Aside from the advice given by others about calling the financial aid office, do remember that your education is in your own hands.
That sound's snarky, but it really is true - you need to take responsibility for funding your education, which means deciding what you are willing to do. You want to work for FOSS, but are you willing to work for the spooks? How about big oil companies? Big pharma? The military? Once you know that, there are several steps. First, get the FAF/FASF forms filled out as soon as possible. They determine what the school will think you can afford. If that is substantially different from what your parents are willing to pay, you need to know that right now.
For example, if your parents are told to pay $15k a year, and they are only going to pay $10k, you need to find another $5K at least. You can probably do that - there are many loan programs, there are decent summer jobs for programming interns, and at the end of the day, there are ROTC programs. Piles and piles of debt kinda suck, but it is better than giving up on an achievable dream. On the other hand, if you walk away with $120K in debt after four years, you are going to have very limited options - forget FOSS, you will need to take the highest paying job you can find just to stay on top of the payments.
Second, be realistic about where you fit in. Take all the assessment tests you can find, and find the norms for the schools you like. You should want to be out of the top 1% - it makes it hard to learn if you are targeting a very different level of kill than your peers. That said, you might want to find a school where your combined math and verbal is around the top third or top half. You then have many people to learn from, but you will also do fairly well and should not drown.
Third, expect college to be different from high school. Colleges have many people who you can learn from. I do think the MIT education is worth it, but you can watch the lectures for free from Stanford, Berkeley, and MIT off of iTunes U. Try doing the entire MIT algorithms course, and see how it feels. Once you know whether it is pitched at your level, you can write an admissions essay that describes in real terms what excites you, what you have to offer, and why it is worth them taking a risk.
Have you considered a Google Summer of Code project? A good one is not a bad way to get some real code checked in. Other companies have their equivalent.
I went to Harvey Mudd, and I would rather screen mudders or MIT people than people from, say, USC or UC whatever. Not because of any inherent quality, but because Mudd, MIT, and other institutions of their caliber generally admit very good people, push them hard, and give them access to the very best resources. Further, they are used to being around smart people, so there is less of the lesser nerd bull - you do not have to prove you are a fripping genius with every statement, and thus we can get on to the work.
Knowing that they came from HMC/MIT increases the odds of finding a very bright generalist. Further, I have found that Mudders tend to be aware of their limitations - they do not know everything, and they know it. I have found that people from the Ivies tend to assume their own genius. I would rather someone know what they need to learn, because people that smart should always be learning.
Some people come out of the best schools barely capable of independent thought. Some people come out of middle or lower tier schools sharp as a tack. My job in hiring is to figure out whether a given candidate is one of those very good people. So, will Mudd on your resume get you in the door? No, but it might just get you a phone screen even with an unexceptional resume, because your resume might not be indicative of real skills.
That said, I _have_ hired people from USC, UCI, and other schools. They tended to be unusual for their schools, in that they were generalists, thoughtful, and just plain interesting, but that is unusual in the general population, and no degree is a guarantee of that flexibility.
You would almost certainly find that an MIT education has value, as does an MIT degree, but if you do not get there, try to excel and try to learn as much as you can about as much as you can, as the world does change over the course of your career, and this is a good chance to learn that flexibility.
So, would I pauper myself for an MIT degree? No. I might take out student loans - I had to when I was an undergrad - but I would make sure I had a sustainable amount of debt and a plan to get it paid off. Don't expect a killer salary, but do expect to find something rewarding with a bit of searching. Debt limits your options, just like a ROTC commitment, but if that is what you need to do to get where you want to go, then by all means, do it. The name has some value, but the work you will have to put in to get it, and the people you will get it with has much more value over the long term.
My own buying habits demonstrate how raising prices will result in lower total revenue.
I play music while gaming. Since we may game for four or five hours over a weekend, I like to get two new movie scores every month. Sometimes it was hard to find two that were good in any given month; finding good music was the limiting factor, not money.
As prices climbed to $18-$20 at Tower, I found I was only buying one a month, or one every two months. This is not a moral stand; I found that the albums cost more than I was willing to pay. The samples did not say "drop $18 on me".
Just after the ITMS opened, I went back to buying albums, mostly at the music store, but often on CD as stores were dropping their prices. I bought one or two a month steadily for over a year. Easy enough to see by looking at the creation date for the music.
