Haha, yeah.. I was trying to demonstrate to my friend how to build gradients over text; a couple years back I made a really cool spheroid, but I couldn't remember how I did it, sadly.
In fact, it's quite the opposite of what you're thinking. A post like that isn't ego driven; it's frustration. Nothing has pissed me off more in life than "trying to find a purpose", ie something to do with my life, since I know I can "do anything if I put my mind to it". The problem isn't capability of our youth. I feel like anyone else; we all can learn to do anything if we choose to, but the question is "how do we choose to?" So instead of teaching us how to choose a task, our schools pack us with so many choices that it's generally impossible to choose.
I chose something because it's what I spent most of my time doing, even if I don't really enjoy it.
Implicit understanding of my philosophies kind of drove your post off into a misunderstanding of what I'm getting at. Not all kids, myself included, know exactly what to do. My high school graduates less than a hundred students a year. Fourty of them leap off into state schools and different colleges. Twenty jump into vocations and tech schools. The other ten of us, the few that never really excelled at any one thing, the few that were completely and totally average in every subject (or in the case of my friends, completely and totally ABOVE average in every single subject) had no clue what to do. So we scatter off into collleges and universities, spending who knows how much on even more education towards even more indecision.
I don't consider myself superior, nor do I find myself below everyone else. I'm just your standard, middle class American with student loans and misunderstandings. Ever seen American Beauty??
Isn't that what they feed everyone in school; "You can do anything if you just put your mind to it"?
In my younger years, I took this to mean "Do everything, because you can". Now that I'm in college, that entire lesson was bunk, and now I'm stuck with a bunch of what I'd consider useless knowledge.
The "Pretender" gene, as I often call it (after the TV series) is something a lot of us are blessed/cursed with. We have the ability to sit down at a computer and code anything, then get up, walk into a garage or workshop, pick up a hammer and build something, then go to a rally and speak about how you can change the world if your party will support you.
The problem with it is futility. Others like me, myself included, find it futile at times to do anything, since we've done everything we're interested in doing. Us general-purpose, disposable task people have to cast ourselves into single purpose, repetitive task people, and that's really hard for us, in college, and in life.
Sadly, I don't see an easy solution. Except I won't be telling my children that "They can do anything". I'll tell them "you can do something. but it's up to you to choose what that something is."
A) I use myself as a predicate; if "I" don't want to use it, and I'm a typical user, then what does that say for other typical users? Exactly.
B) Software engineers are exactly that, engineers. An engineer's job is Design, Supervising Construction, and Redesign once the product is complete. Software engineers are no different than Automotive Engineers except in the medium they work with. Automotive Engineers look at the aspects of everything that goes into a car, not just the chassis, the engine and the steering. Now, that isn't to say they aren't helped by anyone; there are concept artists and fabricators who draw and sculpt and build prototypes, but its the engineer's job to see what works, what doesn't, what is economical, and what is luxury.
Apply this same outlook to software engineering. Look at what's out there. See what's wrong with it, what it does good. Design your application to do all of those things well. Construct your application. Sell your application, then go back and take a look at it. Find what's wrong, what's right, what makes it better than anything else. Design your application to do all of those things well. Sell your application, repeat the loop again.
Modern Engineers have a good idea of how to do most of the steps. But it seems nobody teaches how to look at something that exists, and see what's wrong with it. My software engineering classes as of yet have all been about writing code and solving problems. Very few (only one class period IIRC) have been about looking at which does a job better and why (and it was a logic design class).
No. I think if a code author writes code just to be writing it, he's a code monkey, reinventing the wheel.
A software engineer spends his or her time researching as well as implementing. For most Open Source authors, researching encompasses a Google search coming up blank. Now I'm not saying all of them are. There are a lot of good authors out there, creating things that I could have never dreamed up. But the problem is, these people are good engineers, and often are gulped up into companies like Microsoft for example, which has its margins, which means you do it their way, or its the highway.
If something really sucks, then that's an example of poor design, which means that the engineer did a poor job researching the situation.
Email clients are notoriously hard to get right. They've got to deal with a lot of information coming in, and try to process it in a way that it's useful, hopefully without cluttering the whole world up with useless buttons and scroll bars and chrome. But if you sit down and point out what's important in a mail client (the MAIL perhaps?) and think of how to represent it in a clear, concise way, then you quickly start getting ideas of how to implement it, and you can quickly go through those ideas, choosing which are good, and which are crap. I swear I don't think most of the code developers I know ever go through this step. They just look at something that exists, and clone it. And that's what I'm against.
