By adopting the same strategy as d.net - taking volunteers and working closely with them. I don't think a lack of source code prevents people from looking at the data that the seti@home client sends, and sending similar data back. Why would anyone with the ability want to be such a pain?
As for the optimization, I thought 2.0 was going to fix it. They've been around for a long time, and I didn't see anything on their site about taking volunteers to do the work that they've neglected to do.
Is this the fault of laziness/not-caring-about-appearances on the part of geeks, or just a lack of services from the OS?
If someone came up with a "universal interface language" that was clearly defined and implementation-independent, so that a standard dialog-box descriptor could be used as a group of command-line elements or a group of graphic elements (for example), programmers could quickly design their UI better.
Moreover, if this descriptor language was understood directly by the shell, the interface sources could be developed as a co-project by inclined users. Make a graphic interface editor (a la ResEdit on the Mac) and you've got instant skins in every application.
Seti@Home is crappy closed-source code, with very badly optimized clients. d.net is (partly, and in ideals) open-source, and clients are optimized by volunteers.
In terms of getting the most out of the computing power of those who dovolunteer, I'd definitely take d.net over Seti@Home.
Motion sickness is caused by having your middle ear bumped around too much without other cues about movement. I can't imagine this VR technology is not for people who get carsick easily.
I don't want to play any videogames that leave me feeling like I just took a long car ride. I just get this kind of soggy sleepy feeling. Mebbe more exciting motions would be different, but I just doubt the technology.
The American audience won't put up with this. Facing competition from other sources of music, they should be making CD's look as attractive as possible - "don't lose all your music when your drive crashes!" - "infinite shelf life, storage is free!" - etc. Or, add features. How much could you do with 50K of magnetic tape around the inside/rim of a CD? Keep programs, track titles, whatever. You've already got a laser there, why not make it magneto-optical to prevent data loss? But they're not. They're making CD's look unattractive, and themselves to look hostile, and it'll scare away customers. Mebbe not everybody rips their CD's, but a lot of people like to play them in their computer anyway. Their behavior is typical of companies that know they're getting behind tech & the times. Word'll get out.
Oh no, not "Razorblade Romance!" I wanted that one! Well I guess they're just experimenting w/ unpopular music first...
Just downloaded the new 2.0 client, to see if it adds any features/optimization. It sucks.
Just to give you a grasp of this incredible suckiness: viewed w/ a disassembler, the code appears to be cross-compiled from x86. And it wasn't any incredible TransMeta cross-compiler technology, either. How can I tell? Variables are put on the stack, their registers used for two instructions, then taken back off. They get taken out of the stack and put back for a single add operation, or seemingly just to move somewhere else on the stack. This is fine on a PIII w/ the stack being an alias for the internal registers. Not OK on a RISC processor of any type - I don't want to know how much Alpha power is giong to waste here, too.
Example: the function GLIB_PANEL::FillSolidRect, 41 instructions long, contains 4 instructions that are not load/store operations (ex. the pro/epilogue), and its sole purpose is to call another function. All 4 operations are 'add r0, r3, r0'. And why write this function at all when there's the Mac API?
Apparently they didn't know how to compile Mac programs either, or they wouldn't have exported all the debugging info I used to do this analysis... so much for discouraging hackers! They also included OS libraries directly in the downloadable binary! I'm not sure exactly why or how, tho... Also, it's distributed as a control panel, which means that all that fatty code is kept in RAM at all times!
This isn't a way to put your spare CPU cycles to use, this is robbery!
Non-x86 users, get rc5. And multiply the benchmarks at their site by about 5.
v1.0.6 for Mac SUCKS. A low-level debugger shows they implement their own drawing (rectangle, line,...) code for the screensaver part, and that uses a significant amount of processor time. Not to mention being ugly. What's more, the processor optimizations (when it is doing math) are dismal.
I dunno if they're planning to come out w/ a v2.0 GUI for Win/Mac, but until they clean their code or OSS it, I'm sticking w/ dnetc. All text, Altivec optimized, 3.8MKeys/sec. They know where the computing power is:v).
