Just because you may not live to see the results from the research, it shouldn't be funded? I can see how the idea is attractive, if you are only concerned about your single nearly worthless life.
Also, letting the masses decide how to spend their tax dollars is an equally poor idea, if only because many people have the same shortsighted viewpoint you have.
Without funding experimental research into the unknown, the unknown will remain unknown.
Who knows if the space elevator might allow for improved zero-g research facilities that develop the cure for AIDS and cancer?
Then again, you might be right. That Hummer I could buy with my 'misspent' tax dollars will certainly do humanity a whole lot more good than drastically reducing the cost of access to space. NOT!
This is also why I have much love for LaTeX.
I'm dyslexic and staring at a formatted page drives me nuts. Worrying about the format drives me off my mental path, and it takes me a while to get back onto the train of thought.
With latex I don't have to worry about that much at all. I create a new document, then outline each chapter, section and paragraph. Then i come through and fill in the text for each section.
Combining the outline and the text in a single document is a lifesaver for me. I've gone from struggling through writing docs, to blasting out page after page after I switched to LaTeX
Google tags things as beta for many reasons. I think the term should be banned now (see post below). Really, every google app is subject to frequent and major change.
Beta means that the code is feature complete, but not tested. I am pretty sure that all of googles apps have seen enough use to identify most of their flaws by now.
Google hosts these apps and makes improvements all of the time. The beta tag might be there for legal reasons (google news), and that has some validity to it. I really think that they call everything beta because they want the ability to freely modify it in the future to fit their plans. This includes being able to stop a service at any time. They present services as a beta, and can kill it if they want. Its just a beta, so people should not be creating production code around it. Google Scholar not paying the bills? Can it!
Of course things like Google Maps ARE being incorporated into production codebases. While I don't really think that incorporating a free beta service hosted by a different company into my companies site is a good idea, other people don't mind.
If Maps were to go away, the people relying on it for production apps would be screwed, but Google can say that it was a beta and cover their ass that way.
Really, 90% of googles apps are production ready code simply tagged as Beta for a variety of reasons. None of those reasons relate to bugs or stability of the code.
90% of people don't know what Beta means. Sure corporate support IT folks might know, but your average household probably has no clue.
Plus, the distinction between Beta and General Availability software isn't always clear. I've used GA software that should be beta, and vice versa.
I don't think this point is valid in todays computer world.
See "Contact" for a similar situation. Religious whack-job destroying the space transporter thingie.
I also believe that if a person or group is sufficiently motivated, no amount of process and security will stop them. It only raises the bar on the difficulty of the operation.
Apple is using a version of C that fakes object orientation through a runtime and some clever preprocessing. Nobody expected objective C to be around in the year 2005.
Hey buddy. Objective-C is one of the more object oriented flavors of C out there. I would personally rank them (in object orientation) like this:
1. Objective C
2. C# / Java -- Can't decide which is better, so they tie
3. C++
Objective C's runtime provides features and capabilities that other languages do not have.
FEMA and the federal government can not just decide to basically invade a state. This is a very good thing.
The local governments direct the federal resources in a coordinated effort. That coordination was completely missing in this example.
I agree with many of your points though and believe that all parties share a bit of blame. The good thing to come out of this is that the issues have been brought to light and that those issues will eventually be resolved.
The racism that left FEMA, the Corps of Engineers and the rest of the government response totally unprepared for a totally predicted catastophe. White people, even the poor, in Florida last year got speedy, effective relief (just prior to the election) when 4 hurricanes crossed their state. In New Orleans, the levees were underbuilt, the plans ignored, abandoned and screwed up. The difference? The biggest difference was that most New Orleanians are black, most Floridians are white.
I have a couple of issues with this. The first is that the issue in NOLA was about class more than race. The people who could not leave were the lowest classes, of all races. The people without the funds necessary to evacuate.
The _LOCAL_ government should have had a plan to evacuate the people. Read that again. LOCAL. The blame for NOLA being full of people when the disaster hit lies fully on the Mayor of New Orleans, and to a lesser extent the Governor of LA.
The reason the Florida performed so much better last year was not because the people were white (there are a lot of minorities in FL, more in total than in LA). Florida did well because the local and state governments coordinated very well with FEMA on the response.
