It's worth noting that the patent most likely to be stepped on in dealing with MIPS is US patent 4,814,976, which covers the unaligned load/store instructions lwl, lwr, swl and swr. This patent expires 2006-12-26, which won't be long now. Google for "Lexra" "MIPS" and "Patent" for details of the various spats over the patent.
Apparently some of the more recent extensions fall under other patents, but the basic archetecture will be entirely unencumbered after this one expires. And as a Computer Enginering student I can tell you as ISAs go it's far and away the easiest useful one to impliment.
Since it didn't make it onto the "Related Stories" tab, I'll throw the good old Periodic Table Of Comic Books back into the discussion.
Just to keep it all over the table I am related to one of the people responsable for this.
To me the quality of the community, especially as shown by a distro's Wiki/Forum/IRC Channel is a big determinant in the desirability of the distro. I've been using ArchLinux for years, and one of it's strongest suits is its knowledgeable and within reason, patient and helpful community (along with great package management). If a quick search of the forums and wiki fail to answer your questions, someone on the IRC channel probably can; sometimes I leave the channel up in the background just to learn tricks from the more knowledgeable people hanging around. Keeping an eye on a good distro community can teach you all sorts of useful things. Also, never rule out a simple google search, if you are having a problem, there is a good chance someone else has had it too, so learn from their experience.
I have a stack of five origional Pentium boxes with 32mb of RAM and 2gb harddrives (except for one, with a larger drive for a software repository). Origionally built it to experiment with AFAPI based clustering, but since AFAPI is a reasonably non-invasive setup, it works well for trying other techniques too, everthing from simply running distcc on the nodes to speed up i586 software builds to briefly fiddling about with some of the other clustering options mentioned. Fiddling around with options on a real cluster (running cluster software on a single node really isn't a good impression) that could be reinstalled from scratch in a few hours, and the machines aren't worth enough to matter if it is physically damanged is a great way to learn.
Culture isn't something you create, sell, market, purchase, and/or trade. Culture is something that happens amoung a people. You're eating up the line the entertainment industry keeps making that they "produce" culture, when they at best ride the cultural wave, making products that pander to, and to some degree shape, the prevailing cultural movement. The sucess of indie films, blogs, grassroots movements, and services like YouTube should be evidence enough that they aren't in control. If a large part of the Chinese populance wants western entertainment, then that IS a part of the culture.
Uhm. If we're right shifting letters, that would be VMS->WNT->XOU. Doesnt make qute such a cute statement. XOU appears to be a company that supplies network appliacnes: http://www.xousolutions.com/ Unless we want to give credit to Sun's "The network is the computer", thats not an OS.
Anyone remember BeFS it came out in 1996, supported most of the "difficult and innovative" features WinFS was advertized to have, and WORKS. Its not quite relational, but it has extensive indexed metadata that makes it act as if it were. There's an open-source reimplimentation . Be, Inc. really did have some great technology, pity they couldnt make a buisness of it.
The "Throwing it off the balcony" part is just an extreme version of a common fix for stuck harddrives. Giving a dead drive, especially one that is "ticking", a good firm smack will often get it working again. Works best on small (2.5" or smaller) drives. I've resurrected quite a few drives with the same trick, including the one in my Rio Karma, as someone suggests here at riovolution . The way it works is sometimes the heads and/or platters will get stuck, possibly due to suction between the two, and the smack frees them. It often causes minor physical damage (a couple bad blocks), and a drive resurrected this way's days are probably numbered, but its great for fixing drives long enough to get the data off, or in the case of devices with nothing overwhelmingly important on them (like mp3 players) simply getting a few more weeks/months/years of use out of them.
I hadn't thought of it in those terms, if your information is accurate that ceartianly does place most all the blame on the labels (id say a 29 cent cut for distribution isnt particularly unreasonable, servers and bandwidth aren't free).
It also presents yet another good reason that music celeberties shouldnt be taken as role models...
At last count, the breakdown of where that $.99 goes is (on average):
Apple - $.35
Label - $.53
Artist - $.11
And thats only after the label reclaims whatever they claim they spent in production costs.