The problem is that there are two groups arguing here, and often members are interchangable depending on the specific case. Group A wants innovation and less copying, to attract users. Group B wants everything to look like a commercial product to attract users. The problem is that Group B wins out hands down in the intial run, because most people out there want something that looks and feels familiar to them. If corporations are to take up linux on the desktop, you can bet they are going to go with something that looks and acts a whole lot like microsoft. Group A can get it's way when people have adopted the software philosophy, because they are then more open to trying new things that could potentialy be better than what they are used to.
The problem is that Group A and Group B can both be a single group, if we chose to go that route instead.
Nothing is really stopping us from making Commercial-grade software. The dependent libraries are there. The potential for eye candy is there. The implementation code is there. What we need is design.
When you look at any product, absolutely anything out there, you are presented with choice. For example, the difference between a Segway and a Moped. The Moped was designed to do a very specific task, as simple as possible, and to look something like the little brother to a motorcycle. The Segway was designed to do everything a Moped can do, but to give you better flexability with what you want to do with it.
The problem with Group A+B is that it's a lot of work to get both implementation and design working, and nobody wants to do it for free. Anyone can code a copy to something in a given set of time. It takes a set mount of time, plus probably even two times that to draw out and design your interface, figure out its weaknesses and strengths, do comparisons, etc. That's why the Segway still costs four times that of a good Moped.
Gmail is the perfect example of what can be done for the interface of an email client, even if it is only web only. Content groups, filtering, integrating search technologies, etc.
iCal is another good example, even if it doesn't seem it. It's simple enough for anyone to pick up and understand in minutes, and is compatible enough to work with the open source programs out there that do the same thing. Plus it gives you a lot of room and leeway to do what you want to do with it, such as RSS-like calendar feeds, which was all defined in the standard it was written to project.
Calendaring and email systems in my opinion are the worst programs out there in interface, which is why I stay with separate, very down to earth solutions for both. If a calendaring program could do good group management, automatically set up my contact groups based on who emails me, build social nets, etc etc, I'd switch to it in a heart beat. But these carbon clones of Outlook aren't helping me as a software consumer, which means there is virtually no incentive for me to switch to Open Source.
God forbid anyone thing that Open Source authors learn something about design instead of functionality. That's the difference between Software Engineers and Code Monkeys.
The problem as I see it is that the screen shots don't really show how any one product is better than the other; they're all virtually identical, so why not use any of them, which defeats the point of "choice" anyways.
Compatibility is one thing, but design is entirely another. These apps were designed to be carbon copies, not to be Outlook compatible.
It only seems like the normal thing because most Open Source coders look at something that already exists, and try to mirror its functionality. This is a great example of it.
As for the marketplace, the iPod's interface was design genius, and is it's sole link to fame. Stealing the interface of an iPod is stealing the iPod. The same shouldn't be said for software; the interface and the application should be two very seperate tools. That way, you can use whatever interface you like, and nobody complains. Like Linux, for example.
This just shows how sad innovation is in the Open Source market. These are the kinds of systems we should be innovating most because we use them most; clunky interfaces and useless features suck when it comes to task management and email.
I found it almost shocking to see how closely Kontact mirrored Outlook 2k1. While this might be alright for the come across user, it's not likely to actually/attract/ any users.
Honestly, a monkey and an organ grinder could have predicted this. Hardware was one of the boundries for entering the computer market in the past, but now with Cell phones, in-car GPS systems, and everywhere else you see ubiquitious computing machines, you begin to pay less for the hardware (and ironically, the software too!), and more for the support contracts of those platforms, whether they be cellular service, special GPS-like network access, or radio access.
Just worries me because some day in the future, the need for owning a full fledged computer may be unjustifiable..
Intel processors may be teh suck, but they are also teh ub3r cheap (to speak in such terms).
The POWER chip family was very expensive as IBM found out, and the only viable solution was the Power-core derivatives like the 970. PPC may be more efficient (no wasted CISC translation interface), and more powerful per clock, but it also costs a lot more to develop faster and better chips with. As they phase into normal use, it should get cheaper, but by then, we will be calling PPC slow and the Next Best Thing will be faster, more RISCy (if that's even possible), massively parallel but expensive to develop for. That's the general microprocessing life cycle.
