ya know this whole attitude of:
by having a project which is not-for-profit, and realeases source code under a free license, i am within my rights to infringe on anyone's patent or trademark or any other form of IP.
It's just rubbish. Whether or not you "believe" in Intellectual property has no bearing on whether it exists, and if it is legally binding.
Fighting the system through simple noncompliance is not the answer, talk to your govt reps, and demand action. If nobody complains, then nothing changes. The whole point of democracy is that you stand up and be heard, not just bitch and moan cuz you're not in charge. And do you really believe that gov't cares about anything but money anyway? Well, guess who has more money? The people do, and by the way, corporations can't vote, so if you stop being a sheep and watchin all the political ads, and do something. Then the system might change. If you've never actively participated in the system ( no, just voting is not enough) then you're not doing enough to make sure you're interests are represented. Support a canidate that REALLY represents you, then maybe we wouldn't have all these cookie-cutter, do anything for a buck congressional whores deciding how we're gonna plan our collective futures.
To me it sounds like you are talking about distributing software in a "free" model, but in one which restricts anyone from profiting from its use(and/or)distribution. If that's you're talking about, then that's not free software. That's shareware. "Free software" licensing is designed to protect the intellectual property (or maybe just ordinary property) rights of the owner, while still granting the public an unrestricted right to use the software. Public domain software is essentially the same thing as free, but without the intellectual property protection.
If you want to distribute software, but not unconditionally, then yes, you would need to license it. That's what a license is for, to set the terms under which you make you property available to others. I don't think the need for licenses has anything do do with the real or imagined existence of a "litigation crazed atmosphere in the US" (not that i believe it is imagined), but with the universaly accepted concept of "ownership" and how an owner can protect their property which is made availble to others.
if the whole point here was that farmers need to buy/plant new seed every year to get the crops to re-grow, then even if you plant a whole farm with this seed, next year none of the plans from that seed will grow again. Not a very effective "weed" if you ask me.
> If Libranet isn't making money off their
> distribution then tough. There's no such thing as
>a right-to-profit. You work for a start-up you
>take your risks. Sometimes you get burned. Find a
> new business plan, guys.
i think that's exactly what they have done. Not that you should have noticed;-)
-earl
You know, if you are so impressed with the quality of the OpenSSH/OpenSSL code that you want to use it for your own projects.... you might consider that the code quality is a product of both programming skill, and peer review. If you hope for your own projects to have a similar quality of code, you might consider not just copying the code from those projects, but also emulating the development process they've used.
That said, both codebases are under licenses which allow you to copy code (1 - OpenSSL) as long as you don't claim the finished product to be the original, or use the name of the original to promote your work. (2 - OpenSSL) That derived code include the copyright notice from the original code. And (3 - OpenSHH) that you credit the originators for their work.
So the bottom line is yes, you can use it. GPL code however requires that when redistributing GPLed code, you license it under the GPL, include source code for no additional cost. If you only copy and do not redistribute GPLed code, (i.e. for a project strictly internal to your company/organization) then all bets are off, and you can do with it as you please.
the competition thing is a mixed blessing, on the one hand, Intel and AMD are devoting nearly all of their resources to getting the fastest x86 chip out the door, and it makes them lean and mean. On the other hand, it's a profitable market, but margins are razor thin for both parties. If you don't have the profits to do the R&D, and to build the fabs, it's hard to stay comptitive in that cut-throat market. IBM gets the benefit of having a diversified product strategy, and also because they use PowerPC in alot of their own systems (RS/6000, big iron) they get to sell PPC systems in two higher (than x86 desktop) margin markets: midrange and high-end servers. Plus they get service and support revenues from the product lines as well.
And as we all know, it takes a whole lot more than CPU power to make a powerful server. Ask Sun, they have one of the weakest processor offeings in the server market, but because they build balanced systems, having a weak CPU isn't hurting them. But that is beside the point.
Intel and AMD already own the desktop CPU market, and when they offer 64-bit CPUs, AMD will pull in a big share of the mid-range server market but they'll both expand into the the high-end server market as well.
Don't forget that IBM is part of the PPC triumverate. In my eyes, over the last 5 years, not even Intel can hold a candle to IBM in terms of making advnces in microprocessor fabrication, and design. Things like copper interconnects, and SOI have helped IBM make lots of money licensing technology, or simply manufacturing other people's chips with IBM's state of the art processes.