I have not bought a new score on CD for the last three months, as the prices have climbed back to $18 at Tower. I have three I am considering at the ITMS, but I note with some dismay that many are not available on a per-song basis.
Again, this is not a moral objection, but a practical one. The Sahara DVD I just bought cost me under $20 with a pre-order. I am not willing to pay $18 of that for a CD, and ITMS is not an option. (The soundtrack is there as an album only, and the score is not to be found.) I find that the music I am hearing is not worth the price they want me to pay. I will live with my 600 albums, and nearly a thousand ITMS tracks until the prices drop again.
I must say, if you own stock in a record company, start selling. There is only one bright spot in the marketing reports from every major label, and that is the ITMS. Killing it by raising the prices for the tracks worth having will just cut revenue.
I am living proof that if they start raising the prices, the product will not sell. Period.
The people working on the Intel transition are aware of the risks, and are addressing them like the consummate professionals they are. I talked with a lot of them at WWDC 2005, and they know what they were doing, and they know what roadblocks could come up and bite them.
You cannot repair or replace a G4/1.5 powerbook motherboard with any processor Intel ships. By having supply through 2008, they can satisfy any Applecare requirements they may have. This alone would justify their contract with freescale. After all, Applecare is three years, and they are going to be selling laptops with G4 processors until early 2006.
Further, depending on Intel's price points, there may be a place for G4 iBooks or Mac Minis for a few months after introduction. Apple currently sells laptops and minis ranging from $500 to $3000, which is a very big range. It is not clear to me just how quickly they will change every segment of their laptop and mini offerings. Certainly, the high end laptops will change fast, as Apple is getting creamed in that market. For a mini, though, where battery life is not a critical factor, but price is?
The only concrete statement we have comes from WWDC - they will have something running by WWDC 2006, and will have completed their transition by the end of 2007. That is a two year range, in which every machine must change. I am betting that the G4 will vanish overnight, but I am not going to bet the farm, and having an insured supply of chips means that Apple is not going to either.
It takes confidence to make a change this drastic. I, for one, am in favor of it.
Open source has two benefits: customers of the product have access to the source, and a wider community can read, change, and improve the source. Announcements like the Jasper ones force the community to decide where they stand.
Put another way: if the lead devs decide to move, and get paid for their work, then we find out whether the project was robust or fragile. If the community does not step up to the plate, then they did not care enough.
To me, that is just fine. It makes it clear where we put our time and treasure. Projects that fail for this reason were fragile, depending on the good will of one person.
When it has been fought out, the reporter usually lost, unless that which they were reporting was in some way illegal. Whistle blower laws then usually protected the reporter and the sources.
This is as it should be - the source in this case stole proprietary trade secrets. The reporter is publicizing proprietary information. If the reporter does not list sources, then the fault quickly becomes his.
This is up there with me breaking into your house, stealing your credit card number, and then the rumor sites claiming that they do not have to help the police, because they did not do it.
Reporters are responsible for what they print, and the people do NOT have an intrinsic "right to know". The people may have a right to know any given fact, and we need to protect the ability of the press to publicize the things that are important, but they do not have a God, or Constitutional, right to tell any damn thing without consequences.
When someone breaks the law, the press does not get attorney-client privilege.
Java is what I do for a living, and I use a TiBook/667 to do it. It is a pretty good place to work.
I find Circus Ponies Notebook and BareBones BBEdit to be real helps to Eclipse/IDEA and the command line tools I use for most work.
Do be aware that releases often lag a bit, though, so if you absolutely need to use Java 1.5 as soon as the beta hits the street, you will need a PC or VPC. I choose to run VPC to test my 1.5 beta apps, and to run the rest under MacOS X.
Microsoft clearly makes money from "number 2" by making Mac office. Thus, they _can_ answer the question "what do we get?" with the words "cold, hard cash, and yet another spike in the coffin of potentially competing office products."
Think about it - were MS to stop shipping office for Mac, Apple would have to come up with an office suite of their own. They would probably fail, but MS really does not want to take that chance, given that Apple has enough cash to make it happen.
Why would IBM want to use a processor architecture other than Intel? Because they make oodles of cash on IBM servers, and they do not want to share with Intel.