In short, the Mac geek didn't even matter. He was only talking about archetecture design, not about Windows vs Mac OS. Windows could have easily been written for PPC (and a port exists) just as easily as Mac OS could be written for x86 (and a port also exists here).
The problem is, if we/don't/ go ahead and make this comparison, then Microsoft has lost on all accords, or has won on all accords, depending on which side of the fence you are on.
If you don't make the comparison to future software, Microsoft can claim anything in the world, as they have been with Longhorn to date. WinFS, Avalon,.Net 2, buzzword after buzzword, but no real evidence of anything. This means that either a) Microsoft's ideas are SO advanced that they're YEARS beyond us, or b) Microsoft has nothing. I know which side I'm on if you're looking at it from this perspective.
So what if 19 months is a long time. _Both_ companies have 19 months, and as I recall, the release cycle of OS X has seemingly hit that miraculous 18 month interval, meaning that when Longhorn actually does come out, so will Mac OS X 10.5.
At this point, we can only compare what exists, and what doesn't. Dashboard and Spotlight exist/now/. Microsoft has a fancy alt-tab skin.
You don't have to buy, on either side of the fence. Windows is on practically every machine out there. You can tour Mac OS X at your nearest Mac Mall/Mac Town/Apple Store. Or you can head over to the nearest K-12 school. Or you can head over to a library (college OR public, both tend to have a fair share of the Mackenworld). Both platforms are everywhere, you just have to look hard to see the Macs (they're typically well hidden/sitting on the employee's desks).
The Cell's max heat production should be almost spot on with the G5, plus seven times that of a typical SPU (which could be fairly accurately estimated by relating the thermal production per transistor for the G5 in the unit with your unknown over the transistor count for the SPUs). This is likely to require liquid cooling once you get into prolonged usage, and it's highly likely the blades IBM looks to build with it will be a Blade-enclosed heatpiping system.
What *needs* to happen for the continued production of laptops in Apple's camp is a low-power equivalent of the G5; do like the Intel engineers did. Look at the archetecture. What there is good, and what is bad. What can you take from the G4 that was good and apply it to the G5. Things like the Velocity Engine could be scraped if it's discovered it's thermal production per performance isn't up to par (as I hypothesize it does, along with SSEn). Any last step you can do to reduce the transistor count, but keep the performance up.
The Cell is clearly built _not_ to be portable. It's built for use inside of things that will stay relatively static, and require high amounts of processing power. Things like High Def LCD/Plasma televisions (Picture in Picture, perhaps?), things like Sony's PS3, things like your fancy new Streaming Media Server, or your super fancy new, multi-channel encoding TiVo.
I'll one up your bid; IBM should either 1) Hold a $$million dollar bounty on Open Source Blade Management software, or 2) Buy RLX and open it to us!
It's more likely that the first will happen, as IBM has had problems enough acquiring other's software and using it for any purpose (take a look at the SCO case). And with their investment so deeply in Linux right now, I'd say this is just a bit over the horizion for Big Blue.
While I get your point, it seems both companies are going for the same goal, just in different ways.
Microsoft is pushing their "procedural synthesis" into extremly parallel processors, to offload work from artists, and still make games look better. Sony is pushing their Cell chip as an extremely parallel processor to offload the work of the game, putting all the weight on artists, and still making the games look better. None of this, however, helps gameplay, which is the reason I don't even own a current-gen console.
Secondly, what would stop Sony, or anyone else for that matter, from making a "proceedural synthesis" system of their own? Really, the whole concept is to unroll the concept we had when we first started making video games ("Oh, computers are too slow, let's pre-compute everything for them" vs "Oh, computers are so fast they can afford to compute the data on the fly"). Instead of running your vector algorithms on the production-side, run it on the client side. The machines are there now.
It's funny how the early paradigms for optimization in computers are now starting to work in reverse.. quality verses cost curve is reversing I guess.
PalmOS is like the Apple of the business. It may not be the cheapest (but often is). It may not be in the lead marketshare-wise (but currently is). But the interface is hella streamlined, and it Just Works (tm). Besides that, it's not too bad to code for, and it's got a firm old of the hardware it's on.
Even so, it wouldn't be all that bad to port PalmOS to the XScale chip, or any other archetecture. I'd be interested in seeing it run on x86 natively (emulators already exist).