It wasn't too long ago that AMD had a majority or their chips made by IBM. (or was that cyrix? i don't remember)
yeah checkpoint FW-1 handles these things pretty well, especially in combination with their MetaIP product. I'll let checkpoint toot their own horn, but the bottom line is that MetaIP integrates your DNS with your DHCP, and from there you can create firewall rules based on date from the MetaIP system.
i.e. you can create rules that will effectively be based on MAC address, since MetaIP can tell the firewall what IP is assigned to that MAC address. of course you could create dns based rules with out MetaIP, so that one's a gimme.
Pardon my french, but that's bullshit. The government isn't bankrolling the development or deployment, or anything about "broadband" access to the internet. And the vast majority of the US does not have access to it. So how are you gonna tell me that you're entitled to it. If private industry decides not to offer it to you, you're pretty much SOL, if you are not offered electric power, it's a totally different story.
BTW, So when was the last time there were subsidies for low-income families who couldn't afford T-1's??
unless you live on the moon, you do have a choice of ISPs. You might not have a choice of of providers of "broadband" Internet service, but pretty much anywhere you can use a 56K modem to dial a local number to get to the internet.
If you don't feel the service @home provides is acceptable, stop subscribing to it. If you feel their corporate strategy is not ok, stop supporting it with your money. So many people seem to miss the point that corporations really dont have much cash, what they have is alot of income. When that income no longer comes in, a corporation is just as vulnerable as you or I. And when you or I turn our income into saved cash, we are no longer as vulnerable to the economic power corporations have over us. but alas, i've already spent my paycheck...;-)
Yeah, it was sort of a spinoff type of thing, but I've always considered Debian to be the FSF distro. The project is idealogically very close to the FSF, as seen by Debians practice of seperating software into 2 categories: free and non-free.
-earl
alright,
I've reached my breaking point. Everyone is entitled to their opinion, but what i'm reading here looks much more like a RMS bashfest, than a reasonable discussion of the article at hand.
As I understand it, the Free Software concept is by design an unreversible solution to the issue of proprietary (closed source, restricted, etc) software. Especially how in certain circumstances, companies/organizations with vested interests in proprietary software have co-opted/abused what was intended to be Free Software, and claimed it as a part of their proprietary software. Notorious examples include the ATT/BSD lawsuit, and the original Emacs written by RMS.
What it seems RMS seeks to do, is define the concept of Free Software in a totally unambiguous way, such that both accidental and malicous abuse of "Software Freedom" will stand in clear violation of the software license.
I personally agree with this philosophy, not because I want RMS to "win" any battle for power in the computer software industry, but because in my experience software that is "Free Software" (as in speech) does more to benefit everyone than the major alternatives. If you think of the software world as a matrix (no not the movie), you get something like this:
Well, that took forever.;-) So the point i'm trying to make is this:
RMS does not seek to force anyone to become an advocate of, or developer of any kind of software, what he does is try to persuade software developers and users to become advocates of Free Software, and thus refuse to use other types of software that restrict a persons rights to the following four things:
1) The freedom to run the program, for any purpose (freedom 0).
2) The freedom to study how the program works, and adapt it to your needs (freedom 1).
3) The freedom to redistribute copies so you can help your neighbor (freedom 2).
4) The freedom to improve the program, and release your improvements to the public, so that the whole community benefits. (freedom 3)
I hope i've provided some clarity to those who may have misunderstood the topic, and to those who disagree still, I'm sorry that we don't see it the same way. I see it as a concept similar to political freedom in America, either you have the freedom to vote for a representative of your choice, or you don't. Having either a vote or a (selection of) representatives does not make you politically free, you need both. Just as having either the freedom to redistribute software or the freedom to access the source code is not enough. You need both. And only be guaranteing both, do we ensure that our software will always have both, and that it will provide the most benefit to both the user and developer communities.
on the subject of forking...
whyn do we need one huge kernel anyway? Probably several kernels are needed. One for big-ass servers, one for tiny-ass routers, one for mainstream workstations, and one TBD. Having one all encompasing kernel makes building the kernel a pain in the ass. I've been using linux for four years, and I still have to build my kernels a couple times before I get it right. So many freakin options, i'm bound to get something wrong.