Why do people want to buy Macs? Because they are better, and thus I make more money as a consultant. (You do not have to agree, but this is why I use it, and why my company buys them.)
In all three cases, the answer is "because we make money by doing so."
Contrast this with Real. They are known for dreadful software, spyware, and relatively poor quality. They have addressed some of these issues, but i know very few people who are fond of Real software. Thus - what would Apple get out of this partnership?
The work in making a clone or custom mac is not in the design of the cpu, but in getting all the rest of the hardware integrated with the OS. Add in the thrill of device drivers, and potentially having to reverse engineer any custom Apple HW, and you get a pretty much clone-free market. Opening up the cpu will not change that.
It does mean that the architecture might be used in more places than it now is. Next generation video games might move to the Power architecture if they see a benefit. Similarly, good Linux offerings will make scientists consider clustering PPC boxes rather than x86 boxes for high performance numerical computing.
If these events happen, then new support, and new understanding of the architecture will hit the streets in a way that will benefit Apple. For example, if serious money goes into optimizing gcc by someone other than Apple, releases of OS X will get better without Apple effort. Similarly, more money from other hardware purchasers gives IBM incentive to advance the architecture.
Putting in a battery access door to make it trivial to replace takes room and makes the device more prone to failure. (The seals around doors are never great, unfortunately.) As it is, based on the number of people in the class action suit, battery failure affected a fairly small number of units; it would not surprise me if one whose battery can be replaced easily had a higher failure rate.
The battery is probably as replacable as a standard iPod - about a hundred from Apple, about two thirds that from other vendors, and about half that if you can do it yourself. If it worries you, get Applecare.
For what it is worth, my wife's 5G iPod bought when they had only been out a month is still going strong, and my 20G bought a week after they showed up at the Apple store is doing well too.
Depends on the job. If your job is to keep the machines up without ever talking to a human, sure! Be a condescending ass if you like.
If, on the other hand, your job is like every job I have had - solving technical problems to produce whatever the organization produces - then you do need social skills. More to the point, you need to have people trust you. If you are good at technical things, then you probably know more about them than the decision makers. They are not going to absorb your knowledge in an hour of mentoring, so they will still be making their decision about technical issues based on whether they trust you.
Further, the claim "if I'm condescending its because a lot of people who're above me are there not because they're better than me but because they have the "Oh so called Social Skills." is nonsense. If the people above you are really utter loads, find a new job. If, on the other hand, they have better social skills than you have shown, it may be that they are above you because people trust them.
Ask yourself this question. If your accountant's response to "why did you deduct $4286 as charity this year?" was "I am the accountant, and you are the clueless loser. Don't ask stupid questions.", would you feel happy? More to the point, would you keep trusting him with your finances?
I wouldn't.
Finally, when you stop respecting people, you stop listening to them. In many cases, you are right about what they want, and thus not listening is the right answer, but you will miss things, and you will never know that you missed them. THe people you are filled with contempt for _do_ know when you are writing code/emails/shell scripts during meetings, and are not filled with trust later.
People who do not trust you are not likely to promote you. Blame social skills if you want, but it sure seems likely that they have a real reason.
I note that our consulting group recently extended machine lifetime to three years, while at most of our clients, it has been pushed out to five or more years.
Perhaps this is not the pentium 90 era, but it does show the importance of working well on older hardware. Apple may drop support for some older machines, but if each release gets faster, then they get a steady stream of revenue from users who have not bought new hardware AND they get users with those older machines up to date on software. Up to date users are a lot easier to sell new hardware to, as the software that comes with the machines is the same.
I work for one of the best companies out there. The management has clue, the policies make sense, I work from home when I can, and the work is challenging, interesting, and ever-present. Three years ago, I would not have believed that a company existed that was this on the ball. Is it perfect? No, but I have not been able to come up with any changes that would make it better.
We would love to hire more people, but it is darn hard - few people have the level of talent and self control needed to do what we do. We know they exist, because we do hire them when we find them, but people with our skill set are rare, even in the DotComImplosion. We also know how hard it is to find them, because we have had open job requisitions since I started. There is more work than people to do it right.
