project. Ebb and flow; hackers follow whatever seems to be working, whether its individual run project or some democratic debian. Either one is capable of supporting community.
<b>Reducing open source to little more than "not vendor lock in" is fucking perposterous.</b>
"With prices approaching zero, software developers have two choices when trying to win over users: (1) add features not available elsewhere, and (2) release the source code."
Shame O'reilly, shame! To say that Linux does not innovate! Look at yesterday's ask slashdot about how to stream sounds from one system to another. Linux has countless solutions; esound, jack.udp, gstreamer, vlc. A hundred ways to dice it. Windows requires a $49 program to make a fake sound card. There used to be an open source program to do it, but the drive dev kit is now a couple hundred from MS, so the project died. Real feature added, eh?
Open source is built around the fundamental tenants of technocracy. The most elegant hackable solutions win. <b>Source code is simply our current modus operandi for ensuring our systems are maximally hackable.</b>
Myren
If the problem is no sound card, external USB sound cards are $20. Thats what we call "no excuses". I got mine at a geeks.com sale for $12. 24/96 and 8 channel output. What more do you want? Wait, your computer doesnt have USB? Try upgrading from a 486. Cant do cabling? Now we have an issue.
Look, there's no way in fucking hell you're going to get windows doing this. We all want our Pinto's to fly and do 180 on the highway, but its just not going to happen. There used to be some esound-based virtual sound card, it looked like a normal sound card and it streamed out to esound daemons across network, but it wasnt updated since the Win NT 4 days. The site was gone when I went looking for it again last year and the year before. Dead.
Since that time, MS has started charging for the device driver development kit. it used to be free. people used to be able to make cool hacks. those days are over now. there's a couple costly apps (Virtual Audio Cables is basically taht app) which create fake sound card devices, but harvesting these to make streaming connections to other machines is not simple. Shoutcast is your only bet, and good luck getting anything less than a 1s delay on that front.
This one question IS very interesting, it just so happens that the particulars of the question are quite boring bordering idiotic. We might not get flying pintos, but by-george slashdot will do its damnedest to whip up a flying car. "Where are my flying cars?" The "news for nerds" under the title is the hint that we're going to take your boring "give me software to do it" Suit-wearing question and mix it up to a "what CAN we do", "what IS possible" question. The answer's left wearing leftover's 90's grunge, but you know what, at least its a real answer. Which is something Windows cannot deliver. We've crushed the technical problem best as possible; mission accomplished.
I would've written your frakking software for you too, had windows continued giving away the device developers kit free.
Myren
Allow me to start off by being a bitch;
"What sort of latency is there with this approach?"
What sort of machine are you running? What sort of network cards & network speed? What sort of network congestion? What sort of tasks are you running on your systems?
Jack.udp itself is exceedingly low latency. I would be truly suprised if you found a lower latency streaming method. Take care to setup jack to use smaller buffers to reduce its latency.
On the other hand, I'm not sure how well jack.udp maintains synchronization. Synch is a very complex issue. Xine & all media players take great pains to keep the system clock in sync with the sound card's clock. How this is supposed to happen with a network thrown in between, I have no idea. I'd suggest running a very draconian NTP, but it probably will not help.
There's two main options. VLC will stream anything, simple as day, its fan-freaking-tastic. Really, VLC streaming is just superb.
On the other hand, there's icecasting/shoutcasting. you should be able to get a JACK stream from the FM tuner, in which case you can pickup Oddsock's Oddcast to stream it to an Oddcast/Icecast server; the server will require editing a couple lines of an xml config to get running properly but its not hard. There's plenty of other options for streaming to Icecast, I just happen to like Oddcast/JACK since its so low latency and can be configured while live. JACK is just a low latency patch-board, it can be useful. OTOH, it does take some actual CPU usage to run; liveice could potentially give you better results if you have a static setup and know you wont want to do anything else but listen to the FM stream.
Good luck, happy hacking.
Myren
Holographic will be the first new technology to present a worthy enough upgrade to prompt mass migration. Even 50GB is simply not enough, although it does at least become feasible for me to start doing backups again.