If you're really in the sticks, you're best bet is sattelite, you can get bi-directional (real soon now) at real good speeds from gilat sattelite (www.gilat-to-home.com). They won't release final specs, but i've seen it, and it's fast. Cable is you're second best bet, but you probably pay out the nose for rural cable service, and from my experience, you don't get genuine "internet" from a cable comapany. You'll probably get a few incoming ports blocked(25,80,135,137,139), and you might even see some outgoing soon (napster, gnutella, etc).
Like you said, i wouldn't ever expect to see DSL out your way. GL
I can speak from experience on this one. I took an intro to c++ class a few years ago, and they were using borland c++. Rather than pay $90 at the bookstore, and install windows on my machine, I just used gcc.
Everything worked flawlessly until the final. I had a 98 average in the class, but the final exam was to take a program and modify it. Well wouldn't ya know it, the stoopid program wouldn't compile under gcc, cuz of a few function calls depended on a library that was different in gcc than it was on borland.
So being the procrastinator that i am, i found this out the night before the final was due, and by 8 the next morning i had the program compiling/running with gcc. It was due at 10. So I had to hand my port to gcc, but of course i got squat on the final, and my final grade ended up being a b. sucks huh??
the moral of the story is, yes there are valid reasons for a class to require a consistent work environment. This is a prime example, If i'd used the borland thing, i wouldn't have had that problem, but i'd be out 90 bucks and i'd have to have used windows. I'll take the B.;-)
I have to agree with you on this one. But with exceptions. MS didn't "develop an OS that supports a wide range of hardware". What they did is maintain and improve a codebase (MS-DOS) that was the defacto standard for personal computing. This was accomplished with fairly little effort: a few file system changes a few performance optimizations and a few extensions to provide new functionality (networking, CD-ROM's)
Making NT is Microsoft's single biggest accomplishment. NT is to MS-DOS what today's UNIX is to CPM: Several orders of magnitude more functionality, far better performance under load, a modern feature set, and runs on both destop to server.(Now I know that UNIX is *NOT* a derivative of CPM, and I know somebody would have tried to point that out if I didn't say it). So, they get half credit. They rebuilt thier core OS from scratch, yet maintained compatibility (mostly) for old devices/APIs/interfaces. But they haven't eliminated the usability and stability problems that plagued DOS. Maybe when they become a real "OS" comapany (as opposed to a "monopoly maintaining" company) they'll start to fix some of those bugs that cause all those damn blue screens, and refine that interface a bit, and actually add features in a sensible and consistent way. They might even make some products that can compete head to head with UNIX on servers. It could happen... really!!;-)
i got one. it's excellent. yes it does take time to write the pics to the disc(5-10 sec for hi res or for movies), and i haven't found a way to get the camera to work under linux. The trick is that the camera has a usb port, and when you put the camera in pc mode the pc will just see a usb ls-120 drive with some jpeg's on it. cool huh? no special software required. So far i've only got it to work under 95, no NT, no win2k (yet). If anybody can tell me how to nake this happen in linux, you'll make a new friend;) It's a 1.3Mpixel camera. this picture quality is real good, nice zoom. It can take low/high res pictures, low res can be in (extra) zoom mode, a quick 5 shot sequence, or a 10 sec quicktime movie (with sound!) fsckin cool! the best part is that it lists for about a grand, but i found it for only $600 at www.harmonycomputers.com, they charge too much for shipping, but the total is still at least $100 below the rest.
I didn't get the impression that he's holding his cards close to his chest on this one. It wouldn't really make any sense to do so on slashdot either, especially after agreeing to the interview in the first place. If he just wanted to get some publicity for the project/company, a site more oriented towards managers instead of tech would have ben a smarter move, but i digress... It sounds more like because of his prior experience with implementing clustering technology, he was already well aware of the major pitfalls. That said, You're really just hooking up a bunch of machines to a network, setting up a shared data source, and configuring the parallel procesing software. The design of an effective solution (given the requisite expertise) really can't be that hard. Now if he had designed a major component of the architecture (which he admits was all done by third parties) then we'd have a real case study to look at.