See http://www.downhillbattle.org/itunes/ for details.
If you want good integration with windows (read: PPTP), and want to keep it on a nice cheap *nix box, try Poptop . Runs on most any *nix, entirely compatible with the builtin PPTP support in recent versions of windows. I've been running it for my own purposes (admittedly not on a "small business" scale, only one or two users) for years on a modest linux box and it hasnt given me any trouble connecting from WinXP or linux clients.
When they released a overdesigned computer built from half-baked technology borrowed from the laptop line as the "20th Anniversary Mac". specs and extra pictures
Oh wait. Thats just like the iMac. Steve-o will have to try something else this time.
This is a perfect example of the problem with BSD licencing. Under the various BSD licences, its perfectly OK to take a piece of code and sell it, either modified or exactly as found, without in any way recognising or contrubuting to the project. Run "strings c:\windows\system32\ftp.exe" on a WinXP box and you'll see a perfect example of uncredited work. At least under the GPL if someone sells an unmodified program, the project will get recognition (since it will have to remain open source, and thus the origion of the code will be obvious), and if they sell a modified version the project will get the source for the modifications back. Neither directly equates to funding, but publicity and a better code base both help to attract financial support. Both arrangements depend somewhat on the cooperation and altruism of the entity using the code for a profit, but the GPL isn't quite so hopelessly naive.
Actually, if OpenSSH was GPL software this probably wouldnt be happening. As I understand it, the reason OpenSSH manages to bring so little money home despite its wide use is the BSD licenceing allows venders/leeches/etc. to make minor (or no?) modifications, and sell it without giving back to the project that created the code base. GPL software cannot have non-free forks soaking up all the income and not giving back to the project, because it does not allow closed derivitive works.
The headline made me think of an old sierra game set in the the Earthsiege (WOO GIANT ROBOTS) universe: Misson Force: Cyberstorm (its abadonware, download here). I picked it up from a bargin bin about 2 years after it came out, one of the only turn-based games I've ever enjoyed. Probably not related, but then again the flunkie that came up with the name could well be a gamer.
I'm quite found of Code:Blocks , except for a somewhat weak debugger frontend. Its actually written using wxWidgets, so it works just fine under Windows (wx binds to win32 api) AND Linux (wx binds to gtk2). I was actually just building myself a package for ArchLinux earlier today, as a quick warning the guy that packages up the releases runs Windows, so you have to fix the dos style line endings in the build scripts before building it, but it works just fine.
I've got a 6100/66 DOS across the room, which is another one of the systems with the 486DX based PDS card apple offered. It entertained me for weeks seeing what one can and cannot do with it. Interesting details:
1. It can run with its own memory (good) or borrow some from the mac (painfully slow). Uses the same 72 pin SIMMs as the mac.
2. It does NOT have conventional PC BIOS, so DOS and windows no higher than 95 work, but most other OSes will (mostly)not.
3. it IS possible to boot linux in a limited sort of way, if you use the syslinux bootloader (syslinux uses a windows bootsector to bootstrap your system). I never managed to get much info on the hardware with this trick, and havent had time to try more.
4. The mac-side software allows you limited clipboard (copy/paste) ability between the systems. I don't claim to understand how they pulled this off.
5. The PC can use physical volumes or drive containers (files) on the mac. Drive containers are really more reliable.
6. You switch between environments with Command-Return, the DOS card has a horrible octopus of a cable that connects it directly to the monitor (mac video on a passthrough), offers a gameport, etc.
7. Apparenty (I haven't tried myself) you can upgrade the card with ceartain newer pin-compatible intel and AMD chips.
There were also third party cards (even more modern PCI based ones with newer PC hardware) from Radius and Orange Micro that could be used with most kinds of mac. Not very useful, but great fun to play with now that they don't cost a thing.