I guess you're one of the few that actually like Windows CE or Windows Embedded or whatever they're calling it today; an existing system ported onto a system with ten times the restricted ram, and even more so when you speak of CPU power and battery power. Why not let PalmOS, the operating system designed to fit embedded PDA systems, do the job it was created to do?
As a student, I looked into many different fields before I chose computer science. One of which was digital film making. I spent a summer shadowing a documentary film crew, and I can tell honestly tell you that the amount of work that goes into even the simplest of films is insane.
When we start talking digital animation, the amount of work leaps exponentially. Long hours of modeling, shading, color checks, lighting checks, triangle counts, waiting for renders, etc. It's a tough business.
The "suicidal" part comes in when someone suggests making a feature length film, animated, basically with no money to pay people to come and work for you. You're looking at a group of 10 to 20 dedicated people, spending a great deal of their lives for the next year or two, churning away at scenes, storyboards, models, textures, etc, until finally they come up with something, instead of Pixar's or Dreamwork's thousands of support personel. You're looking at 10 to 20, midrange servers whereas Pixar or Dreamworks has hundreds, possibly thousands of highrange servers in their rendering farms.
Now, will the final product be worth it? Hell yes if it's a good story, looks good, and feels good. Put it in theaters, get a couple million in ticket sales and you've instantly paid for your venture. But the problem is getting even that far. And for that, I would call you suicidal, but I would commmend your work.
Difference between a client, web based application, and a client, client based application. My idea was a program like a Dashboard widget. AJAX is ***only*** web-based and will never have file system access.
This is beyond just using Linux on servers; this is an entire movie made using only open source tools.
Honestly I don't see the point. Who cares if the data is created with Open or Closed source tools, just as long as the resultant data is worth the resources their spending on it, at least that's my opinion about it all. I guess to those in the project, it simply proves the tools are "good enough", and that it can be done, if anyone was suicidal enough to attempt it.
Slashdot Mirror
Haha, yeah.. I was trying to demonstrate to my friend how to build gradients over text; a couple years back I made a really cool spheroid, but I couldn't remember how I did it, sadly.
In fact, it's quite the opposite of what you're thinking. A post like that isn't ego driven; it's frustration. Nothing has pissed me off more in life than "trying to find a purpose", ie something to do with my life, since I know I can "do anything if I put my mind to it". The problem isn't capability of our youth. I feel like anyone else; we all can learn to do anything if we choose to, but the question is "how do we choose to?" So instead of teaching us how to choose a task, our schools pack us with so many choices that it's generally impossible to choose.
I chose something because it's what I spent most of my time doing, even if I don't really enjoy it.
Implicit understanding of my philosophies kind of drove your post off into a misunderstanding of what I'm getting at. Not all kids, myself included, know exactly what to do. My high school graduates less than a hundred students a year. Fourty of them leap off into state schools and different colleges. Twenty jump into vocations and tech schools. The other ten of us, the few that never really excelled at any one thing, the few that were completely and totally average in every subject (or in the case of my friends, completely and totally ABOVE average in every single subject) had no clue what to do. So we scatter off into collleges and universities, spending who knows how much on even more education towards even more indecision.
I don't consider myself superior, nor do I find myself below everyone else. I'm just your standard, middle class American with student loans and misunderstandings. Ever seen American Beauty??
Isn't that what they feed everyone in school; "You can do anything if you just put your mind to it"?
In my younger years, I took this to mean "Do everything, because you can". Now that I'm in college, that entire lesson was bunk, and now I'm stuck with a bunch of what I'd consider useless knowledge.
The "Pretender" gene, as I often call it (after the TV series) is something a lot of us are blessed/cursed with. We have the ability to sit down at a computer and code anything, then get up, walk into a garage or workshop, pick up a hammer and build something, then go to a rally and speak about how you can change the world if your party will support you.
The problem with it is futility. Others like me, myself included, find it futile at times to do anything, since we've done everything we're interested in doing. Us general-purpose, disposable task people have to cast ourselves into single purpose, repetitive task people, and that's really hard for us, in college, and in life.
Sadly, I don't see an easy solution. Except I won't be telling my children that "They can do anything". I'll tell them "you can do something. but it's up to you to choose what that something is."
Yeah, I had a friend that did the same thing.. and I didn't have the heart to say a word.
He's currently working for Apple pulling in a quarter million a year, while I sit here in Engineering school.