Somehow I don't see this building Transmeta into an instant giant. They will need time to grow, and in the meantime Motorola (for example) can implement at least some of these ideas into the 683xx/Coldfire series (the processor used in PalmPilots). Staying with a dedicated ISA (68K) that's easy to do in a small number of transistors (say 68,000) in the 1st place could probably reduce the wattage/transistor count while keeping another variable - manufacturing process cost - down. Or mebbe the ARM or PPC embedded lines would be better for high-power/low power (gag) apps. In this way, Transmeta, being independent, is the worst company to come out with a Crusoe - they have to keep their processors needlessly flexible. Not all of their tech is patentable - VLIW has been floating around for a while. Do we really want to depend on a company that depends on patent enforcement?
Cool that this comes from the same people who brought us RISC, tho. They said thatwasn't practical...
Bloat causes unnoticed bugs. I'll bet the few (kilo)bytes of machine language in a Crusoe are probably pretty easy to debug, and any problems would show up. I think the processor engineers at TM will be a little more careful than the average MS programmer, too.
And don't forget the original Pentium division error - not having a net to fall back on doesn't make engineers behave better, either.
I also spent a lot of elementary school time in "studies," and it earned me a lot of picking-on. When I went into the Big Middle School in 5th grade, I rebelled against my former image and started hanging out with people. This was important in my life. I'm glad I took the vacation from worrying about the future - that's what being a kid is all about.
But, soon I realized that the years are draining away. Now, I've picked my reference manuals back up and I'm programming again. I want to be a very good programmer, w/ all that $$, or mebbe even not. Computer programming is a very good skill to have.
From what you say, it looks like you've been living your life day-to-day, and now sometimes you feel flashes of regret about this. Wanting to change is the first step. Feeling the want to change at a time that there's a book there, or the preliminary notes for a project, makes actually sitting down easy.
Give in to your fears. Pick up a good programming book, or start a project, from where you left off. At 14, there's really nothing lost.
And the child-prodigy-lost thing is nothing. Child prodigies get screwed over in the long-term. No one wants to deal w/ an employee with that screwed up of a view of life.
P.S. I'm 15 now. Since I got re-interested in programming, a little over a year ago, I've learned just about every detail about how processors work, to program in asm, and much of how OS's operate. And I haven't exactly been dedicating/reserving time. Someone with a will to teach themselves can learn a lot faster than some tutored "prodigy" bastard, IMHO.
People! It's simple. Be allows us to download it from them, while retaining all legal rights. The word for this is
FREEWARE!
And the reason they're doing this is probably not because they want to get on the bandwagon, but 'cuz years ago, before the final release, it was available for free. Years of people begging them to continue this method of distribution have paid off, that's all.
The goal is to establish a user base, because people are most likely to try an OS w/ a simple installer & no cost. In political terms, this means that they realize they don't have a good contender for the home market and they want to win over more non-embedded users.
Free as in beer, not as in speech. BeOS is still proprietary, it's just freeware. And as for "proprietary apples", Be originally made its own proprietary machines, then ported their OS to Apple's hardware.
Slashdot Mirror
Eyyy, I shouldn't be such a troll. But I was really pissed when I disassembled that client. There should be a warning on it. See the link.
Where is my mind?
By adopting the same strategy as d.net - taking volunteers and working closely with them.
I don't think a lack of source code prevents people from looking at the data that the seti@home client sends, and sending similar data back. Why would anyone with the ability want to be such a pain?
As for the optimization, I thought 2.0 was going to fix it. They've been around for a long time, and I didn't see anything on their site about taking volunteers to do the work that they've neglected to do.
Where is my mind?
Is this the fault of laziness/not-caring-about-appearances on the part of geeks, or just a lack of services from the OS?