In NOLA, for the first few days FEMA was trying to work with the locals, but the communication was not there. Eventually FEMA recognized that the mayor was a Loon and that the governor needed to be told what to do in very short sentences. FEMA eventually took over the situation, and within 24 hours things drastically improved.
The levies would have failed even if _every_ proposed project for flood control was initiated. None of them addressed anything over a class 3 hurricane.
At the end of the day, you should be looking to you LOCAL government officials. City, County and State. Those are the people who are going to evacuate people and manage the crisis for the first 72 hours.
From an economic standpoint, it really makes since to toss billing out the window right now. It is in _everyones_ benefit to get this area economically back on the ground.
The entire US is going to feel the pain from this hit. The Mississippi is the single largest shipping port in the US. This means BILLIONS of dollars for incoming and outgoing products.
Just Oil and Gas alone is going to drastically effect the economy.
Basically anything these companies do to minimize the damage and turn around the economic base more quickly is going to pay off huge dividends in the medium to long term.
Everyone in this country should be doing whatever they can to ease the pain, because it is in their economic best interest.
Ummm, your using three partitions on two disks to create a raid 5 array? Thats idiotic and totally defeats the point of raid 5. You need to have your data disks on seperate spindles from your parity disks. At best you should do a raid 1 on the two drives to get some redundant space.
This too is an old idea. Basically when dealing with any sort of foil (such as a wing or rotor) you can't really use the last two or three feet effectively, as the low pressure air from the top slips around the end of the wing instead of over it. Putting some sort of vertical or 90 degree winglet on the end traps the air so that is must flow over the wing surface as designed.
By placing the end pieces on the wing you are effectively adding two or three feet to the length of the wind, and only taking a small increase in friction. Good times all around.
I think we are both arguing the same point.
Basically, I am saying that if you are going this route, and have a manageable load, the best way to get a 'true redundant' power supply is with two large PSes, each capable of running the full load.
Going with more than two, in my experience, leads to more downtime. Granted, I am not a hardware engineer, but I have worked with a great deal of equipment, and the two PS solution appears to be more reliable.
I think you are almost right, but are missing it by just a bit.
The problem with these types of setups are that when you finally do lose a single power supply, and the other two step up to fill in the gap, you will often lose one or two of those as well.
If you had more than three supplies (say 6 or 7) then the increase when one dies is negligible. In the 2+1 configuration you talked about, you'll be looking at a 50% load increase when one of the three die.
For example, you have a server that requires 1000w of power, and have three 500w power supplies. Most of the time, each is only providing 333w, but if one dies, the remaining two shoot up to 500w a pieice.
The problem is that one of the other two might have been doing fine at 333w, but struggle with 500w. After working hard for a few minutes and heating up, it too will die. This leaves you with a server that is just as dead as one with a non-redundant power supply. Actually, maybe slightly more dead, as you will have to get more than one PS to get back into production.
The two options are to either provide many more smaller power supplies, or go to 2 power supplies that are each capable of taking the entire load of the machine, and running the box single legged on one PS until it fails over to the redundant supply.
Since things like power supplies tend to get more efficient as they grow larger, having two of these large supplies is better in the long run than more of the smaller supplies.
Mission critical servers I work with have two power supplies that are each hooked to separate redundant circuits (220v or 440v AC). If one of the PSes take a dump, the other picks up the load. Admins receive an alarm and swap out the failed PS, leaving the server in production with no outage.
This is just great. Lets teach our kids how to close their brains even more. Instead of using the wonderfully flexible english language, these kids are going to down to a couple of choices. A, B or C.
You speak the truth. I personally know of one Quadra 900 which has been working continuously as the only mail server for a publishing company of around 75 employees.
The server was purchased in 91, and moved into the mail server role in 93. For 12 straight years, this server (and its measly 9gb external Lacie SCSI hard drive) has been plunking along. I mean, this box has an AAUI adapter to connect it to a 10-BaseT network!
No OS updates, and no hardware issues. The only reason the box has been rebooted in all this time is after an extended power failure which takes the UPSes offline.
And that really changes everything. I believe that given the current state (stale) of windows, and the number of real and perceived issues people are having with Windows, Apple could make a good run of selling OS X to the masses to run on standard PCs.
The problem is that OS X would be in direct competition with Microsoft at that point. Selling OS X to run only on Apple Intel powered PCs is one thing, but full scale OEM licensing to any x86 manufacturer is a totally different ball-game.