I do have sympathy for the unemployed programmers who were hoping to get by on HTML and Perl. Unemployment is not fun at all. It is, though, a clear sign that it is time to learn new skills. I try to read a new tech book every few months, and at least a few good tech papers every month, and I always feel that I am behind. I am aghast at people I have met who seem to feel that they can get by with only a current set of skills. Especially if those skills are only the ones currently hyped.
So, am I sympathetic? YES! Do I have a pancea? No! The best I can suggest is to stay current, and remember that no job is forever, just like the lack of one.
I went to Harvery Mudd College as a Physicist with Econ and CS concentrations. That I did not have a CS degree hurt me a bit in getting my entry level jobs, but the exposure to lots of computational physics, computational engineering, and just plain solid thought really helped. I learned how to break down problems into core difficulties and crank work, and to remember the important bits, without cluttering the mind with trivia. (My definition - trivia is what can be looked up without breaking workflow. Important knowledge changes how I approach a problem.)
I ended up with an MS in applied mathematics and an MS in Industrial Engineering in later years, which did help, but which were the same kind of toolbox expansion that I got in CS classes.
I would reccomend against being a liberal arts major, but I would take a fair number of liberal arts classes. They do broaden the mind, but it helps to have gotten practical salable skills as well.
A good way to do the above is to get a solid programming class or two at a junior college before heading off to college, then trying to get programming jobs either at the school or elsewhere during your summers. Then, if you find you like computers, you will have solid job skills. If you discover that you hate it, then you will find out in time to pick another life direction while still in college.
Scott
Re:Who really needs a lesson
on
Lawsuits Suck
·
· Score: 1
Please, folks. Look at the populations before being surprised at their actions.
Legislators in America tend to be older men, usually white. Also often lawyers.
The race and gender balance is changing, but I suspect that legislators will always be older lawyers predominantly, as it is hard to get elected if you do not know the people who are, and that takes both time and involvement.
If we want to get our agenda in front of them, we need to work with them where they live. Coming off like a bunch of ill mannered teenagers is not going to endear us to them. We want them to consider us as voters and as interested parties with money to spend, not spoiled children.
This will be true in virtually every country, not just the US, but the methods will differ. Money does talk, but so does simple presence, in the US, as we do have votes every so often. For the US, expressing views to federal, state, and local legislators can make a difference, but it must be done before they have taken a stand, and made promises. Once they have promised, and horse traded with other legislators, they are hard to sway.
If other countries want to run the Internet their way, they likely can, and their own executive powers will do it as they feel their populations demand. After all, US patent law is not nearly as universal as, say, the Bern convention. That said, we are going to see a convergence between different countries, because if one does something that does not work, then they will loose out to others who have a working system.
Therefore, the hegemony of the US lawyers is not that they control the net, but that many others will follow suit. To fix that, try to influence your own authorities to go in the direction that you want, rather than the herd.
Bringing it back to the US, while many congresscritters care only about money, others honestly want what they feel is good for the voters, the district, or the country. They may not agree with us on what that is, but they have that aim. If you, as an individual reader, help present your aims to them in a way they can understand and accept, you will influence them.
Scott
Re:Check Josh Bloch's book, and do not optimize ea
on
Optimizing Java?
·
· Score: 1
Apologies for the complete lack of formatting in my last post. I hit submit rather than preview.
Mea Culpa
Scott
Check Josh Bloch's book, and do not optimize early
on
Optimizing Java?
·
· Score: 1
I have been doing a lot of biostatistics work lately, and have found a combination of the default hprof and the commercial NuMega DevPartner for Java a good pair. Our code usually ran for about an hour and a half per run, so dropping the time by 10% saved us almost 10 minutes.
The others who have said "do not optimize early" are quite correct - it is always worth a quick profile once the code works to see if there are any easy points, but once those are done, it is a hard call. Is it is worth a few days rewriting a method that takes 5% of the time to drop it to 4.95%? Given what a coder costs, likely not.
On the other hand, it is probably worth it to drop a 60% down to 5%, since then you start to see major performance improvements.
As always, algorithmic improvements are probably better anyway. Looking for ways to precalculate things is handy, as it getting calculations out of a loop, but even that might well be done by your jitc.
This does not mean you should code foolishly and hope the jitc gets it. For example, StringBuffers are faster than concatenating strings in all cases, so use them as soon as you have enough plus signs to justify it. I do not usually use them in exceptions, but I do in any main line code with more than one plus sign.
