High Frequency Traders (HFTs) are not investors, they are market makers. They find a willing buyer and a willing seller, arrange the transaction, and execute the trade. They make a profit on the spread between the buy price and the sell price.
Except that a neutral third party (the exchange) could connect the willing buyer and willing seller who both are willing to perform the transaction at higher prices, and split the difference. i.e. the Seller is selling for $1.05 and the buyer is willing to pay $1.06. HFT makes money by buying from the seller, selling to the buyer, and pocketing the $0.01 (minus expenses and trading costs.)
But wouldn't the buyer and seller be better off if the exchange, who is taking a transaction fee to perform the service, closed the transaction at $1.055? And if the buyer and seller are serious about wanting to make the trade, requiring the offer to be valid for a full second (for example) would give the market plenty of time to guarantee that the trade occurs.
Since the introduction of high frequency trading, transaction costs have fallen considerably, saving plenty of people a lot of money
It is not clear if this is a case of cause or effect. Without cheap transactions, HFT doesn't make sense. (And if the number of transactions were an order of magnitude smaller without the HFT noise, the cost of the infrastructure to execute the trades would be dramatically cheaper as well, which would reduce the cost of trades.)
But the fact remains that HFT acts as a hidden transaction fee on every trade. All of the money they make comes at the expense of someone executing an order.
We have about 10,000 firearm homicides in the US per year (and similar numbers of suicides), compared to 600 accidental deaths. Your "1/3 of deaths being an accident" is seriously miscalibrated.
Overall suicide rates have not changed in countries that banned firearms (Australia), so those probably wouldn't change even if guns are banned.
But if were into advocacy for any controversial subject, it would be cool to have an account on G+ that I could post everything publicly. (That being the whole POINT of advocacy vs. limited postings for my friends.)
Blocking my cow-orkers only deals with half the equation.
And I'd still have to worry about who's subscribing, and if it's someone's 3rd cousin, etc. (Not to mention people re-sharing with my name attached, etc.)
I'd be happy to pay money for it. Or put up a bond. Or whatever. But as long as it handles the case where someone who's being stalked, abused, persecuted or prosecuted can set up an account for free and some people can elect to see it, I'd be satisfied.
I'm firmly atheist. I don't choose to participate in skepticism advocacy, but if I choose to, I might well prefer a pseudonym. (There are several fundamentalists in my management chain at work.)
I know of several people who are involved in the burner and pagan communities, who keep all of their non-mundane activities under pseudonyms. I don't currently know anyone involved in the SCA, but in the past I understood that many people didn't mix their role and real life.
I've been told that the BDSM community uses nicknames almost exclusively.
Many actors, musicians, authors, etc. work under pseudonyms, and would probably prefer not to mix their personal and public identities.
People being stalked (in real life or online) might have something to say, but not want to post under their real name.
Activists in the middle east and china certain prefer not to be forced to post under their real name.
Pretty much any individual or community that is subject to personal, social, legal, or political harassment may have motivation to operate in a public space, but use pseudonyms.
How about having a couple of flags, for "anonymous accounts" and "pseudonymous" accounts (the latter being "google knows my name and has verified it as much as anything else, but it isn't posted associated with this account). And indicate anony/psuedo accounts at the top of the profile screen. Add a security setting to block them entirely.
I have read about an equipment recall on aircraft due to errors caused by cosmic radiation at altitude causing problems with a specific sensor.
But I agree with you that this would much more likely cause unintended stops or stalls, or engine misfires, or computer resets, rather than just unintended acceleration on hundreds of vehicles.
This may be due to a failure mode, but it will likely be due to something like corrosion of a sensor or bus (as suggested by one professor) rather than single bit errors.
I assume from your description that the company isn't likely to go public any time soon, and that it is probably not likely to be acquired. (You'll want the agreement to cover those circumstances, but I don't think that's your major focus.) So ignore everyone above talking about going public, options, etc.
I'll bet that the company has a half-dozen to a dozen employees, and basically the owner / CEO is offering you a share of the company that they think you've earned over the last 6 years. With a half-dozen employees, figure his profits go down by 20% if you leave. (Until / unless he can successfully replace you.) And if you're good, and perhaps competing with them, it might hurt even more.
So basically, this sounds to me like you're becoming a minor partner in the firm. That's not a bad deal in theory. You get to be the owner of a consulting firm without having to deal with starting it from scratch, finding accountants, setting up accounts receivable, payroll, etc.
So follow up with a lawyer and a tax accountant to verify the deal is legit. And ask for company financials to get an idea of what 10% of the company will be worth on an ongoing basis. (i.e. how much would dividend you'd have been paid the last 2 or 3 years.)