While MS has been having a bit of trouble executing in the last couple of years, I wouldn't want to be Apple in a direct OS competition war. MS has massive (un-ending?) resources, and many of the smartest people in the world working for them.
I believe that for the next couple of years, Apple is going to carve small pieces of MS customers base away, moving them to Apple built Intel hardware. Once Longhorn and its predecessors finally ship, the situation will change somewhat.
Any way you look at it, Apple has a very wide road ahead of them for the next few years, and they are going to grow the platform. Will that make them strong enough to compete head to head with MS by 2010? Who knows.
Yeah, it has really been killing Microsoft's Bottom Line:)
Software is not a burden. Apple uses their software to sell their hardware. I wouldn't buy a computer if I had to code everything I wanted myself (i'd install Linux if I wanted that:) Just a joke, don't kill me).
I think that Apples future lies in their software. Really, people are in love with their Mac because of OS X, not because of the silicon and components that make up the hardware.
Apple is transitioning to a future where they can produce more higher quality hardware, and they are going to use OS X to sell the pants off of it.
People think that Apple (the stock) is over-valued. I think that the market has merely recognized that Apple is now in a position to increase its sales 25%+ Year over Year for more than a couple straight quarters.
It is really odd that Apple is now in a position to leverage.... Windows XP to break into a larger market. Think about that for a second.
Believe it or not, there are tons of people out there with more than enough money to buy just about any kind of computer they want. Paying 2%0 more or even 80% more than the cheapest available comparable system is no biggie for these people.
What they do want tho is a system that they know will work for them. Many, many people are afraid of the big switch to PPC Mac OSX because they don't know if OS X will work for them, and if it doesn't, the hardware is basically useless for them (i.e. they aren't going to install Linux and be happy with it).
With the new Intel machines, these people can now buy apple gear knowing that they can go back to XP if need be. I'd bet that a big majority of these people never get around to even installing XP.
I agree that the numbers are fishy, but I also disagree with your numbers as well.
Any company over 2,500+ workstations is going to have _at least_ one mac. Be it for testing external websites, publishing, or the crotchety manager who only uses macs.
So, 12-21% can not represent the number of companies with at least one mac, that number is going to be much higher.
Slashdot Mirror
Fantastic idea.
Just because you may not live to see the results from the research, it shouldn't be funded? I can see how the idea is attractive, if you are only concerned about your single nearly worthless life.
Also, letting the masses decide how to spend their tax dollars is an equally poor idea, if only because many people have the same shortsighted viewpoint you have.
Without funding experimental research into the unknown, the unknown will remain unknown.
Who knows if the space elevator might allow for improved zero-g research facilities that develop the cure for AIDS and cancer?
Then again, you might be right. That Hummer I could buy with my 'misspent' tax dollars will certainly do humanity a whole lot more good than drastically reducing the cost of access to space. NOT!
The kind of companies that would worry about this have already banned cell phones in sensitive areas.
This is a non-issue.
This is also why I have much love for LaTeX. I'm dyslexic and staring at a formatted page drives me nuts. Worrying about the format drives me off my mental path, and it takes me a while to get back onto the train of thought. With latex I don't have to worry about that much at all. I create a new document, then outline each chapter, section and paragraph. Then i come through and fill in the text for each section. Combining the outline and the text in a single document is a lifesaver for me. I've gone from struggling through writing docs, to blasting out page after page after I switched to LaTeX
Google tags things as beta for many reasons. I think the term should be banned now (see post below). Really, every google app is subject to frequent and major change.
Beta means that the code is feature complete, but not tested. I am pretty sure that all of googles apps have seen enough use to identify most of their flaws by now.
Google hosts these apps and makes improvements all of the time. The beta tag might be there for legal reasons (google news), and that has some validity to it. I really think that they call everything beta because they want the ability to freely modify it in the future to fit their plans. This includes being able to stop a service at any time. They present services as a beta, and can kill it if they want. Its just a beta, so people should not be creating production code around it. Google Scholar not paying the bills? Can it!
Of course things like Google Maps ARE being incorporated into production codebases. While I don't really think that incorporating a free beta service hosted by a different company into my companies site is a good idea, other people don't mind.
If Maps were to go away, the people relying on it for production apps would be screwed, but Google can say that it was a beta and cover their ass that way.