Joshua Bloch, co-author of the Java2 collections classes, has a new book entitled "Effective Java" that is on my must buy list. He may well have some good usage tips.
Scott
Slashdot Mirror
Aside from the advice given by others about calling the financial aid office, do remember that your education is in your own hands.
That sound's snarky, but it really is true - you need to take responsibility for funding your education, which means deciding what you are willing to do. You want to work for FOSS, but are you willing to work for the spooks? How about big oil companies? Big pharma? The military? Once you know that, there are several steps. First, get the FAF/FASF forms filled out as soon as possible. They determine what the school will think you can afford. If that is substantially different from what your parents are willing to pay, you need to know that right now.
For example, if your parents are told to pay $15k a year, and they are only going to pay $10k, you need to find another $5K at least. You can probably do that - there are many loan programs, there are decent summer jobs for programming interns, and at the end of the day, there are ROTC programs. Piles and piles of debt kinda suck, but it is better than giving up on an achievable dream. On the other hand, if you walk away with $120K in debt after four years, you are going to have very limited options - forget FOSS, you will need to take the highest paying job you can find just to stay on top of the payments.
Second, be realistic about where you fit in. Take all the assessment tests you can find, and find the norms for the schools you like. You should want to be out of the top 1% - it makes it hard to learn if you are targeting a very different level of kill than your peers. That said, you might want to find a school where your combined math and verbal is around the top third or top half. You then have many people to learn from, but you will also do fairly well and should not drown.
Third, expect college to be different from high school. Colleges have many people who you can learn from. I do think the MIT education is worth it, but you can watch the lectures for free from Stanford, Berkeley, and MIT off of iTunes U. Try doing the entire MIT algorithms course, and see how it feels. Once you know whether it is pitched at your level, you can write an admissions essay that describes in real terms what excites you, what you have to offer, and why it is worth them taking a risk.
Have you considered a Google Summer of Code project? A good one is not a bad way to get some real code checked in. Other companies have their equivalent.
Scott
I went to Harvey Mudd, and I would rather screen mudders or MIT people than people from, say, USC or UC whatever. Not because of any inherent quality, but because Mudd, MIT, and other institutions of their caliber generally admit very good people, push them hard, and give them access to the very best resources. Further, they are used to being around smart people, so there is less of the lesser nerd bull - you do not have to prove you are a fripping genius with every statement, and thus we can get on to the work.
Knowing that they came from HMC/MIT increases the odds of finding a very bright generalist. Further, I have found that Mudders tend to be aware of their limitations - they do not know everything, and they know it. I have found that people from the Ivies tend to assume their own genius. I would rather someone know what they need to learn, because people that smart should always be learning.
Some people come out of the best schools barely capable of independent thought. Some people come out of middle or lower tier schools sharp as a tack. My job in hiring is to figure out whether a given candidate is one of those very good people. So, will Mudd on your resume get you in the door? No, but it might just get you a phone screen even with an unexceptional resume, because your resume might not be indicative of real skills.
That said, I _have_ hired people from USC, UCI, and other schools. They tended to be unusual for their schools, in that they were generalists, thoughtful, and just plain interesting, but that is unusual in the general population, and no degree is a guarantee of that flexibility.
You would almost certainly find that an MIT education has value, as does an MIT degree, but if you do not get there, try to excel and try to learn as much as you can about as much as you can, as the world does change over the course of your career, and this is a good chance to learn that flexibility.
So, would I pauper myself for an MIT degree? No. I might take out student loans - I had to when I was an undergrad - but I would make sure I had a sustainable amount of debt and a plan to get it paid off. Don't expect a killer salary, but do expect to find something rewarding with a bit of searching. Debt limits your options, just like a ROTC commitment, but if that is what you need to do to get where you want to go, then by all means, do it. The name has some value, but the work you will have to put in to get it, and the people you will get it with has much more value over the long term.
My own buying habits demonstrate how raising prices will result in lower total revenue.
I play music while gaming. Since we may game for four or five hours over a weekend, I like to get two new movie scores every month. Sometimes it was hard to find two that were good in any given month; finding good music was the limiting factor, not money.