A) I use myself as a predicate; if "I" don't want to use it, and I'm a typical user, then what does that say for other typical users? Exactly.
B) Software engineers are exactly that, engineers. An engineer's job is Design, Supervising Construction, and Redesign once the product is complete. Software engineers are no different than Automotive Engineers except in the medium they work with. Automotive Engineers look at the aspects of everything that goes into a car, not just the chassis, the engine and the steering. Now, that isn't to say they aren't helped by anyone; there are concept artists and fabricators who draw and sculpt and build prototypes, but its the engineer's job to see what works, what doesn't, what is economical, and what is luxury.
Apply this same outlook to software engineering. Look at what's out there. See what's wrong with it, what it does good. Design your application to do all of those things well. Construct your application. Sell your application, then go back and take a look at it. Find what's wrong, what's right, what makes it better than anything else. Design your application to do all of those things well. Sell your application, repeat the loop again.
Modern Engineers have a good idea of how to do most of the steps. But it seems nobody teaches how to look at something that exists, and see what's wrong with it. My software engineering classes as of yet have all been about writing code and solving problems. Very few (only one class period IIRC) have been about looking at which does a job better and why (and it was a logic design class).
No. I think if a code author writes code just to be writing it, he's a code monkey, reinventing the wheel.
A software engineer spends his or her time researching as well as implementing. For most Open Source authors, researching encompasses a Google search coming up blank. Now I'm not saying all of them are. There are a lot of good authors out there, creating things that I could have never dreamed up. But the problem is, these people are good engineers, and often are gulped up into companies like Microsoft for example, which has its margins, which means you do it their way, or its the highway.
If something really sucks, then that's an example of poor design, which means that the engineer did a poor job researching the situation.
Email clients are notoriously hard to get right. They've got to deal with a lot of information coming in, and try to process it in a way that it's useful, hopefully without cluttering the whole world up with useless buttons and scroll bars and chrome. But if you sit down and point out what's important in a mail client (the MAIL perhaps?) and think of how to represent it in a clear, concise way, then you quickly start getting ideas of how to implement it, and you can quickly go through those ideas, choosing which are good, and which are crap. I swear I don't think most of the code developers I know ever go through this step. They just look at something that exists, and clone it. And that's what I'm against.
The problem is that there are two groups arguing here, and often members are interchangable depending on the specific case. Group A wants innovation and less copying, to attract users. Group B wants everything to look like a commercial product to attract users. The problem is that Group B wins out hands down in the intial run, because most people out there want something that looks and feels familiar to them. If corporations are to take up linux on the desktop, you can bet they are going to go with something that looks and acts a whole lot like microsoft. Group A can get it's way when people have adopted the software philosophy, because they are then more open to trying new things that could potentialy be better than what they are used to.
The problem is that Group A and Group B can both be a single group, if we chose to go that route instead.
Nothing is really stopping us from making Commercial-grade software. The dependent libraries are there. The potential for eye candy is there. The implementation code is there. What we need is design.
When you look at any product, absolutely anything out there, you are presented with choice. For example, the difference between a Segway and a Moped. The Moped was designed to do a very specific task, as simple as possible, and to look something like the little brother to a motorcycle. The Segway was designed to do everything a Moped can do, but to give you better flexability with what you want to do with it.
The problem with Group A+B is that it's a lot of work to get both implementation and design working, and nobody wants to do it for free. Anyone can code a copy to something in a given set of time. It takes a set mount of time, plus probably even two times that to draw out and design your interface, figure out its weaknesses and strengths, do comparisons, etc. That's why the Segway still costs four times that of a good Moped.
Gmail is the perfect example of what can be done for the interface of an email client, even if it is only web only. Content groups, filtering, integrating search technologies, etc.
iCal is another good example, even if it doesn't seem it. It's simple enough for anyone to pick up and understand in minutes, and is compatible enough to work with the open source programs out there that do the same thing. Plus it gives you a lot of room and leeway to do what you want to do with it, such as RSS-like calendar feeds, which was all defined in the standard it was written to project.
Calendaring and email systems in my opinion are the worst programs out there in interface, which is why I stay with separate, very down to earth solutions for both. If a calendaring program could do good group management, automatically set up my contact groups based on who emails me, build social nets, etc etc, I'd switch to it in a heart beat. But these carbon clones of Outlook aren't helping me as a software consumer, which means there is virtually no incentive for me to switch to Open Source.