So I'm not sure how pre-teen you're looking but some authentic juvenile fiction:
Timothy Zahn has a new series (currently on book 5 of 6?) called "Dragon and Thief" (and each sequel is "Dragon and _____".) Lots of interesting stuff, and they have study and discussion guides at the end if you care.
Phillip Pullman is good stuff.
We've really enjoyed the Cornelia Funke, and a movie is coming out soon.
As a kid in 4th-6th grade, I devoured the Dark is Rising books. I need to re-read them.
Ditto the Madeline L'engle books.
The Dragon Drums series by Anne McCaffery is targetted at a younger audience. Some of the other novels in that series (and her other series) have more adult themes. (Not that this is bad, but might be a bit much for a 9 or 10 year old.)
However, if the cancellation fees are pro-rated for the term of the contract, then the carrier isn't hurt by subsidizing the phone. However, most carriers today charge you the full fee if you cancel 23 months into a 24 month contract.
Also, the FCC could restrict the cancellation fees to when you DO get a new phone. Extending your contract, even on a "special rate" shouldn't extend a cancellation fee.
At any accredited university, you get a slew of general requirements. In 12 quarters, you had to finish 6 humanities (English, Languages, etc.) and 6 social sciences (History, Sociology, etc.) classes at my engineering college. So you will take a couple of upper level reading classes, some anthropology, and maybe a foreign language.
It is just that you aren't going to have tons of free elective hours, as they'll be full of math, science, and your major area work.
First, I assume that this person is talking about an engineering or science focused school, such as an MIT, Berkeley, Georgia Tech, NC State, etc. These are institutions that focus on research and granting advanced degrees. They are not "technical trade schools".
And maybe this is what differentiates real engineering schools from "liberal arts" schools. Most engineering programs, and most CS programs associated with engineering schools, have only suffered from a little bit of grade inflation over the last 20 years. And a BS from an MIT (or even a Georgia Tech or NC State) and a bit of experience/demonstrated clue is plenty to get you a decent job.
If you get a BS or BA in CS from a non-engineering school, you'll be under a bit more pressure to prove you actually learned something in your CS classes in an interview, and that you can do something.
If you get a degree from a "Vocational" tech school, I'd seriously quiz you to see if you actually understand the principles of CS. While some sharp people attend these schools, I generally assume that most top-tier people will be attending a "real college".
You will probably get a significantly better CS education from an Engineering school, and you'll probably have more options to specialize in what interests you. You'll also have a larger group of fellow students, which may help you in your future career.
And Engineering schools (MIT, Georgia Tech, UIUC, Texas A&M) do have a healthy dose of "liberal arts".
However, most Engineering schools have relatively limited alternatives for other degree programs. That means that you won't be able to easily transfer to another degree program if you don't want to do something technical.
When a High School honor student had bad grades at my Alma Mater (Georgia Tech) and wanted to transfer to a liberal arts school, one of the advisors had a letter that went something like this:
"This students was an excellent performer in high school and would have likely been accepted to your school. [Long section about how Georgia Tech gives lower grades than most schools and that it is a very intense program that just doesn't fit some students.] In summary, I feel that this student would perform better in a more humane university environment."
So, if you want the best CS degree possible, surrounded by the sharpest people doing leading edge research in the field, go to the engineering school. If you want a more humane university environment, go to the liberal arts school.
How about a bench next to the sidewalk in someone's front yard?
It is, certainly, private property, but it is not at all clear by the placement that owner does not intend the public to use it.
Further, the owner is harmed by the temporary use of a bench about as much as the use of her WAP access point.
Of course sleeping on the bench overnight is pretty rude, as is damaging the bench, but taking a rest in the middle of a walk around the neighborhood doesn't cause the least bit of harm.
Assuming you're not discounting the wholesale price to other retailers, and assuming that the preliminary buzz is reasonable, I suspect Gamestop senior management isn't very picky about when they get paid. (i.e. at the sale or in advance.)
Of course, you can't discount the price to other retailers, because then they could sell it for $2 over wholesale, and suddenly Gamestop won't sell any...
And you'd have to check with a lawyer to see if structuring the deal in any specific way causes collusion / anti-trust problems..
Actually dams generally sequester a great deal of carbon under the water, when they flood out the trees. There was a great article on a new robot that can harvest the underwater forests of the world for lumber.
Mind you, any tree doesn't really sequester carbon long term, because they WILL die, and then be burned or decompose.
I don't see how various people justify "carbon credits" for trees unless they can guarantee (through a trust, etc.) that the land will never be de-forested again.
If you go with a pure hand-counting system, you put each vote on it's own piece of paper.