As prices climbed to $18-$20 at Tower, I found I was only buying one a month, or one every two months. This is not a moral stand; I found that the albums cost more than I was willing to pay. The samples did not say "drop $18 on me".
Just after the ITMS opened, I went back to buying albums, mostly at the music store, but often on CD as stores were dropping their prices. I bought one or two a month steadily for over a year. Easy enough to see by looking at the creation date for the music.
I have not bought a new score on CD for the last three months, as the prices have climbed back to $18 at Tower. I have three I am considering at the ITMS, but I note with some dismay that many are not available on a per-song basis.
Again, this is not a moral objection, but a practical one. The Sahara DVD I just bought cost me under $20 with a pre-order. I am not willing to pay $18 of that for a CD, and ITMS is not an option. (The soundtrack is there as an album only, and the score is not to be found.) I find that the music I am hearing is not worth the price they want me to pay. I will live with my 600 albums, and nearly a thousand ITMS tracks until the prices drop again.
I must say, if you own stock in a record company, start selling. There is only one bright spot in the marketing reports from every major label, and that is the ITMS. Killing it by raising the prices for the tracks worth having will just cut revenue.
I am living proof that if they start raising the prices, the product will not sell. Period.
Scott
Apple is very confident.
The people working on the Intel transition are aware of the risks, and are addressing them like the consummate professionals they are. I talked with a lot of them at WWDC 2005, and they know what they were doing, and they know what roadblocks could come up and bite them.
You cannot repair or replace a G4/1.5 powerbook motherboard with any processor Intel ships. By having supply through 2008, they can satisfy any Applecare requirements they may have. This alone would justify their contract with freescale. After all, Applecare is three years, and they are going to be selling laptops with G4 processors until early 2006.
Further, depending on Intel's price points, there may be a place for G4 iBooks or Mac Minis for a few months after introduction. Apple currently sells laptops and minis ranging from $500 to $3000, which is a very big range. It is not clear to me just how quickly they will change every segment of their laptop and mini offerings. Certainly, the high end laptops will change fast, as Apple is getting creamed in that market. For a mini, though, where battery life is not a critical factor, but price is?
The only concrete statement we have comes from WWDC - they will have something running by WWDC 2006, and will have completed their transition by the end of 2007. That is a two year range, in which every machine must change. I am betting that the G4 will vanish overnight, but I am not going to bet the farm, and having an insured supply of chips means that Apple is not going to either.
It takes confidence to make a change this drastic. I, for one, am in favor of it.
Scott
This is how it should be.
Open source has two benefits: customers of the product have access to the source, and a wider community can read, change, and improve the source. Announcements like the Jasper ones force the community to decide where they stand.
Put another way: if the lead devs decide to move, and get paid for their work, then we find out whether the project was robust or fragile. If the community does not step up to the plate, then they did not care enough.
To me, that is just fine. It makes it clear where we put our time and treasure. Projects that fail for this reason were fragile, depending on the good will of one person.
Scott
When it has been fought out, the reporter usually lost, unless that which they were reporting was in some way illegal. Whistle blower laws then usually protected the reporter and the sources.
This is as it should be - the source in this case stole proprietary trade secrets. The reporter is publicizing proprietary information. If the reporter does not list sources, then the fault quickly becomes his.
This is up there with me breaking into your house, stealing your credit card number, and then the rumor sites claiming that they do not have to help the police, because they did not do it.
Reporters are responsible for what they print, and the people do NOT have an intrinsic "right to know". The people may have a right to know any given fact, and we need to protect the ability of the press to publicize the things that are important, but they do not have a God, or Constitutional, right to tell any damn thing without consequences.
When someone breaks the law, the press does not get attorney-client privilege.
Scott
Java is what I do for a living, and I use a TiBook/667 to do it. It is a pretty good place to work.
I find Circus Ponies Notebook and BareBones BBEdit to be real helps to Eclipse/IDEA and the command line tools I use for most work.
Do be aware that releases often lag a bit, though, so if you absolutely need to use Java 1.5 as soon as the beta hits the street, you will need a PC or VPC. I choose to run VPC to test my 1.5 beta apps, and to run the rest under MacOS X.
Scott
Bat puckey.
