Rubbish, both are similar, and similarly daft as the eye-candy in Vista.
As the GP points out Apple's eye candy is there for functional reasons. Expose zooms out and lets you find the window you're after quickly. It looks great but that's a side effect to being there for a purpose. It's a whole different element to pointless but pretty stuff like translucent window decorations.
Fixed that for you. But in all seriousness my Uni had a shiny new campus mainly for CS which sounded great until I got there and realised there weren't exactly many women there - the ratio was about 2-1 on campus and on the CS course much worse.
Agreed that the tax shelter isn't illegal and a company of their size is likely to have made sure they are only avoiding tax legitimately.
Perhaps though, US courts won't look as favourably on a company that avoids paying US taxes. In theory it shouldn't matter, but in practice I'm sure it could affect the court's view.
The Linux open-source model is fundamentally open and this sort of thing is a consequence. A group of interested parties have got together to discuss the problems getting Linux adopted in an area they are interested in. Hopefully they will decide what they can improve and go away and do it. With companies like IBM involved, there isn't great need for the community to implement the stuff.
They aren't breaking Linux on the desktop - just improving it on big-iron servers. There's no need for it to be 'representative'. It's quite valid for a few companies to hold a closed meeting and do what they want without outside interruption.
The source code will make its way into the world and if the key players who weren't invited / represented think it's doing something useful it will get further modified and brought into distros like Ubuntu.
Recent black immigrants to the USA and their children are far more successful
Here in the UK the same is very much true with people of Indian descent. On the slightly dubious basis of accents, the last two doctors I have seen have been recent immigrants as are some of the software engineers I work with, while typically the Indian people working at the local supermarket speak in a broad Lancashire accent, suggesting their families have been here at least a couple of generations.
It doesn't prove there is no discrimination, just that a lot of recent immigrants have the training and skills to overcome it. This is mainly due to the points based immigration systems and quotas now used in so many countries.
If you assume the 'average' American is European descent and then look at the success of recent generations of European immigrants you'll find exactly the same pattern because you have to have a good degree - possibly even a PhD, just to get in.
I'd at least do both of those - it's a bit of a hassle to be handed a web address, go home, switch on the PC, type in the address, wait for the files to download etc.
For a quick listen to see if you like the music, the CD is more convenient.
MySQL is a database that is very well suited to running a web application. That's why. Many web apps are generally going to have many more reads than writes. Additionally, the writes tend to be pretty simple and not require complex locking mechanisms.
E.g. my photos website uses gallery2 which works easiest on the LAMP stack. The main database queries are simple - what albums are there, does the user have permission to see them, what photos are in this album etc. The updates are similarly easy - add a new album, add a new photo, add a new user. For this, MySQL is perfect, simple and fast.
However once you start writing more complex systems with tougher constraints things become a lot harder. Typically, you have a record in a state that needs notifying to another system. You need to read that record, send a message to the other system and then modify that record to say the message is sent and the other system then has control over it. For this you need the various locking mechanisms and guarantees from the database or the two systems get out of line. MySQL just isn't the database to use for that sort of application.
Leaving aside the CPU / GPU power, it's not exactly comfortable playing games with a trackpad and tiny cramped keyboard. Also, powerful graphics takes a lot of power and results in lower battery life. Even in many high-end laptops, that's a compromise many people just don't want to take.
I've got a big 17" laptop with a decent graphics card - it has plenty of power for games, but I'd need to cart an external keyboard and mouse / other peripheral around with me to actually play games. I've tried playing strategy games on it while on the road but gave up pretty quick.
We aren't just seeing the death of PC gaming, but the death of the desktop PC. Laptop sales definitely overtook desktops on total value a while back - not sure if it's true on number of units shipped just yet. There just aren't all that many new desktop PCs being sold compared to a while back. People are deciding laptops are good enough and have so many advantages.
One area that they aren't good enough though is gaming. Hence more and more people who would like to game on a PC are having to consider buying a whole new PC to play games. I don't believe you need a $1200 graphics card to play new PC games, but I suspect you need at least a $1200 PC.
I think that quote is an exaggeration, but the general point is still true.