Either way though, i'd like to know more about the particulars of the cluster management software. The configuration management piece sounds particularly useful for large groups of identical machines. does anyone know if this is publicly available software, or if there is a publicly available quivalent?? -earl
Re:Please answer these questions?
on
SuSE For PPC
·
· Score: 3
well there's already a project in progress to build an open powerpc motherboard that has all the standard pc components. IBM has basicly given to any serious takers the full specs for their reference ppc board. The board had pci/isa, pc100 ram, udma33, up to 1GB RAM, and the standard ppc bridge chips (north and south bridge). Take a look at ibm's linux news page for more details: They've got linux booting on it now, and the process is takin time, but once a real board hits the market, we'll probably see several ambitious pc board makers jump on board trying to differetiate them selves in this commodity market. Now, as this will clearly have lower volume than the x86 board market, we should expect relatively higher prices, but as people come to realize that powerpc without apples pricing is actually very price competitive, and the initial costs are recovered, the prices ought to come down a bit. I've been looking forward to a ppc board market similar to the x86 board market for awhile. Fortunately the one-two punch of linux's cross platform compatability, and the fact that most of the apps that run on linux are open source means that having linux on ppc become more of an equal to intel and alpha linux is a realistic goal on the short term. That was the biggest problem with NT on non-intel platforms, sw companies had to invest resources in porting their x86 centric apps to non-intel, and then attempt to support them on very low volume platforms. With open source, the user community can ensure that the apps have cross platform compatibility and support, while the developers have the option of doing that work, developing new features, or even doing nothing. The code lives on regardless.
so, I'll get off my risc soapbox now, but I hope i've made the point that we can expect to see much broader support for linux on powerpc real soon now, and who knows, with the kind of performance you could get from altivec with intensive apps like encryption, packet filtering, compression, rendering, etc should make it easy to see that there are genuine advantages to working with alternative technologies. appliance wise, you would think that one of these security vendors would jump all over the opportunity to make a high performance security appliance that can do some serious vpn-ing at a relatively low cost, or even if you think about Be selling ultra low overhead (and cheap) rendering farms of dual G4s to graphics designers or digital animators or whatever they call themselves these days. That would be fscking cool! (never mind the mandatory references to beowulf clusters, or even just seeing natalie portman straddling one of those bad boys;)
i know we've all talked this one to death, but I'm really conflicted on this one. Apple opening the source to Darwin is a good thing, and ought to be encouraged, if they are diligent about putting together regular updaters for approved patches, and stay^H^H^H^Hget on top of security issues then there's a real benefit to having the source available. Otherwise it's just a bonus for app developers and porters.
What is(?can?) apple gonna do to make this really significant? If they don't open up the corresponding api's, then security remains obscure, and writing real carbonized apps remains the same as ever. Can apple provide a real benefit to users/developers with just darwin, or would they have to go further? I'm curious what y'all think. -earl
no, no, no you're misunderstanding the GPL. When you GPL program, you continue to reatin full copyright, what you do is allow unrevokable, unlimited redistribution and modification by anyone. The one stipulation is that when you redistribute it to a third party, you must also redristribute the source, at no additional cost. you still own the software, and you can re-release it under another license if you wish, or you can even make changes, and redristribute without source. Only those who you license it to (everyone who doesen't own the copyright to the software) must abide by the licensing terms. Just like microsoft still owns the software that they license. They could even change the license (specifically, license terms to new licensees) on a whim. But it dosen't make sense that they'd have to pay for a copy right? They don't license it to themselves.
Slashdot Mirror
by having a project which is not-for-profit, and realeases source code under a free license, i am within my rights to infringe on anyone's patent or trademark or any other form of IP.
It's just rubbish. Whether or not you "believe" in Intellectual property has no bearing on whether it exists, and if it is legally binding.
Fighting the system through simple noncompliance is not the answer, talk to your govt reps, and demand action. If nobody complains, then nothing changes. The whole point of democracy is that you stand up and be heard, not just bitch and moan cuz you're not in charge.
And do you really believe that gov't cares about anything but money anyway? Well, guess who has more money? The people do, and by the way, corporations can't vote, so if you stop being a sheep and watchin all the political ads, and do something. Then the system might change. If you've never actively participated in the system ( no, just voting is not enough) then you're not doing enough to make sure you're interests are represented. Support a canidate that REALLY represents you, then maybe we wouldn't have all these cookie-cutter, do anything for a buck congressional whores deciding how we're gonna plan our collective futures.
by the way, no i'm not bitter.