You sort the ballots into races, and then sort each race into a stack for each candidate.
Then you count and re-count the stack. Some countries use bank tellers, who, in those countries, get the day off. And they are, of course, very experienced in counting the number of little bits of paper and don't make a ton of money.
The optical scan machines are probably worth it, although I'd resist the temptation to have them be programmed to understand which races are which, and to automatically report things. The more configuration you have, the more opportunities to corrupt the process.
But I recall a case where mis-calibration of the optical scan machines caused a number of SAT scores to be reported in error. I think I would be willing to accept a very small error rate by human counters (keeping in mind that the error rate on handling money is VERY low), to prevent occasional large-scale mechanical failures.
Let us say that a person making $10 / hour can count 1000 votes an hour. That's one cent per vote counted.
Let us assume that a person can enter one vote in 20 seconds on a voting machine. Let us assume that voting machines are busy 10 hours on voting day. Each voting machine will "count" 1800 votes in a day. So for $20, you can count more votes than the voting machine.
If each voting machine costs $400, it will take 20 elections to recoup your investment. And while there are multiple elections a year, you have to buy enough machines to handle the presidential elections every 4 years. Most elections (i.e. primaries, run-offs, etc.), the machines will be significantly underutilized, so so only register a few dozen or hundred votes on average.
It may not be quite as fast as a mechanical system, but it certainly would be a heck of a lot cheaper.
Optical scan, i.e. standardized test style, isn't a bad way to have a machine count ballots, and leaves a paper trail, and is cheaper than the video machines (since you can use one machine for all of the votes), but probably isn't really that much better.
Google is set up so that only the founders have much control over the board and direction of the company. They don't report to the common shareholders, and they're all billionaires, so hopefully won't be primarily motivated by money.
No guarantees, but they have better odds than some other companies of maintaining 'do no evil'.
One more good article, with references, that talks about 100:1, 22:1, and the fact that 30% of the programmers on a typical product are "net negative producing programmers":
Steve McConnell's _Rapid_Development_, p 12, cites 7 academic studies claiming 10:1 productivity measures. Variations of teams vary by 3-5:1.
I've personally helped manage someone very senior who was consistently produced 1/5th of what I would. We'd schedule a problem for 2 days for me, or 2 weeks for him.
And as Joel pointed out in the first article, the problem isn't just that lesser programmers aren't as productive, but that there are some things that they just can't do.
I'm firmly convinced that many of the less productive programmers wash out relatively early in their career, before they start hitting system architecture and other challenges that greatly exceed their abilities.
In the programming business, the top 10% of the programmers are about 10x as productive as the middle 50%. The middle 50% are 10x as productive as the bottom 10%. Experience plays a big factor in productivity, but if you aren't in the top 50%, 12 years of experience won't make you as good as a person with 2 years of experience who's top 10%.
At some point, if your salary requirements increase much at all, you're priced out of the market.
And if you're at the lower end of the productivity spectrum, it can cost more to manage and support you than you produce.
However, you may just need a more challenging job, and an environment where you can advance. Getting too comfortable happens all too frequently.
Slashdot Mirror
High Frequency Traders (HFTs) are not investors, they are market makers. They find a willing buyer and a willing seller, arrange the transaction, and execute the trade. They make a profit on the spread between the buy price and the sell price.
Except that a neutral third party (the exchange) could connect the willing buyer and willing seller who both are willing to perform the transaction at higher prices, and split the difference. i.e. the Seller is selling for $1.05 and the buyer is willing to pay $1.06. HFT makes money by buying from the seller, selling to the buyer, and pocketing the $0.01 (minus expenses and trading costs.)
But wouldn't the buyer and seller be better off if the exchange, who is taking a transaction fee to perform the service, closed the transaction at $1.055? And if the buyer and seller are serious about wanting to make the trade, requiring the offer to be valid for a full second (for example) would give the market plenty of time to guarantee that the trade occurs.
Since the introduction of high frequency trading, transaction costs have fallen considerably, saving plenty of people a lot of money
It is not clear if this is a case of cause or effect. Without cheap transactions, HFT doesn't make sense. (And if the number of transactions were an order of magnitude smaller without the HFT noise, the cost of the infrastructure to execute the trades would be dramatically cheaper as well, which would reduce the cost of trades.)
But the fact remains that HFT acts as a hidden transaction fee on every trade. All of the money they make comes at the expense of someone executing an order.
However, you can't mold honeycombs, foams, or other objects with complex internal structures that might be more efficient.
Then it's just a question of comparing the costs of industrial printing a part vs. the other ways you can improve vehicle efficiency.