Here in the US we have the "Patriot Act" from 2001, which grants the government a rather extrordinary collection of survalence powers. 16 sections of the bill expire on Dec. 31 of this year, including many of the electronic survalence portions that would be of interest to the/. crowd. The Republican party is trying to push a 4-year renewal through congress by the end of the year, but the 9/11 carte blanche that got it passed initally has run out and there is serious resistance this time around, including threats of a filibuster. Its worth noting that there are some minor improvements in the 4 year extension, and a more substantal overhaul, called the safe act, currently stalled in comitte. For those interested, some stories on cnet and msnbc (look, balance for the news bias whiners!)
ULi also appears to be the only company other than ATI making chipsets that support CrossFire (ATI's multi-GPU solution, competing with Nvidia's SLI, for the one person who doesnt know but cares) in the form of the ULi M1575. I cant imagine Nvidia will let that continue.
And if we focous all our attention on the "more important" things, after we are done with this round of "the war in $country, Social Security, Health Care, Education, the Economy and about a dozen other topics that deserve some attention", we will discover that all the little things have broken while we were crusading. Just because something isnt THE issue of the day doesnt mean it doesn't need to be taken care of.
I agree that old C-based OS design has issues, and that OOP design should be applied to OSes, for speed, security, and clarity reasons. However, it is my strong personal opinion that bytecode is never the right solution. To explain by metaphor:
Writing an OS in C is like rigging a ship starting from hemp. It takes forever, forces you to reinvent the wheel, and adds whole new layers of potential for accidents.
Writing an OS in Java (or C# or any bytecode language) is like rigging a ship with twine because you can hang yourself with rope. (also note: both are impractical, twine isn't strong enough, and bytecode needs an interpreter)
Writing an OS in an object-oriented memory protecting language (like C++) makes sense.
I'm certainly not saying they're doing this for the greater good, but look at Sony v. Universal (aka the betamax decision ), Sony was fighting for profits, but ended up establishing the substantial non-infringing use argument, inadvertently doing something for the greater good as a result of their "profit protecting". Also notable, the betamax format failed anyway, and the decision is now reviled by their entertainment divisions, the good part outlasted the greed.
Although this is immediately disgusting, in the not-so-long-run this might end up being a good thing, this is putting a kink in Sprint/Verison and Cingular's (the big mean companies with nearly inexhaustible legal resources) business model, who will likely lash out against it. If all goes well for them, it will end up creating a substantial precedent against this kind of business-method patent, which would inadvertently improve the patent law situation in the U.S., if we're lucky it might even catalyze a wider reform.
Slashdot Mirror
It's worth noting that the patent most likely to be stepped on in dealing with MIPS is US patent 4,814,976, which covers the unaligned load/store instructions lwl, lwr, swl and swr. This patent expires 2006-12-26, which won't be long now. Google for "Lexra" "MIPS" and "Patent" for details of the various spats over the patent.
Apparently some of the more recent extensions fall under other patents, but the basic archetecture will be entirely unencumbered after this one expires. And as a Computer Enginering student I can tell you as ISAs go it's far and away the easiest useful one to impliment.
Since it didn't make it onto the "Related Stories" tab, I'll throw the good old Periodic Table Of Comic Books back into the discussion.
Just to keep it all over the table I am related to one of the people responsable for this.
To me the quality of the community, especially as shown by a distro's Wiki/Forum/IRC Channel is a big determinant in the desirability of the distro. I've been using ArchLinux for years, and one of it's strongest suits is its knowledgeable and within reason, patient and helpful community (along with great package management). If a quick search of the forums and wiki fail to answer your questions, someone on the IRC channel probably can; sometimes I leave the channel up in the background just to learn tricks from the more knowledgeable people hanging around. Keeping an eye on a good distro community can teach you all sorts of useful things. Also, never rule out a simple google search, if you are having a problem, there is a good chance someone else has had it too, so learn from their experience.
I have a stack of five origional Pentium boxes with 32mb of RAM and 2gb harddrives (except for one, with a larger drive for a software repository). Origionally built it to experiment with AFAPI based clustering, but since AFAPI is a reasonably non-invasive setup, it works well for trying other techniques too, everthing from simply running distcc on the nodes to speed up i586 software builds to briefly fiddling about with some of the other clustering options mentioned. Fiddling around with options on a real cluster (running cluster software on a single node really isn't a good impression) that could be reinstalled from scratch in a few hours, and the machines aren't worth enough to matter if it is physically damanged is a great way to learn.