Reiser's stated goal is to increase throughput through additional CPU usage. If you're not using the CPU anything, since, say, you're blocking for I/O to go through, it makes sense to expense some CPU if you can get I/O done faster.
On the other hand, if you're running a database, yes, you need all the cpu you have. The question then becomes, how much I/O do you get per unit of CPU, but the situation is very complex; its not necessarily some easily reducable linear system. It could be that you get, for example, twice the I/O at ten times the CPU usage, but this doesnt imply you'll only get 1/5 the CPU at one times the cpu usage. Again, these systems dont have to be linear... scalability has many dimensions to it.
But yes, clearly the goal of ReiserFS is to serve a desktop platform where desktop apps are often going to be awaiting data to proceed.
Forgive me sir, but I seriously do not think so.
d uct_xp120.htm">XP-120</a>... hilarious. What a differnt age.
Celeron 300A was Spring/Summer of `98. I cant imagine it took Intel a full 18 months to go from 450 mhz on the cheap end to 500 mhz on the fast end.
Those were the days. I still have a couple old Alpha heatsinks with what must've been an AMAZING 50 mm fan, weighing almost a whole third of a pound of solid aluminum. HA, fantastic. Two of the BP6's I built are still alive and kicking today. Dual powered celerons, whoo.
Compare that to my <a href="http://www.thermalright.com/a_page/main_pro
What are you grievances with Mr. Pennington. Just curious, thanks.
Myren
<blink> Test is flawed! </blink>
/w and w/o softupdate benched.
Checkout the CPU utilizations; reiserfs is pegged at 100% cpu utilization for ~8 tests. For a FS which describes itself as willing to use more CPU in order to achieve better I/O than the competition, running the benches on an antiquated 700 mhz machine is simply not fair.
OTOH, Untarring and tarring are notably NOT cpu limited, and still pretty lackluster for Reisers case. Disappointing, very disappointing. I was extremely impressed in the ext's; I simply had no idea how consistently well performing they were.
I'd also like to see FreeBSD's UFS
Myren
v frevbhfyl qbhog vg
I'll have to look for the demos. I wasnt aware of heirarchy. I thought it was just a size scaling replacement for toolbars. Helping people find stuff is defiantely important.
;-]
Anyways, we all know screen editing is the one true path, right?
reiserfs is no fun without the meta...
does that count as a double entranda?
i guess i didnt make my original post clear.
the kernel dev's still think of linux as a monolithic kernel. join the kernel if you want a working project; we will garuntee you no stable external facing API's... thats what they said to video drivers, and i'm amazed they made any consolation at all for fuse. but what happens when something like reiserfs comes along? it has to be shoehorned in and trimmed down to fit with the existing picture kernel-dev's have for the kernel. thats what i mean when i talk about stifling innovation. its not only a monolithic kernel, its a monolithic development process. its an entirely different form of monopoly, but it is one none the less, and its because linux is inherently monolithic. i'd love to see examples of how this has changed, but besides fuse, i'm not aware of many.
winfs is one of those big infinity projects, but it could be said that reiserfs is a ripoff off it (at least of some of the earlier different incarnations). conversely, singularity may be a rip off of HURD, another great infinity project, but it looks like some of singularity is actually going to make it to market. you're absolutely right, its a shame microsoft doesnt give credit where credit is due, but if your interested in isolation, interested in security, you know about EROS, HURD, & microkernel, and you've probably read some of the papers of people who did actual research.
i really dont see microsoft as using a more microkernel alike approach as means of suppressing others, i see it as the logical choice now that we have the performance to use microkernel systems. i'm not a shill for ms, i'm a shill for microkernel. i dont think i ever stated anything about ms doing goodwill, i just happen to be a technical purist; whether its money-sniffing ms or oss-touting linux following smart tracks, i think its important to note someone is doing something intelligent. i'm psyched to see someone actually delivering the one true microkernel path, even if i'm never going to frakking touch it.
Give and take, its all shades of gray.
Myren