Microsoft clearly makes money from "number 2" by making Mac office. Thus, they _can_ answer the question "what do we get?" with the words "cold, hard cash, and yet another spike in the coffin of potentially competing office products."
Think about it - were MS to stop shipping office for Mac, Apple would have to come up with an office suite of their own. They would probably fail, but MS really does not want to take that chance, given that Apple has enough cash to make it happen.
Why would IBM want to use a processor architecture other than Intel? Because they make oodles of cash on IBM servers, and they do not want to share with Intel.
Why do people want to buy Macs? Because they are better, and thus I make more money as a consultant. (You do not have to agree, but this is why I use it, and why my company buys them.)
In all three cases, the answer is "because we make money by doing so."
Contrast this with Real. They are known for dreadful software, spyware, and relatively poor quality. They have addressed some of these issues, but i know very few people who are fond of Real software. Thus - what would Apple get out of this partnership?
The work in making a clone or custom mac is not in the design of the cpu, but in getting all the rest of the hardware integrated with the OS. Add in the thrill of device drivers, and potentially having to reverse engineer any custom Apple HW, and you get a pretty much clone-free market. Opening up the cpu will not change that.
It does mean that the architecture might be used in more places than it now is. Next generation video games might move to the Power architecture if they see a benefit. Similarly, good Linux offerings will make scientists consider clustering PPC boxes rather than x86 boxes for high performance numerical computing.
If these events happen, then new support, and new understanding of the architecture will hit the streets in a way that will benefit Apple. For example, if serious money goes into optimizing gcc by someone other than Apple, releases of OS X will get better without Apple effort. Similarly, more money from other hardware purchasers gives IBM incentive to advance the architecture.
Scott
The battery can be replaced, and people do.
Putting in a battery access door to make it trivial to replace takes room and makes the device more prone to failure. (The seals around doors are never great, unfortunately.) As it is, based on the number of people in the class action suit, battery failure affected a fairly small number of units; it would not surprise me if one whose battery can be replaced easily had a higher failure rate.
The battery is probably as replacable as a standard iPod - about a hundred from Apple, about two thirds that from other vendors, and about half that if you can do it yourself. If it worries you, get Applecare.
For what it is worth, my wife's 5G iPod bought when they had only been out a month is still going strong, and my 20G bought a week after they showed up at the Apple store is doing well too.
Depends on the job. If your job is to keep the machines up without ever talking to a human, sure! Be a condescending ass if you like.
If, on the other hand, your job is like every job I have had - solving technical problems to produce whatever the organization produces - then you do need social skills. More to the point, you need to have people trust you. If you are good at technical things, then you probably know more about them than the decision makers. They are not going to absorb your knowledge in an hour of mentoring, so they will still be making their decision about technical issues based on whether they trust you.
Further, the claim "if I'm condescending its because a lot of people who're above me are there not because they're better than me but because they have the "Oh so called Social Skills." is nonsense. If the people above you are really utter loads, find a new job. If, on the other hand, they have better social skills than you have shown, it may be that they are above you because people trust them.
Ask yourself this question. If your accountant's response to "why did you deduct $4286 as charity this year?" was "I am the accountant, and you are the clueless loser. Don't ask stupid questions.", would you feel happy? More to the point, would you keep trusting him with your finances?
I wouldn't.
Finally, when you stop respecting people, you stop listening to them. In many cases, you are right about what they want, and thus not listening is the right answer, but you will miss things, and you will never know that you missed them. THe people you are filled with contempt for _do_ know when you are writing code/emails/shell scripts during meetings, and are not filled with trust later.
People who do not trust you are not likely to promote you. Blame social skills if you want, but it sure seems likely that they have a real reason.
I note that our consulting group recently extended machine lifetime to three years, while at most of our clients, it has been pushed out to five or more years.
Perhaps this is not the pentium 90 era, but it does show the importance of working well on older hardware. Apple may drop support for some older machines, but if each release gets faster, then they get a steady stream of revenue from users who have not bought new hardware AND they get users with those older machines up to date on software. Up to date users are a lot easier to sell new hardware to, as the software that comes with the machines is the same.
Scott
We would love to hire more people, but it is darn hard - few people have the level of talent and self control needed to do what we do. We know they exist, because we do hire them when we find them, but people with our skill set are rare, even in the DotComImplosion. We also know how hard it is to find them, because we have had open job requisitions since I started. There is more work than people to do it right.