If someone came up with a "universal interface language" that was clearly defined and implementation-independent, so that a standard dialog-box descriptor could be used as a group of command-line elements or a group of graphic elements (for example), programmers could quickly design their UI better.
Moreover, if this descriptor language was understood directly by the shell, the interface sources could be developed as a co-project by inclined users. Make a graphic interface editor (a la ResEdit on the Mac) and you've got instant skins in every application.
Where is my mind?
Seti@Home is crappy closed-source code, with very badly optimized clients.
d.net is (partly, and in ideals) open-source, and clients are optimized by volunteers.
In terms of getting the most out of the computing power of those who dovolunteer, I'd definitely take d.net over Seti@Home.
Where is my mind?
Motion sickness is caused by having your middle ear bumped around too much without other cues about movement. I can't imagine this VR technology is not for people who get carsick easily.
I don't want to play any videogames that leave me feeling like I just took a long car ride. I just get this kind of soggy sleepy feeling. Mebbe more exciting motions would be different, but I just doubt the technology.
Where is my mind?
Slashdot: Science: Marriot to Open Resort on Cruithne
Great, the way he said "the earth now has a second moon. Its name is Cruithne. Luv dat.
Where is my mind?
The American audience won't put up with this. Facing competition from other sources of music, they should be making CD's look as attractive as possible - "don't lose all your music when your drive crashes!" - "infinite shelf life, storage is free!" - etc.
Or, add features. How much could you do with 50K of magnetic tape around the inside/rim of a CD? Keep programs, track titles, whatever. You've already got a laser there, why not make it magneto-optical to prevent data loss?
But they're not. They're making CD's look unattractive, and themselves to look hostile, and it'll scare away customers. Mebbe not everybody rips their CD's, but a lot of people like to play them in their computer anyway. Their behavior is typical of companies that know they're getting behind tech & the times. Word'll get out.
Oh no, not "Razorblade Romance!" I wanted that one! Well I guess they're just experimenting w/ unpopular music first...
Where is my mind?
But it's already archived! For the first time I can think of, there is NO WAY TO POST ON TACOHELL right now!
:v) .
I guess now Malda's feeling the 14-day rule
Where is my mind?
This story was already on TacoHell.
Where is my mind?
isn't a sneaky virus, it's a sneaky tech help guy. In Linux, hired help can recompile/rewrite anything, no?
Where is my mind?
Hey, I replied to your post before you made it! To put this trick to a useful task, see my "homepage."
And please don't let #150 be another AC!
Where is my mind?
Go to this page.
Where is my mind?
Hey, it did work! You just have to look at it the right way...
Where is my mind?
Noo, that didn't work. Sorry. But go to the page anyway. That did work, and it's muy interesante.
Where is my mind?
Hello, caller 100! Notice that I replied to your message before you posted it. To see how this happened and more magic, see this page.
Where is my mind?
Just downloaded the new 2.0 client, to see if it adds any features/optimization. It sucks.
Just to give you a grasp of this incredible suckiness: viewed w/ a disassembler, the code appears to be cross-compiled from x86. And it wasn't any incredible TransMeta cross-compiler technology, either. How can I tell? Variables are put on the stack, their registers used for two instructions, then taken back off. They get taken out of the stack and put back for a single add operation, or seemingly just to move somewhere else on the stack. This is fine on a PIII w/ the stack being an alias for the internal registers. Not OK on a RISC processor of any type - I don't want to know how much Alpha power is giong to waste here, too.
Example: the function GLIB_PANEL::FillSolidRect, 41 instructions long, contains 4 instructions that are not load/store operations (ex. the pro/epilogue), and its sole purpose is to call another function. All 4 operations are 'add r0, r3, r0'. And why write this function at all when there's the Mac API?
Apparently they didn't know how to compile Mac programs either, or they wouldn't have exported all the debugging info I used to do this analysis... so much for discouraging hackers! They also included OS libraries directly in the downloadable binary! I'm not sure exactly why or how, tho... Also, it's distributed as a control panel, which means that all that fatty code is kept in RAM at all times!