-earl
To me it sounds like you are talking about distributing software in a "free" model, but in one which restricts anyone from profiting from its use(and/or)distribution. If that's you're talking about, then that's not free software. That's shareware.
"Free software" licensing is designed to protect the intellectual property (or maybe just ordinary property) rights of the owner, while still granting the public an unrestricted right to use the software. Public domain software is essentially the same thing as free, but without the intellectual property protection.
If you want to distribute software, but not unconditionally, then yes, you would need to license it. That's what a license is for, to set the terms under which you make you property available to others. I don't think the need for licenses has anything do do with the real or imagined existence of a "litigation crazed atmosphere in the US" (not that i believe it is imagined), but with the universaly accepted concept of "ownership" and how an owner can protect their property which is made availble to others.
if the whole point here was that farmers need to buy/plant new seed every year to get the crops to re-grow, then even if you plant a whole farm with this seed, next year none of the plans from that seed will grow again. Not a very effective "weed" if you ask me.
-earl
> distribution then tough. There's no such thing as
>a right-to-profit. You work for a start-up you
>take your risks. Sometimes you get burned. Find a
> new business plan, guys.
i think that's exactly what they have done. ;-)
Not that you should have noticed
-earl
-earl
That said, both codebases are under licenses which allow you to copy code (1 - OpenSSL) as long as you don't claim the finished product to be the original, or use the name of the original to promote your work. (2 - OpenSSL) That derived code include the copyright notice from the original code. And (3 - OpenSHH) that you credit the originators for their work.
So the bottom line is yes, you can use it. GPL code however requires that when redistributing GPLed code, you license it under the GPL, include source code for no additional cost. If you only copy and do not redistribute GPLed code, (i.e. for a project strictly internal to your company/organization) then all bets are off, and you can do with it as you please.
-earl
the competition thing is a mixed blessing, on the one hand, Intel and AMD are devoting nearly all of their resources to getting the fastest x86 chip out the door, and it makes them lean and mean. On the other hand, it's a profitable market, but margins are razor thin for both parties. If you don't have the profits to do the R&D, and to build the fabs, it's hard to stay comptitive in that cut-throat market. IBM gets the benefit of having a diversified product strategy, and also because they use PowerPC in alot of their own systems (RS/6000, big iron) they get to sell PPC systems in two higher (than x86 desktop) margin markets: midrange and high-end servers. Plus they get service and support revenues from the product lines as well.
And as we all know, it takes a whole lot more than CPU power to make a powerful server. Ask Sun, they have one of the weakest processor offeings in the server market, but because they build balanced systems, having a weak CPU isn't hurting them. But that is beside the point.
Intel and AMD already own the desktop CPU market, and when they offer 64-bit CPUs, AMD will pull in a big share of the mid-range server market but they'll both expand into the the high-end server market as well.
It wasn't too long ago that AMD had a majority or their chips made by IBM. (or was that cyrix? i don't remember)
-earl
i.e. you can create rules that will effectively be based on MAC address, since MetaIP can tell the firewall what IP is assigned to that MAC address. of course you could create dns based rules with out MetaIP, so that one's a gimme.
-earl
BTW, So when was the last time there were subsidies for low-income families who couldn't afford T-1's??
-earl
If you don't feel the service @home provides is acceptable, stop subscribing to it. If you feel their corporate strategy is not ok, stop supporting it with your money.
So many people seem to miss the point that corporations really dont have much cash, what they have is alot of income. When that income no longer comes in, a corporation is just as vulnerable as you or I. And when you or I turn our income into saved cash, we are no longer as vulnerable to the economic power corporations have over us. but alas, i've already spent my paycheck...
-earl
Yeah, it was sort of a spinoff type of thing, but I've always considered Debian to be the FSF distro. The project is idealogically very close to the FSF, as seen by Debians practice of seperating software into 2 categories: free and non-free.
-earl
I've reached my breaking point. Everyone is entitled to their opinion, but what i'm reading here looks much more like a RMS bashfest, than a reasonable discussion of the article at hand.
As I understand it, the Free Software concept is by design an unreversible solution to the issue of proprietary (closed source, restricted, etc) software. Especially how in certain circumstances, companies/organizations with vested interests in proprietary software have co-opted/abused what was intended to be Free Software, and claimed it as a part of their proprietary software. Notorious examples include the ATT/BSD lawsuit, and the original Emacs written by RMS.