A lot of people now have a 5 year old PC that will do fine for everything except gaming - in the past PCs always could do with being faster and I think for many people that time is now gone. Many companies operated 3 year redundancy cycles for the office PCs, that have now stretched out to 5 years and longer. Where I work, people started opting to upgrade the RAM and occasionally the CPUs and we have a number of P3 era machines that people (software developers even) are perfectly happy with.
You'll have to spend $200 on a graphics card to get good results with a lot of games, and you'll probably go through at least 2 generations of cards in the time of one console generation. If the rest of your PC isn't up to scratch you might have to buy a whole new PC for gaming. Either way you're looking at similar prices to just buying a standalone console and being rid of the hassles of Windows.
For the interface, I disagree very strongly! I'm new to console gaming but I already prefer the PS3 controller for most games.
I think FPSs are much better with keyboard and mouse. I was never too sure how the Xbox manages to have so many popular titles in this genre when you can't use a keyboard and mouse. The same applies to strategy games and I still use my PC for these.
However, for just about every other genre I've played, a console-style controller is much better. Having standard buttons, analog directional pads and analog triggers all right under your fingers is just fantastic and brings a lot more depth and skill into control of most games. The motion sensing on the PS3 controller isn't really good enough for fine control but it is put to great use in Folklore and Tony Hawks.
I would guess the interface is a prime reason for a lot of people abandoning their PCs for consoles. Sure you can add specialist controllers to your PC, but a lot of games won't be designed for them.
The problem is the base level of hardware you must aim for on a PC is substantially lower than the power of a console. If you want to sell plenty of units, you'll probably need your game to run on Radeon 9000 class hardware. A Radeon 9000 is about 7 years old, but there are a lot of modern PCs with integrated graphics that would still struggle to match that performance.
So you start off with a game that has to run on a system like that, and add in the extra touches for people with the hardware to handle it. The result of that is your game never looks as good as you'd like because the niceties are bolted on optional extras, and the base of your game has to run on something with 10% of the performance.
My PC has a GF 8600GT which is pretty similar to PS3 power. Some games can match my PS3 on graphical quality, but they are typically big-budget ones where the developers can properly program multiple code paths for older and newer hardware.
Vista really is terrible though. I was forced by company policy to get Vista on my new PC, and it's just full of bugs. Some of them pretty major. I've just done the SP1 upgrade this morning - yet to see if this is overall better or worse. They've still not fixed the Explorer bug where it can't read any CDs/DVDs, and VMWare now takes about 10 minutes to start instead of 10 seconds. I can now finally use Remote Desktop from an XP machine though.
Not many people thinking of switching to Vista have first hand experience. Us techie geeks might have a quick play with someone else's computer when we see a new thing on it, but for most people it's not interesting. Their opinion is always based on second hand reports from other people.
In my own experience, almost every Vista user has reported headaches of one kind or another.
Blu-Ray is not Sony's format. They are just a contributor and do not have that level of control to stop MS adding a Blu-Ray drive for the Xbox 360 or anything else. Microsoft were never happy about the Java based interactive features on Blu-Ray and supported HD-DVD mainly because they managed to stuff their sort-of proprietary HDi technology into that format.
More importantly, you are looking at the situation in reverse. Although I think having Blu-Ray is an advantage for PS3, the hardware cost increase is a disadvantage. To a greater degree, the PS3 is an advantage for Blu-Ray. Looking at standalone unit sales, PS3 won the format war for Blu-Ray single handed. Sony will earn some royalties from Blu-Ray. From this point of view, I think Sony would be quite happy for cheap Blu-Ray addon drives for the Xbox. The PS3 hardware can only be considered profitable because it's designed to earn money from sales of games and movies over the next decade. The hardware itself will never be that profitable because moving more units brings in more money down the line.
1) It scales down a bit at least. I'm pretty they were marketing it last year for projector bulbs at around 150W. Not sure whether it scales further down than that.
2) 6000K is very close to sunlight so yeah it's a nice warm sunny light - should in theory be nicer than incandescent light anyway.
3) No - it's a noble gas (unreactive) and naturally present in the atmosphere, making up nearly 1% of it in fact.