Culture isn't something you create, sell, market, purchase, and/or trade. Culture is something that happens amoung a people. You're eating up the line the entertainment industry keeps making that they "produce" culture, when they at best ride the cultural wave, making products that pander to, and to some degree shape, the prevailing cultural movement. The sucess of indie films, blogs, grassroots movements, and services like YouTube should be evidence enough that they aren't in control. If a large part of the Chinese populance wants western entertainment, then that IS a part of the culture.
Uhm. If we're right shifting letters, that would be VMS->WNT->XOU. Doesnt make qute such a cute statement. XOU appears to be a company that supplies network appliacnes: http://www.xousolutions.com/ Unless we want to give credit to Sun's "The network is the computer", thats not an OS.
Anyone remember BeFS it came out in 1996, supported most of the "difficult and innovative" features WinFS was advertized to have, and WORKS. Its not quite relational, but it has extensive indexed metadata that makes it act as if it were. There's an open-source reimplimentation . Be, Inc. really did have some great technology, pity they couldnt make a buisness of it.
As Something Positive reminds us, we should all take a moment to talk about the dangers of molasses.
[part1] and [part2]
The "Throwing it off the balcony" part is just an extreme version of a common fix for stuck harddrives. Giving a dead drive, especially one that is "ticking", a good firm smack will often get it working again. Works best on small (2.5" or smaller) drives. I've resurrected quite a few drives with the same trick, including the one in my Rio Karma, as someone suggests here at riovolution . The way it works is sometimes the heads and/or platters will get stuck, possibly due to suction between the two, and the smack frees them. It often causes minor physical damage (a couple bad blocks), and a drive resurrected this way's days are probably numbered, but its great for fixing drives long enough to get the data off, or in the case of devices with nothing overwhelmingly important on them (like mp3 players) simply getting a few more weeks/months/years of use out of them.
I hadn't thought of it in those terms, if your information is accurate that ceartianly does place most all the blame on the labels (id say a 29 cent cut for distribution isnt particularly unreasonable, servers and bandwidth aren't free).
It also presents yet another good reason that music celeberties shouldnt be taken as role models...
At last count, the breakdown of where that $.99 goes is (on average):
Apple - $.35
Label - $.53
Artist - $.11
And thats only after the label reclaims whatever they claim they spent in production costs.
See http://www.downhillbattle.org/itunes/ for details.
If you want good integration with windows (read: PPTP), and want to keep it on a nice cheap *nix box, try Poptop . Runs on most any *nix, entirely compatible with the builtin PPTP support in recent versions of windows. I've been running it for my own purposes (admittedly not on a "small business" scale, only one or two users) for years on a modest linux box and it hasnt given me any trouble connecting from WinXP or linux clients.
Wow. Now /. is posting stories as obfuscated by blogs as the average post on digg.
When they released a overdesigned computer built from half-baked technology borrowed from the laptop line as the "20th Anniversary Mac". specs and extra pictures
Oh wait. Thats just like the iMac. Steve-o will have to try something else this time.
This is a perfect example of the problem with BSD licencing. Under the various BSD licences, its perfectly OK to take a piece of code and sell it, either modified or exactly as found, without in any way recognising or contrubuting to the project. Run "strings c:\windows\system32\ftp.exe" on a WinXP box and you'll see a perfect example of uncredited work. At least under the GPL if someone sells an unmodified program, the project will get recognition (since it will have to remain open source, and thus the origion of the code will be obvious), and if they sell a modified version the project will get the source for the modifications back. Neither directly equates to funding, but publicity and a better code base both help to attract financial support. Both arrangements depend somewhat on the cooperation and altruism of the entity using the code for a profit, but the GPL isn't quite so hopelessly naive.
Actually, if OpenSSH was GPL software this probably wouldnt be happening. As I understand it, the reason OpenSSH manages to bring so little money home despite its wide use is the BSD licenceing allows venders/leeches/etc. to make minor (or no?) modifications, and sell it without giving back to the project that created the code base. GPL software cannot have non-free forks soaking up all the income and not giving back to the project, because it does not allow closed derivitive works.