What it seems RMS seeks to do, is define the concept of Free Software in a totally unambiguous way, such that both accidental and malicous abuse of "Software Freedom" will stand in clear violation of the software license.
I personally agree with this philosophy, not because I want RMS to "win" any battle for power in the computer software industry, but because in my experience software that is "Free Software" (as in speech) does more to benefit everyone than the major alternatives. If you think of the software world as a matrix (no not the movie), you get something like this:
================================================== ============================ ============================ ============================ ===========================
====================== Source code available ====== Binary only ========================
=================================================
== Freely redistributable ===== "Free Software" ======== "Freeware" ======================
=================================================
== Licensed by contract ===== "Open Source" ========= "Billware" =======================
=================================================
Well, that took forever. ;-) So the point i'm trying to make is this:
RMS does not seek to force anyone to become an advocate of, or developer of any kind of software, what he does is try to persuade software developers and users to become advocates of Free Software, and thus refuse to use other types of software that restrict a persons rights to the following four things:
1) The freedom to run the program, for any purpose (freedom 0).
2) The freedom to study how the program works, and adapt it to your needs (freedom 1).
3) The freedom to redistribute copies so you can help your neighbor (freedom 2).
4) The freedom to improve the program, and release your improvements to the public, so that the whole community benefits. (freedom 3)
I hope i've provided some clarity to those who may have misunderstood the topic, and to those who disagree still, I'm sorry that we don't see it the same way. I see it as a concept similar to political freedom in America, either you have the freedom to vote for a representative of your choice, or you don't. Having either a vote or a (selection of) representatives does not make you politically free, you need both. Just as having either the freedom to redistribute software or the freedom to access the source code is not enough. You need both. And only be guaranteing both, do we ensure that our software will always have both, and that it will provide the most benefit to both the user and developer communities.
-earl
-earl
whyn do we need one huge kernel anyway? Probably several kernels are needed. One for big-ass servers, one for tiny-ass routers, one for mainstream workstations, and one TBD. Having one all encompasing kernel makes building the kernel a pain in the ass. I've been using linux for four years, and I still have to build my kernels a couple times before I get it right. So many freakin options, i'm bound to get something wrong.
but that's just my opioion...
-earl
there are lots of ways someone can id your box over the network, regardless of the service headers.
And BTW, if you're gonna print identifying headers, why print sendmail? print postfix, or qmail, or nothing! ;-)
-earl
Like you said, i wouldn't ever expect to see DSL out your way. GL
-earl
Everything worked flawlessly until the final. I had a 98 average in the class, but the final exam was to take a program and modify it. Well wouldn't ya know it, the stoopid program wouldn't compile under gcc, cuz of a few function calls depended on a library that was different in gcc than it was on borland.
So being the procrastinator that i am, i found this out the night before the final was due, and by 8 the next morning i had the program compiling/running with gcc. It was due at 10. So I had to hand my port to gcc, but of course i got squat on the final, and my final grade ended up being a b. sucks huh??
the moral of the story is, yes there are valid reasons for a class to require a consistent work environment. This is a prime example, If i'd used the borland thing, i wouldn't have had that problem, but i'd be out 90 bucks and i'd have to have used windows. I'll take the B.
a few file system changes
a few performance optimizations
and a few extensions to provide new functionality (networking, CD-ROM's)
Making NT is Microsoft's single biggest accomplishment. NT is to MS-DOS what today's UNIX is to CPM: Several orders of magnitude more functionality, far better performance under load, a modern feature set, and runs on both destop to server.(Now I know that UNIX is *NOT* a derivative of CPM, and I know somebody would have tried to point that out if I didn't say it). ;-)
So, they get half credit. They rebuilt thier core OS from scratch, yet maintained compatibility (mostly) for old devices/APIs/interfaces. But they haven't eliminated the usability and stability problems that plagued DOS. Maybe when they become a real "OS" comapany (as opposed to a "monopoly maintaining" company) they'll start to fix some of those bugs that cause all those damn blue screens, and refine that interface a bit, and actually add features in a sensible and consistent way. They might even make some products that can compete head to head with UNIX on servers. It could happen... really!!
nope, if you have another ls-120 drive (which i don't) you can just put the "superdisk" in there, and go to town.