4 and 5) Dunno. I was just searching for the projector bulb version and couldn't find any actually for sale, which given that it was announced half a year ago isn't great going:(
No, it does reach that heat, but in a very small area and doesn't radiate too much heat. Normal incandescents reach about 3000 C and don't fry everyone because only the filament is at that temperature - the glass bulb is hot enough to burn your fingers but nowhere near the temperature of the filament.
Somewhat paradoxically, at a higher temperature, more radiation is of the visible light spectrum and less is waste heat. So these lights will radiate less heat because they are hotter.
I believe you are partly right. I do work with multi-process applications, but I think part of the reason these are simpler is that you really have to sit there and decide on data to be shared and explicitly create shared memory segments for this data and then attach each new process to those segments (in C at least).
When I have used multi-threaded programming (in Java), it is very tempting, when lots of extra data is shared to just say 'OK I'll use this variable just as a one off to get this done'. Usually in the long run that's a very bad choice because the data was not designed to be shared.
Actually designing ALL of the data that can be shared is an important part of getting the design right and avoiding a lot of the little bugs that crop up. From this point of view I really prefer multi-process work than multi-threading as it is more natural to do things 'right'.
I think many or even most applications can make use of parallel programming with relatively little real effort. Locks and other mechanisms are tricky to get your head around - the real trick is coming up with a design that doesn't need them, or at least minimises that need.
The systems I write store records in a database and minimising locking issues is really about thinking through the broad states that different tasks will deal with. So anything in state A is managed by task X and all other areas of the system should ignore state A (or at least, not change things out of this state).
This results in a much tidier system and much smaller and more readable code. The system I am currently working on has only about 5 background processes that will do most of the work, and the target hardware has 8 cores, so this leaves 3 for other bits and pieces but mainly for dealing with user interaction. If I need to aim at more cores, I know I can easily split the main 5 into about 12 processes without even having to think about locking.
Those calculators compare money equivalency based on inflation figures only - the change in the value of 'money' over time. The bit that is really missing is earnings growth i.e. actual changes in productivity.
For instance if a cheap pair of shoes cost $10 in 1917 and now costs $30, and this sort of measurement was true over a lot of goods, it is true to say $10 then is the equivalent of $30 now.
However this says nothing about how affordable things are. As you say, if the average salary in 1917 was around $700 / year, a pair of shoes is almost a weeks wages. Nowadays it's a couple of hours wages.
Cars are a tricky example because lots of people are happy to spend on them. The price doesn't really drop as much because we instead demand better cars. The cheapest new cars around today are a small fraction of yearly earnings and still bigger, faster, more efficient, more comfortable, safer and more reliable than any car from 1917.
Exactly my opinion also. We are just reaching the point when a properly useful ultra-portable can be had at a decent price. I looked longingly at the Sony TZ series but couldn't justify spending £2000. I've been seriously considering getting the old EEE, but in the end the screen resolution was a bit too low. The new one is just enough and I will also be buying one. I don't care if Walmart sell a similarly priced full size and full featured laptop - I've got one of those and that's not what I am after.
When I sold my house the estate agent website had an SWF that downloaded the images and made it tricky to save. I ran Firefox's tamper data add on to see where it was getting them and found a nice image server with the full high-res wide-angle photos their photographer had taken.
When I bought my last house, it came with a stack of old documents dating back to its building in 1885. It was quite interesting seeing the different styles of legalese that were prevalent at various times. A significant number of the documents had absolutely no punctuation, because even adding pauses into a sentence can change its meaning e.g. Lionel Hutz' advert on the Simpsons saying
Works on commission No money down
gets corrected to 'Works on commission? No, money down!' That's clearly a joke but if someone does similar with the deeds of a house it's not quite as funny. The laws of a country are considerably more important again.
I think that's a question you could spend more than 2 years thinking about and not get an answer. How do you know when you're overcomplicating an issue, and when you're just doing something complex?
Developers always horribly underestimate the difficulty of a task. Every solution starts clean and simple, and it's incredibly hard to keep the finished product looking anything similar to the one which did 95% of the job.
You always look at a nasty solution and think it should be much simpler done this other way. Half the time you're right, but half the time you add in all the little niggles to your overall work of art and you're back at the starting point with an ugly solution that works.
How do you figure out in advance which half you're heading for?
Slashdot Mirror
Rubbish, both are similar, and similarly daft as the eye-candy in Vista.