The headline made me think of an old sierra game set in the the Earthsiege (WOO GIANT ROBOTS) universe: Misson Force: Cyberstorm (its abadonware, download here). I picked it up from a bargin bin about 2 years after it came out, one of the only turn-based games I've ever enjoyed. Probably not related, but then again the flunkie that came up with the name could well be a gamer.
I'm quite found of Code:Blocks , except for a somewhat weak debugger frontend. Its actually written using wxWidgets, so it works just fine under Windows (wx binds to win32 api) AND Linux (wx binds to gtk2). I was actually just building myself a package for ArchLinux earlier today, as a quick warning the guy that packages up the releases runs Windows, so you have to fix the dos style line endings in the build scripts before building it, but it works just fine.
I've got a 6100/66 DOS across the room, which is another one of the systems with the 486DX based PDS card apple offered. It entertained me for weeks seeing what one can and cannot do with it. Interesting details: 1. It can run with its own memory (good) or borrow some from the mac (painfully slow). Uses the same 72 pin SIMMs as the mac. 2. It does NOT have conventional PC BIOS, so DOS and windows no higher than 95 work, but most other OSes will (mostly)not. 3. it IS possible to boot linux in a limited sort of way, if you use the syslinux bootloader (syslinux uses a windows bootsector to bootstrap your system). I never managed to get much info on the hardware with this trick, and havent had time to try more. 4. The mac-side software allows you limited clipboard (copy/paste) ability between the systems. I don't claim to understand how they pulled this off. 5. The PC can use physical volumes or drive containers (files) on the mac. Drive containers are really more reliable. 6. You switch between environments with Command-Return, the DOS card has a horrible octopus of a cable that connects it directly to the monitor (mac video on a passthrough), offers a gameport, etc. 7. Apparenty (I haven't tried myself) you can upgrade the card with ceartain newer pin-compatible intel and AMD chips. There were also third party cards (even more modern PCI based ones with newer PC hardware) from Radius and Orange Micro that could be used with most kinds of mac. Not very useful, but great fun to play with now that they don't cost a thing.
Here in the US we have the "Patriot Act" from 2001, which grants the government a rather extrordinary collection of survalence powers. 16 sections of the bill expire on Dec. 31 of this year, including many of the electronic survalence portions that would be of interest to the /. crowd. The Republican party is trying to push a 4-year renewal through congress by the end of the year, but the 9/11 carte blanche that got it passed initally has run out and there is serious resistance this time around, including threats of a filibuster. Its worth noting that there are some minor improvements in the 4 year extension, and a more substantal overhaul, called the safe act, currently stalled in comitte. For those interested, some stories on cnet and msnbc (look, balance for the news bias whiners!)
ULi also appears to be the only company other than ATI making chipsets that support CrossFire (ATI's multi-GPU solution, competing with Nvidia's SLI, for the one person who doesnt know but cares) in the form of the ULi M1575. I cant imagine Nvidia will let that continue.
And if we focous all our attention on the "more important" things, after we are done with this round of "the war in $country, Social Security, Health Care, Education, the Economy and about a dozen other topics that deserve some attention", we will discover that all the little things have broken while we were crusading. Just because something isnt THE issue of the day doesnt mean it doesn't need to be taken care of.
I'm certainly not saying they're doing this for the greater good, but look at Sony v. Universal (aka the betamax decision ), Sony was fighting for profits, but ended up establishing the substantial non-infringing use argument, inadvertently doing something for the greater good as a result of their "profit protecting". Also notable, the betamax format failed anyway, and the decision is now reviled by their entertainment divisions, the good part outlasted the greed.
Although this is immediately disgusting, in the not-so-long-run this might end up being a good thing, this is putting a kink in Sprint/Verison and Cingular's (the big mean companies with nearly inexhaustible legal resources) business model, who will likely lash out against it. If all goes well for them, it will end up creating a substantial precedent against this kind of business-method patent, which would inadvertently improve the patent law situation in the U.S., if we're lucky it might even catalyze a wider reform.