-earl
The trick is that the camera has a usb port, and when you put the camera in pc mode the pc will just see a usb ls-120 drive with some jpeg's on it. cool huh? no special software required. So far i've only got it to work under 95, no NT, no win2k (yet). If anybody can tell me how to nake this happen in linux, you'll make a new friend
It's a 1.3Mpixel camera. this picture quality is real good, nice zoom. It can take low/high res pictures, low res can be in (extra) zoom mode, a quick 5 shot sequence, or a 10 sec quicktime movie (with sound!) fsckin cool!
the best part is that it lists for about a grand, but i found it for only $600 at www.harmonycomputers.com, they charge too much for shipping, but the total is still at least $100 below the rest.
-earl
p.s. here's my favorite pic favorite pic so far:
It sounds more like because of his prior experience with implementing clustering technology, he was already well aware of the major pitfalls.
That said, You're really just hooking up a bunch of machines to a network, setting up a shared data source, and configuring the parallel procesing software. The design of an effective solution (given the requisite expertise) really can't be that hard. Now if he had designed a major component of the architecture (which he admits was all done by third parties) then we'd have a real case study to look at.
Either way though, i'd like to know more about the particulars of the cluster management software. The configuration management piece sounds particularly useful for large groups of identical machines. does anyone know if this is publicly available software, or if there is a publicly available quivalent??
-earl
They've got linux booting on it now, and the process is takin time, but once a real board hits the market, we'll probably see several ambitious pc board makers jump on board trying to differetiate them selves in this commodity market. Now, as this will clearly have lower volume than the x86 board market, we should expect relatively higher prices, but as people come to realize that powerpc without apples pricing is actually very price competitive, and the initial costs are recovered, the prices ought to come down a bit.
I've been looking forward to a ppc board market similar to the x86 board market for awhile. Fortunately the one-two punch of linux's cross platform compatability, and the fact that most of the apps that run on linux are open source means that having linux on ppc become more of an equal to intel and alpha linux is a realistic goal on the short term. That was the biggest problem with NT on non-intel platforms, sw companies had to invest resources in porting their x86 centric apps to non-intel, and then attempt to support them on very low volume platforms. With open source, the user community can ensure that the apps have cross platform compatibility and support, while the developers have the option of doing that work, developing new features, or even doing nothing. The code lives on regardless.
so, I'll get off my risc soapbox now, but I hope i've made the point that we can expect to see much broader support for linux on powerpc real soon now, and who knows, with the kind of performance you could get from altivec with intensive apps like encryption, packet filtering, compression, rendering, etc should make it easy to see that there are genuine advantages to working with alternative technologies. appliance wise, you would think that one of these security vendors would jump all over the opportunity to make a high performance security appliance that can do some serious vpn-ing at a relatively low cost, or even if you think about Be selling ultra low overhead (and cheap) rendering farms of dual G4s to graphics designers or digital animators or whatever they call themselves these days. That would be fscking cool! (never mind the mandatory references to beowulf clusters, or even just seeing natalie portman straddling one of those bad boys ;)
Apple opening the source to Darwin is a good thing, and ought to be encouraged, if they are diligent about putting together regular updaters for approved patches, and stay^H^H^H^Hget on top of security issues then there's a real benefit to having the source available. Otherwise it's just a bonus for app developers and porters.
What is(?can?) apple gonna do to make this really significant? If they don't open up the corresponding api's, then security remains obscure, and writing real carbonized apps remains the same as ever. Can apple provide a real benefit to users/developers with just darwin, or would they have to go further? I'm curious what y'all think.
-earl
no, no, no
you're misunderstanding the GPL. When you GPL program, you continue to reatin full copyright, what you do is allow unrevokable, unlimited redistribution and modification by anyone. The one stipulation is that when you redistribute it to a third party, you must also redristribute the source, at no additional cost. you still own the software, and you can re-release it under another license if you wish, or you can even make changes, and redristribute without source. Only those who you license it to (everyone who doesen't own the copyright to the software) must abide by the licensing terms.
Just like microsoft still owns the software that they license. They could even change the license (specifically, license terms to new licensees) on a whim. But it dosen't make sense that they'd have to pay for a copy right? They don't license it to themselves.