As the GP points out Apple's eye candy is there for functional reasons. Expose zooms out and lets you find the window you're after quickly. It looks great but that's a side effect to being there for a purpose. It's a whole different element to pointless but pretty stuff like translucent window decorations.
The women will be at a liberal arts college.
Fixed that for you. But in all seriousness my Uni had a shiny new campus mainly for CS which sounded great until I got there and realised there weren't exactly many women there - the ratio was about 2-1 on campus and on the CS course much worse.
Agreed that the tax shelter isn't illegal and a company of their size is likely to have made sure they are only avoiding tax legitimately.
Perhaps though, US courts won't look as favourably on a company that avoids paying US taxes. In theory it shouldn't matter, but in practice I'm sure it could affect the court's view.
The Linux open-source model is fundamentally open and this sort of thing is a consequence. A group of interested parties have got together to discuss the problems getting Linux adopted in an area they are interested in. Hopefully they will decide what they can improve and go away and do it. With companies like IBM involved, there isn't great need for the community to implement the stuff. They aren't breaking Linux on the desktop - just improving it on big-iron servers. There's no need for it to be 'representative'. It's quite valid for a few companies to hold a closed meeting and do what they want without outside interruption. The source code will make its way into the world and if the key players who weren't invited / represented think it's doing something useful it will get further modified and brought into distros like Ubuntu.
Here in the UK the same is very much true with people of Indian descent. On the slightly dubious basis of accents, the last two doctors I have seen have been recent immigrants as are some of the software engineers I work with, while typically the Indian people working at the local supermarket speak in a broad Lancashire accent, suggesting their families have been here at least a couple of generations.
It doesn't prove there is no discrimination, just that a lot of recent immigrants have the training and skills to overcome it. This is mainly due to the points based immigration systems and quotas now used in so many countries.
If you assume the 'average' American is European descent and then look at the success of recent generations of European immigrants you'll find exactly the same pattern because you have to have a good degree - possibly even a PhD, just to get in.
news - there's a great tag. Can someone also tag it slashdot in case we forget?
My iPod came with a big Apple sticker which for some reason I did stick on my PC. Guess I'm OK to use Safari then.
I'd at least do both of those - it's a bit of a hassle to be handed a web address, go home, switch on the PC, type in the address, wait for the files to download etc.
For a quick listen to see if you like the music, the CD is more convenient.
MySQL is a database that is very well suited to running a web application. That's why. Many web apps are generally going to have many more reads than writes. Additionally, the writes tend to be pretty simple and not require complex locking mechanisms.
E.g. my photos website uses gallery2 which works easiest on the LAMP stack. The main database queries are simple - what albums are there, does the user have permission to see them, what photos are in this album etc. The updates are similarly easy - add a new album, add a new photo, add a new user. For this, MySQL is perfect, simple and fast.
However once you start writing more complex systems with tougher constraints things become a lot harder. Typically, you have a record in a state that needs notifying to another system. You need to read that record, send a message to the other system and then modify that record to say the message is sent and the other system then has control over it. For this you need the various locking mechanisms and guarantees from the database or the two systems get out of line. MySQL just isn't the database to use for that sort of application.
Leaving aside the CPU / GPU power, it's not exactly comfortable playing games with a trackpad and tiny cramped keyboard. Also, powerful graphics takes a lot of power and results in lower battery life. Even in many high-end laptops, that's a compromise many people just don't want to take.
I've got a big 17" laptop with a decent graphics card - it has plenty of power for games, but I'd need to cart an external keyboard and mouse / other peripheral around with me to actually play games. I've tried playing strategy games on it while on the road but gave up pretty quick.
We aren't just seeing the death of PC gaming, but the death of the desktop PC. Laptop sales definitely overtook desktops on total value a while back - not sure if it's true on number of units shipped just yet. There just aren't all that many new desktop PCs being sold compared to a while back. People are deciding laptops are good enough and have so many advantages.
One area that they aren't good enough though is gaming. Hence more and more people who would like to game on a PC are having to consider buying a whole new PC to play games. I don't believe you need a $1200 graphics card to play new PC games, but I suspect you need at least a $1200 PC.
I think that quote is an exaggeration, but the general point is still true.