I do have sympathy for the unemployed programmers who were hoping to get by on HTML and Perl. Unemployment is not fun at all. It is, though, a clear sign that it is time to learn new skills. I try to read a new tech book every few months, and at least a few good tech papers every month, and I always feel that I am behind. I am aghast at people I have met who seem to feel that they can get by with only a current set of skills. Especially if those skills are only the ones currently hyped.
So, am I sympathetic? YES! Do I have a pancea? No! The best I can suggest is to stay current, and remember that no job is forever, just like the lack of one.
Scott
I went to Harvery Mudd College as a Physicist with Econ and CS concentrations. That I did not have a CS degree hurt me a bit in getting my entry level jobs, but the exposure to lots of computational physics, computational engineering, and just plain solid thought really helped. I learned how to break down problems into core difficulties and crank work, and to remember the important bits, without cluttering the mind with trivia. (My definition - trivia is what can be looked up without breaking workflow. Important knowledge changes how I approach a problem.) I ended up with an MS in applied mathematics and an MS in Industrial Engineering in later years, which did help, but which were the same kind of toolbox expansion that I got in CS classes. I would reccomend against being a liberal arts major, but I would take a fair number of liberal arts classes. They do broaden the mind, but it helps to have gotten practical salable skills as well. A good way to do the above is to get a solid programming class or two at a junior college before heading off to college, then trying to get programming jobs either at the school or elsewhere during your summers. Then, if you find you like computers, you will have solid job skills. If you discover that you hate it, then you will find out in time to pick another life direction while still in college. Scott
Legislators in America tend to be older men, usually white. Also often lawyers.
The race and gender balance is changing, but I suspect that legislators will always be older lawyers predominantly, as it is hard to get elected if you do not know the people who are, and that takes both time and involvement.
If we want to get our agenda in front of them, we need to work with them where they live. Coming off like a bunch of ill mannered teenagers is not going to endear us to them. We want them to consider us as voters and as interested parties with money to spend, not spoiled children.
This will be true in virtually every country, not just the US, but the methods will differ. Money does talk, but so does simple presence, in the US, as we do have votes every so often. For the US, expressing views to federal, state, and local legislators can make a difference, but it must be done before they have taken a stand, and made promises. Once they have promised, and horse traded with other legislators, they are hard to sway.
If other countries want to run the Internet their way, they likely can, and their own executive powers will do it as they feel their populations demand. After all, US patent law is not nearly as universal as, say, the Bern convention. That said, we are going to see a convergence between different countries, because if one does something that does not work, then they will loose out to others who have a working system.
Therefore, the hegemony of the US lawyers is not that they control the net, but that many others will follow suit. To fix that, try to influence your own authorities to go in the direction that you want, rather than the herd.
Bringing it back to the US, while many congresscritters care only about money, others honestly want what they feel is good for the voters, the district, or the country. They may not agree with us on what that is, but they have that aim. If you, as an individual reader, help present your aims to them in a way they can understand and accept, you will influence them.
Scott
Mea Culpa
Scott
I have been doing a lot of biostatistics work lately, and have found a combination of the default hprof and the commercial NuMega DevPartner for Java a good pair. Our code usually ran for about an hour and a half per run, so dropping the time by 10% saved us almost 10 minutes. The others who have said "do not optimize early" are quite correct - it is always worth a quick profile once the code works to see if there are any easy points, but once those are done, it is a hard call. Is it is worth a few days rewriting a method that takes 5% of the time to drop it to 4.95%? Given what a coder costs, likely not. On the other hand, it is probably worth it to drop a 60% down to 5%, since then you start to see major performance improvements. As always, algorithmic improvements are probably better anyway. Looking for ways to precalculate things is handy, as it getting calculations out of a loop, but even that might well be done by your jitc. This does not mean you should code foolishly and hope the jitc gets it. For example, StringBuffers are faster than concatenating strings in all cases, so use them as soon as you have enough plus signs to justify it. I do not usually use them in exceptions, but I do in any main line code with more than one plus sign. Joshua Bloch, co-author of the Java2 collections classes, has a new book entitled "Effective Java" that is on my must buy list. He may well have some good usage tips. Scott