Really, 90% of googles apps are production ready code simply tagged as Beta for a variety of reasons. None of those reasons relate to bugs or stability of the code.
90% of people don't know what Beta means. Sure corporate support IT folks might know, but your average household probably has no clue. Plus, the distinction between Beta and General Availability software isn't always clear. I've used GA software that should be beta, and vice versa. I don't think this point is valid in todays computer world.
See "Contact" for a similar situation. Religious whack-job destroying the space transporter thingie. I also believe that if a person or group is sufficiently motivated, no amount of process and security will stop them. It only raises the bar on the difficulty of the operation.
Why start investing in a project when you don't even know if you will legally be able to do it? Get approval first.
As little as 6 milliamps through your heart will kill you though. So he lucky he radiated that energy rather than piping it thought his trunk.
Fact stolen from tonight's mythbusters.
Apple is using a version of C that fakes object orientation through a runtime and some clever preprocessing. Nobody expected objective C to be around in the year 2005. Hey buddy. Objective-C is one of the more object oriented flavors of C out there. I would personally rank them (in object orientation) like this: 1. Objective C 2. C# / Java -- Can't decide which is better, so they tie 3. C++ Objective C's runtime provides features and capabilities that other languages do not have.
FEMA and the federal government can not just decide to basically invade a state. This is a very good thing. The local governments direct the federal resources in a coordinated effort. That coordination was completely missing in this example. I agree with many of your points though and believe that all parties share a bit of blame. The good thing to come out of this is that the issues have been brought to light and that those issues will eventually be resolved.
I have a couple of issues with this. The first is that the issue in NOLA was about class more than race. The people who could not leave were the lowest classes, of all races. The people without the funds necessary to evacuate.
The _LOCAL_ government should have had a plan to evacuate the people. Read that again. LOCAL. The blame for NOLA being full of people when the disaster hit lies fully on the Mayor of New Orleans, and to a lesser extent the Governor of LA.
The reason the Florida performed so much better last year was not because the people were white (there are a lot of minorities in FL, more in total than in LA). Florida did well because the local and state governments coordinated very well with FEMA on the response.
In NOLA, for the first few days FEMA was trying to work with the locals, but the communication was not there. Eventually FEMA recognized that the mayor was a Loon and that the governor needed to be told what to do in very short sentences. FEMA eventually took over the situation, and within 24 hours things drastically improved.
The levies would have failed even if _every_ proposed project for flood control was initiated. None of them addressed anything over a class 3 hurricane.
At the end of the day, you should be looking to you LOCAL government officials. City, County and State. Those are the people who are going to evacuate people and manage the crisis for the first 72 hours.
_A_ business? Try entire industries. For example, how many companies fortunes have been made with digital computers?
Apollo brought a massive boost to the technological knowhow and capabilities of the entire world.
From an economic standpoint, it really makes since to toss billing out the window right now. It is in _everyones_ benefit to get this area economically back on the ground.
The entire US is going to feel the pain from this hit. The Mississippi is the single largest shipping port in the US. This means BILLIONS of dollars for incoming and outgoing products.
Just Oil and Gas alone is going to drastically effect the economy.
Basically anything these companies do to minimize the damage and turn around the economic base more quickly is going to pay off huge dividends in the medium to long term.
Everyone in this country should be doing whatever they can to ease the pain, because it is in their economic best interest.
Ummm, your using three partitions on two disks to create a raid 5 array? Thats idiotic and totally defeats the point of raid 5. You need to have your data disks on seperate spindles from your parity disks. At best you should do a raid 1 on the two drives to get some redundant space.
By placing the end pieces on the wing you are effectively adding two or three feet to the length of the wind, and only taking a small increase in friction. Good times all around.
I think we are both arguing the same point. Basically, I am saying that if you are going this route, and have a manageable load, the best way to get a 'true redundant' power supply is with two large PSes, each capable of running the full load. Going with more than two, in my experience, leads to more downtime. Granted, I am not a hardware engineer, but I have worked with a great deal of equipment, and the two PS solution appears to be more reliable.
The problem with these types of setups are that when you finally do lose a single power supply, and the other two step up to fill in the gap, you will often lose one or two of those as well.
If you had more than three supplies (say 6 or 7) then the increase when one dies is negligible. In the 2+1 configuration you talked about, you'll be looking at a 50% load increase when one of the three die.