A lot of people now have a 5 year old PC that will do fine for everything except gaming - in the past PCs always could do with being faster and I think for many people that time is now gone. Many companies operated 3 year redundancy cycles for the office PCs, that have now stretched out to 5 years and longer. Where I work, people started opting to upgrade the RAM and occasionally the CPUs and we have a number of P3 era machines that people (software developers even) are perfectly happy with.
You'll have to spend $200 on a graphics card to get good results with a lot of games, and you'll probably go through at least 2 generations of cards in the time of one console generation. If the rest of your PC isn't up to scratch you might have to buy a whole new PC for gaming. Either way you're looking at similar prices to just buying a standalone console and being rid of the hassles of Windows.
For the interface, I disagree very strongly! I'm new to console gaming but I already prefer the PS3 controller for most games.
I think FPSs are much better with keyboard and mouse. I was never too sure how the Xbox manages to have so many popular titles in this genre when you can't use a keyboard and mouse. The same applies to strategy games and I still use my PC for these.
However, for just about every other genre I've played, a console-style controller is much better. Having standard buttons, analog directional pads and analog triggers all right under your fingers is just fantastic and brings a lot more depth and skill into control of most games. The motion sensing on the PS3 controller isn't really good enough for fine control but it is put to great use in Folklore and Tony Hawks.
I would guess the interface is a prime reason for a lot of people abandoning their PCs for consoles. Sure you can add specialist controllers to your PC, but a lot of games won't be designed for them.
The problem is the base level of hardware you must aim for on a PC is substantially lower than the power of a console. If you want to sell plenty of units, you'll probably need your game to run on Radeon 9000 class hardware. A Radeon 9000 is about 7 years old, but there are a lot of modern PCs with integrated graphics that would still struggle to match that performance.
So you start off with a game that has to run on a system like that, and add in the extra touches for people with the hardware to handle it. The result of that is your game never looks as good as you'd like because the niceties are bolted on optional extras, and the base of your game has to run on something with 10% of the performance.
My PC has a GF 8600GT which is pretty similar to PS3 power. Some games can match my PS3 on graphical quality, but they are typically big-budget ones where the developers can properly program multiple code paths for older and newer hardware.
Vista really is terrible though. I was forced by company policy to get Vista on my new PC, and it's just full of bugs. Some of them pretty major. I've just done the SP1 upgrade this morning - yet to see if this is overall better or worse. They've still not fixed the Explorer bug where it can't read any CDs/DVDs, and VMWare now takes about 10 minutes to start instead of 10 seconds. I can now finally use Remote Desktop from an XP machine though.
Not many people thinking of switching to Vista have first hand experience. Us techie geeks might have a quick play with someone else's computer when we see a new thing on it, but for most people it's not interesting. Their opinion is always based on second hand reports from other people.
In my own experience, almost every Vista user has reported headaches of one kind or another.
Blu-Ray is not Sony's format. They are just a contributor and do not have that level of control to stop MS adding a Blu-Ray drive for the Xbox 360 or anything else. Microsoft were never happy about the Java based interactive features on Blu-Ray and supported HD-DVD mainly because they managed to stuff their sort-of proprietary HDi technology into that format.
More importantly, you are looking at the situation in reverse. Although I think having Blu-Ray is an advantage for PS3, the hardware cost increase is a disadvantage. To a greater degree, the PS3 is an advantage for Blu-Ray. Looking at standalone unit sales, PS3 won the format war for Blu-Ray single handed. Sony will earn some royalties from Blu-Ray. From this point of view, I think Sony would be quite happy for cheap Blu-Ray addon drives for the Xbox. The PS3 hardware can only be considered profitable because it's designed to earn money from sales of games and movies over the next decade. The hardware itself will never be that profitable because moving more units brings in more money down the line.
1) It scales down a bit at least. I'm pretty they were marketing it last year for projector bulbs at around 150W. Not sure whether it scales further down than that.
:(
2) 6000K is very close to sunlight so yeah it's a nice warm sunny light - should in theory be nicer than incandescent light anyway.
3) No - it's a noble gas (unreactive) and naturally present in the atmosphere, making up nearly 1% of it in fact.