This isn't a way to put your spare CPU cycles to use, this is robbery!
Non-x86 users, get rc5. And multiply the benchmarks at their site by about 5.
v1.0.6 for Mac SUCKS. A low-level debugger shows they implement their own drawing (rectangle, line, ...) code for the screensaver part, and that uses a significant amount of processor time. Not to mention being ugly.
:v) .
What's more, the processor optimizations (when it is doing math) are dismal.
I dunno if they're planning to come out w/ a v2.0 GUI for Win/Mac, but until they clean their code or OSS it, I'm sticking w/ dnetc. All text, Altivec optimized, 3.8MKeys/sec. They know where the computing power is
Somehow I don't see this building Transmeta into an instant giant. They will need time to grow, and in the meantime Motorola (for example) can implement at least some of these ideas into the 683xx/Coldfire series (the processor used in PalmPilots). Staying with a dedicated ISA (68K) that's easy to do in a small number of transistors (say 68,000) in the 1st place could probably reduce the wattage/transistor count while keeping another variable - manufacturing process cost - down.
Or mebbe the ARM or PPC embedded lines would be better for high-power/low power (gag) apps.
In this way, Transmeta, being independent, is the worst company to come out with a Crusoe - they have to keep their processors needlessly flexible.
Not all of their tech is patentable - VLIW has been floating around for a while. Do we really want to depend on a company that depends on patent enforcement?
Cool that this comes from the same people who brought us RISC, tho. They said thatwasn't practical...
Bloat causes unnoticed bugs. I'll bet the few (kilo)bytes of machine language in a Crusoe are probably pretty easy to debug, and any problems would show up. I think the processor engineers at TM will be a little more careful than the average MS programmer, too.
And don't forget the original Pentium division error - not having a net to fall back on doesn't make engineers behave better, either.
Hey.
I also spent a lot of elementary school time in "studies," and it earned me a lot of picking-on. When I went into the Big Middle School in 5th grade, I rebelled against my former image and started hanging out with people. This was important in my life. I'm glad I took the vacation from worrying about the future - that's what being a kid is all about.
But, soon I realized that the years are draining away. Now, I've picked my reference manuals back up and I'm programming again. I want to be a very good programmer, w/ all that $$, or mebbe even not. Computer programming is a very good skill to have.
From what you say, it looks like you've been living your life day-to-day, and now sometimes you feel flashes of regret about this. Wanting to change is the first step. Feeling the want to change at a time that there's a book there, or the preliminary notes for a project, makes actually sitting down easy.
Give in to your fears. Pick up a good programming book, or start a project, from where you left off. At 14, there's really nothing lost.
And the child-prodigy-lost thing is nothing. Child prodigies get screwed over in the long-term. No one wants to deal w/ an employee with that screwed up of a view of life.
P.S. I'm 15 now. Since I got re-interested in programming, a little over a year ago, I've learned just about every detail about how processors work, to program in asm, and much of how OS's operate. And I haven't exactly been dedicating/reserving time. Someone with a will to teach themselves can learn a lot faster than some tutored "prodigy" bastard, IMHO.
People! It's simple. Be allows us to download it from them, while retaining all legal rights. The word for this is
FREEWARE!
And the reason they're doing this is probably not because they want to get on the bandwagon, but 'cuz years ago, before the final release, it was available for free. Years of people begging them to continue this method of distribution have paid off, that's all.
The goal is to establish a user base, because people are most likely to try an OS w/ a simple installer & no cost.
In political terms, this means that they realize they don't have a good contender for the home market and they want to win over more non-embedded users.
Freeware.
Free as in beer, not as in speech. BeOS is still proprietary, it's just freeware.
And as for "proprietary apples", Be originally made its own proprietary machines, then ported their OS to Apple's hardware.
And they can't even make a toaster oven that doesn't get impossibly hot on the outside. Please!
Barcode! Now that's what I call a short attention span.
Just don't get it dirty...