God forbid anyone thing that Open Source authors learn something about design instead of functionality. That's the difference between Software Engineers and Code Monkeys.
The problem as I see it is that the screen shots don't really show how any one product is better than the other; they're all virtually identical, so why not use any of them, which defeats the point of "choice" anyways.
Compatibility is one thing, but design is entirely another. These apps were designed to be carbon copies, not to be Outlook compatible.
It only seems like the normal thing because most Open Source coders look at something that already exists, and try to mirror its functionality. This is a great example of it.
As for the marketplace, the iPod's interface was design genius, and is it's sole link to fame. Stealing the interface of an iPod is stealing the iPod. The same shouldn't be said for software; the interface and the application should be two very seperate tools. That way, you can use whatever interface you like, and nobody complains. Like Linux, for example.
This just shows how sad innovation is in the Open Source market. These are the kinds of systems we should be innovating most because we use them most; clunky interfaces and useless features suck when it comes to task management and email.
/attract/ any users.
I found it almost shocking to see how closely Kontact mirrored Outlook 2k1. While this might be alright for the come across user, it's not likely to actually
I'll stick with Gmail and iCal for now.
Honestly, a monkey and an organ grinder could have predicted this. Hardware was one of the boundries for entering the computer market in the past, but now with Cell phones, in-car GPS systems, and everywhere else you see ubiquitious computing machines, you begin to pay less for the hardware (and ironically, the software too!), and more for the support contracts of those platforms, whether they be cellular service, special GPS-like network access, or radio access.
Just worries me because some day in the future, the need for owning a full fledged computer may be unjustifiable..
Intel processors may be teh suck, but they are also teh ub3r cheap (to speak in such terms).
The POWER chip family was very expensive as IBM found out, and the only viable solution was the Power-core derivatives like the 970. PPC may be more efficient (no wasted CISC translation interface), and more powerful per clock, but it also costs a lot more to develop faster and better chips with. As they phase into normal use, it should get cheaper, but by then, we will be calling PPC slow and the Next Best Thing will be faster, more RISCy (if that's even possible), massively parallel but expensive to develop for. That's the general microprocessing life cycle.
In short, the Mac geek didn't even matter. He was only talking about archetecture design, not about Windows vs Mac OS. Windows could have easily been written for PPC (and a port exists) just as easily as Mac OS could be written for x86 (and a port also exists here).
The problem is, if we /don't/ go ahead and make this comparison, then Microsoft has lost on all accords, or has won on all accords, depending on which side of the fence you are on.
.Net 2, buzzword after buzzword, but no real evidence of anything. This means that either a) Microsoft's ideas are SO advanced that they're YEARS beyond us, or b) Microsoft has nothing. I know which side I'm on if you're looking at it from this perspective.
/now/. Microsoft has a fancy alt-tab skin.
If you don't make the comparison to future software, Microsoft can claim anything in the world, as they have been with Longhorn to date. WinFS, Avalon,
So what if 19 months is a long time. _Both_ companies have 19 months, and as I recall, the release cycle of OS X has seemingly hit that miraculous 18 month interval, meaning that when Longhorn actually does come out, so will Mac OS X 10.5.
At this point, we can only compare what exists, and what doesn't. Dashboard and Spotlight exist
You don't have to buy, on either side of the fence. Windows is on practically every machine out there. You can tour Mac OS X at your nearest Mac Mall/Mac Town/Apple Store. Or you can head over to the nearest K-12 school. Or you can head over to a library (college OR public, both tend to have a fair share of the Mackenworld). Both platforms are everywhere, you just have to look hard to see the Macs (they're typically well hidden/sitting on the employee's desks).
Typically, I like to keep my winky hidden. When good days come, he still likes to stay hidden, just not in my pants...
For me, its more like:
It's time to make that tin foil body armor I was promised by my parents I would never need.. whodve thunk?
Sorry, but that's extremely not likely to happen.
The Cell's max heat production should be almost spot on with the G5, plus seven times that of a typical SPU (which could be fairly accurately estimated by relating the thermal production per transistor for the G5 in the unit with your unknown over the transistor count for the SPUs). This is likely to require liquid cooling once you get into prolonged usage, and it's highly likely the blades IBM looks to build with it will be a Blade-enclosed heatpiping system.