4 and 5) Dunno. I was just searching for the projector bulb version and couldn't find any actually for sale, which given that it was announced half a year ago isn't great going
No, it does reach that heat, but in a very small area and doesn't radiate too much heat. Normal incandescents reach about 3000 C and don't fry everyone because only the filament is at that temperature - the glass bulb is hot enough to burn your fingers but nowhere near the temperature of the filament.
Somewhat paradoxically, at a higher temperature, more radiation is of the visible light spectrum and less is waste heat. So these lights will radiate less heat because they are hotter.
I believe you are partly right. I do work with multi-process applications, but I think part of the reason these are simpler is that you really have to sit there and decide on data to be shared and explicitly create shared memory segments for this data and then attach each new process to those segments (in C at least).
When I have used multi-threaded programming (in Java), it is very tempting, when lots of extra data is shared to just say 'OK I'll use this variable just as a one off to get this done'. Usually in the long run that's a very bad choice because the data was not designed to be shared.
Actually designing ALL of the data that can be shared is an important part of getting the design right and avoiding a lot of the little bugs that crop up. From this point of view I really prefer multi-process work than multi-threading as it is more natural to do things 'right'.
I think many or even most applications can make use of parallel programming with relatively little real effort. Locks and other mechanisms are tricky to get your head around - the real trick is coming up with a design that doesn't need them, or at least minimises that need.
The systems I write store records in a database and minimising locking issues is really about thinking through the broad states that different tasks will deal with. So anything in state A is managed by task X and all other areas of the system should ignore state A (or at least, not change things out of this state).
This results in a much tidier system and much smaller and more readable code. The system I am currently working on has only about 5 background processes that will do most of the work, and the target hardware has 8 cores, so this leaves 3 for other bits and pieces but mainly for dealing with user interaction. If I need to aim at more cores, I know I can easily split the main 5 into about 12 processes without even having to think about locking.
Those calculators compare money equivalency based on inflation figures only - the change in the value of 'money' over time. The bit that is really missing is earnings growth i.e. actual changes in productivity.
For instance if a cheap pair of shoes cost $10 in 1917 and now costs $30, and this sort of measurement was true over a lot of goods, it is true to say $10 then is the equivalent of $30 now.
However this says nothing about how affordable things are. As you say, if the average salary in 1917 was around $700 / year, a pair of shoes is almost a weeks wages. Nowadays it's a couple of hours wages.
Cars are a tricky example because lots of people are happy to spend on them. The price doesn't really drop as much because we instead demand better cars. The cheapest new cars around today are a small fraction of yearly earnings and still bigger, faster, more efficient, more comfortable, safer and more reliable than any car from 1917.
Exactly my opinion also. We are just reaching the point when a properly useful ultra-portable can be had at a decent price. I looked longingly at the Sony TZ series but couldn't justify spending £2000. I've been seriously considering getting the old EEE, but in the end the screen resolution was a bit too low. The new one is just enough and I will also be buying one. I don't care if Walmart sell a similarly priced full size and full featured laptop - I've got one of those and that's not what I am after.
When I sold my house the estate agent website had an SWF that downloaded the images and made it tricky to save. I ran Firefox's tamper data add on to see where it was getting them and found a nice image server with the full high-res wide-angle photos their photographer had taken.
When I bought my last house, it came with a stack of old documents dating back to its building in 1885. It was quite interesting seeing the different styles of legalese that were prevalent at various times. A significant number of the documents had absolutely no punctuation, because even adding pauses into a sentence can change its meaning e.g. Lionel Hutz' advert on the Simpsons saying
Works on commission
No money down
gets corrected to 'Works on commission? No, money down!' That's clearly a joke but if someone does similar with the deeds of a house it's not quite as funny. The laws of a country are considerably more important again.
What is the cost of switching?
I think that's a question you could spend more than 2 years thinking about and not get an answer. How do you know when you're overcomplicating an issue, and when you're just doing something complex?
Developers always horribly underestimate the difficulty of a task. Every solution starts clean and simple, and it's incredibly hard to keep the finished product looking anything similar to the one which did 95% of the job.
You always look at a nasty solution and think it should be much simpler done this other way. Half the time you're right, but half the time you add in all the little niggles to your overall work of art and you're back at the starting point with an ugly solution that works.
How do you figure out in advance which half you're heading for?