For example, you have a server that requires 1000w of power, and have three 500w power supplies. Most of the time, each is only providing 333w, but if one dies, the remaining two shoot up to 500w a pieice.
The problem is that one of the other two might have been doing fine at 333w, but struggle with 500w. After working hard for a few minutes and heating up, it too will die. This leaves you with a server that is just as dead as one with a non-redundant power supply. Actually, maybe slightly more dead, as you will have to get more than one PS to get back into production.
The two options are to either provide many more smaller power supplies, or go to 2 power supplies that are each capable of taking the entire load of the machine, and running the box single legged on one PS until it fails over to the redundant supply.
Since things like power supplies tend to get more efficient as they grow larger, having two of these large supplies is better in the long run than more of the smaller supplies.
Mission critical servers I work with have two power supplies that are each hooked to separate redundant circuits (220v or 440v AC). If one of the PSes take a dump, the other picks up the load. Admins receive an alarm and swap out the failed PS, leaving the server in production with no outage.
Not the one with 80%+ of the marketshare!
This is just great. Lets teach our kids how to close their brains even more. Instead of using the wonderfully flexible english language, these kids are going to down to a couple of choices. A, B or C.
Fantastic.
The server was purchased in 91, and moved into the mail server role in 93. For 12 straight years, this server (and its measly 9gb external Lacie SCSI hard drive) has been plunking along. I mean, this box has an AAUI adapter to connect it to a 10-BaseT network!
No OS updates, and no hardware issues. The only reason the box has been rebooted in all this time is after an extended power failure which takes the UPSes offline.
Now thats ROI.
The problem is that OS X would be in direct competition with Microsoft at that point. Selling OS X to run only on Apple Intel powered PCs is one thing, but full scale OEM licensing to any x86 manufacturer is a totally different ball-game.
While MS has been having a bit of trouble executing in the last couple of years, I wouldn't want to be Apple in a direct OS competition war. MS has massive (un-ending?) resources, and many of the smartest people in the world working for them.
I believe that for the next couple of years, Apple is going to carve small pieces of MS customers base away, moving them to Apple built Intel hardware. Once Longhorn and its predecessors finally ship, the situation will change somewhat.
Any way you look at it, Apple has a very wide road ahead of them for the next few years, and they are going to grow the platform. Will that make them strong enough to compete head to head with MS by 2010? Who knows.
Yeah, it has really been killing Microsoft's Bottom Line :)
Software is not a burden. Apple uses their software to sell their hardware. I wouldn't buy a computer if I had to code everything I wanted myself (i'd install Linux if I wanted that :) Just a joke, don't kill me).
I think that Apples future lies in their software. Really, people are in love with their Mac because of OS X, not because of the silicon and components that make up the hardware.
Apple is transitioning to a future where they can produce more higher quality hardware, and they are going to use OS X to sell the pants off of it.
People think that Apple (the stock) is over-valued. I think that the market has merely recognized that Apple is now in a position to increase its sales 25%+ Year over Year for more than a couple straight quarters.
Apple did _NOT_ leave OS 9 users in the dust. That is a complete fabrication.
OS X had the classic environment, which you can still use in 10.4. Also OS X installs on just about any Mac sold in the last 5 years or so.
Maybe you consider only supporting people for 8 years after they bought their computer leaving them out in the cold, but I don't.
It is really odd that Apple is now in a position to leverage.... Windows XP to break into a larger market. Think about that for a second.
Believe it or not, there are tons of people out there with more than enough money to buy just about any kind of computer they want. Paying 2%0 more or even 80% more than the cheapest available comparable system is no biggie for these people.
What they do want tho is a system that they know will work for them. Many, many people are afraid of the big switch to PPC Mac OSX because they don't know if OS X will work for them, and if it doesn't, the hardware is basically useless for them (i.e. they aren't going to install Linux and be happy with it).
With the new Intel machines, these people can now buy apple gear knowing that they can go back to XP if need be. I'd bet that a big majority of these people never get around to even installing XP.
I agree that the numbers are fishy, but I also disagree with your numbers as well.
Any company over 2,500+ workstations is going to have _at least_ one mac. Be it for testing external websites, publishing, or the crotchety manager who only uses macs.
So, 12-21% can not represent the number of companies with at least one mac, that number is going to be much higher.