What *needs* to happen for the continued production of laptops in Apple's camp is a low-power equivalent of the G5; do like the Intel engineers did. Look at the archetecture. What there is good, and what is bad. What can you take from the G4 that was good and apply it to the G5. Things like the Velocity Engine could be scraped if it's discovered it's thermal production per performance isn't up to par (as I hypothesize it does, along with SSEn). Any last step you can do to reduce the transistor count, but keep the performance up.
The Cell is clearly built _not_ to be portable. It's built for use inside of things that will stay relatively static, and require high amounts of processing power. Things like High Def LCD/Plasma televisions (Picture in Picture, perhaps?), things like Sony's PS3, things like your fancy new Streaming Media Server, or your super fancy new, multi-channel encoding TiVo.
I'll one up your bid; IBM should either 1) Hold a $$million dollar bounty on Open Source Blade Management software, or 2) Buy RLX and open it to us!
It's more likely that the first will happen, as IBM has had problems enough acquiring other's software and using it for any purpose (take a look at the SCO case). And with their investment so deeply in Linux right now, I'd say this is just a bit over the horizion for Big Blue.
While I get your point, it seems both companies are going for the same goal, just in different ways.
Microsoft is pushing their "procedural synthesis" into extremly parallel processors, to offload work from artists, and still make games look better. Sony is pushing their Cell chip as an extremely parallel processor to offload the work of the game, putting all the weight on artists, and still making the games look better. None of this, however, helps gameplay, which is the reason I don't even own a current-gen console.
Secondly, what would stop Sony, or anyone else for that matter, from making a "proceedural synthesis" system of their own? Really, the whole concept is to unroll the concept we had when we first started making video games ("Oh, computers are too slow, let's pre-compute everything for them" vs "Oh, computers are so fast they can afford to compute the data on the fly"). Instead of running your vector algorithms on the production-side, run it on the client side. The machines are there now.
It's funny how the early paradigms for optimization in computers are now starting to work in reverse.. quality verses cost curve is reversing I guess.
PalmOS is like the Apple of the business. It may not be the cheapest (but often is). It may not be in the lead marketshare-wise (but currently is). But the interface is hella streamlined, and it Just Works (tm). Besides that, it's not too bad to code for, and it's got a firm old of the hardware it's on.
Even so, it wouldn't be all that bad to port PalmOS to the XScale chip, or any other archetecture. I'd be interested in seeing it run on x86 natively (emulators already exist).
I guess you're one of the few that actually like Windows CE or Windows Embedded or whatever they're calling it today; an existing system ported onto a system with ten times the restricted ram, and even more so when you speak of CPU power and battery power. Why not let PalmOS, the operating system designed to fit embedded PDA systems, do the job it was created to do?
Read about this at 8:40am yesterday morning.. Slashdot, you're almost a day behind on this one.
Besides this being cool news (no more "Palm!!1One"), I wonder if it'll really end up affecting anything.
As a student, I looked into many different fields before I chose computer science. One of which was digital film making. I spent a summer shadowing a documentary film crew, and I can tell honestly tell you that the amount of work that goes into even the simplest of films is insane.
When we start talking digital animation, the amount of work leaps exponentially. Long hours of modeling, shading, color checks, lighting checks, triangle counts, waiting for renders, etc. It's a tough business.
The "suicidal" part comes in when someone suggests making a feature length film, animated, basically with no money to pay people to come and work for you. You're looking at a group of 10 to 20 dedicated people, spending a great deal of their lives for the next year or two, churning away at scenes, storyboards, models, textures, etc, until finally they come up with something, instead of Pixar's or Dreamwork's thousands of support personel. You're looking at 10 to 20, midrange servers whereas Pixar or Dreamworks has hundreds, possibly thousands of highrange servers in their rendering farms.
Now, will the final product be worth it? Hell yes if it's a good story, looks good, and feels good. Put it in theaters, get a couple million in ticket sales and you've instantly paid for your venture. But the problem is getting even that far. And for that, I would call you suicidal, but I would commmend your work.
Difference between a client, web based application, and a client, client based application. My idea was a program like a Dashboard widget. AJAX is ***only*** web-based and will never have file system access.
This is beyond just using Linux on servers; this is an entire movie made using only open source tools.
Honestly I don't see the point. Who cares if the data is created with Open or Closed source tools, just as long as the resultant data is worth the resources their spending on it, at least that's my opinion about it all. I guess to those in the project, it simply proves the tools are "good enough", and that it can be done, if anyone was suicidal enough to attempt it.