We have about 10,000 firearm homicides in the US per year (and similar numbers of suicides), compared to 600 accidental deaths. Your "1/3 of deaths being an accident" is seriously miscalibrated.
Overall suicide rates have not changed in countries that banned firearms (Australia), so those probably wouldn't change even if guns are banned.
But if were into advocacy for any controversial subject, it would be cool to have an account on G+ that I could post everything publicly. (That being the whole POINT of advocacy vs. limited postings for my friends.)
Blocking my cow-orkers only deals with half the equation.
And I'd still have to worry about who's subscribing, and if it's someone's 3rd cousin, etc. (Not to mention people re-sharing with my name attached, etc.)
I'd be happy to pay money for it. Or put up a bond. Or whatever. But as long as it handles the case where someone who's being stalked, abused, persecuted or prosecuted can set up an account for free and some people can elect to see it, I'd be satisfied.
I'm firmly atheist. I don't choose to participate in skepticism advocacy, but if I choose to, I might well prefer a pseudonym. (There are several fundamentalists in my management chain at work.)
I know of several people who are involved in the burner and pagan communities, who keep all of their non-mundane activities under pseudonyms. I don't currently know anyone involved in the SCA, but in the past I understood that many people didn't mix their role and real life.
I've been told that the BDSM community uses nicknames almost exclusively.
Many actors, musicians, authors, etc. work under pseudonyms, and would probably prefer not to mix their personal and public identities.
People being stalked (in real life or online) might have something to say, but not want to post under their real name.
Activists in the middle east and china certain prefer not to be forced to post under their real name.
Pretty much any individual or community that is subject to personal, social, legal, or political harassment may have motivation to operate in a public space, but use pseudonyms.
How about having a couple of flags, for "anonymous accounts" and "pseudonymous" accounts (the latter being "google knows my name and has verified it as much as anything else, but it isn't posted associated with this account). And indicate anony/psuedo accounts at the top of the profile screen. Add a security setting to block them entirely.
I have read about an equipment recall on aircraft due to errors caused by cosmic radiation at altitude causing problems with a specific sensor.
But I agree with you that this would much more likely cause unintended stops or stalls, or engine misfires, or computer resets, rather than just unintended acceleration on hundreds of vehicles.
This may be due to a failure mode, but it will likely be due to something like corrosion of a sensor or bus (as suggested by one professor) rather than single bit errors.
I assume from your description that the company isn't likely to go public any time soon, and that it is probably not likely to be acquired. (You'll want the agreement to cover those circumstances, but I don't think that's your major focus.) So ignore everyone above talking about going public, options, etc.
I'll bet that the company has a half-dozen to a dozen employees, and basically the owner / CEO is offering you a share of the company that they think you've earned over the last 6 years. With a half-dozen employees, figure his profits go down by 20% if you leave. (Until / unless he can successfully replace you.) And if you're good, and perhaps competing with them, it might hurt even more.
So basically, this sounds to me like you're becoming a minor partner in the firm. That's not a bad deal in theory. You get to be the owner of a consulting firm without having to deal with starting it from scratch, finding accountants, setting up accounts receivable, payroll, etc.
So follow up with a lawyer and a tax accountant to verify the deal is legit. And ask for company financials to get an idea of what 10% of the company will be worth on an ongoing basis. (i.e. how much would dividend you'd have been paid the last 2 or 3 years.)
But if you like the work, go for it!
So I'm not sure how pre-teen you're looking but some authentic juvenile fiction:
Timothy Zahn has a new series (currently on book 5 of 6?) called "Dragon and Thief" (and each sequel is "Dragon and _____".) Lots of interesting stuff, and they have study and discussion guides at the end if you care.
Phillip Pullman is good stuff.
We've really enjoyed the Cornelia Funke, and a movie is coming out soon.
As a kid in 4th-6th grade, I devoured the Dark is Rising books. I need to re-read them.
Ditto the Madeline L'engle books.
The Dragon Drums series by Anne McCaffery is targetted at a younger audience. Some of the other novels in that series (and her other series) have more adult themes. (Not that this is bad, but might be a bit much for a 9 or 10 year old.)
However, if the cancellation fees are pro-rated for the term of the contract, then the carrier isn't hurt by subsidizing the phone. However, most carriers today charge you the full fee if you cancel 23 months into a 24 month contract.
Also, the FCC could restrict the cancellation fees to when you DO get a new phone. Extending your contract, even on a "special rate" shouldn't extend a cancellation fee.
If you are a sys-admin, planning on sabotaging their systems, would you do it before or after turning in your notice?
If you wanted a phone list of the best employees, would you get it before or after turning in your notice?
If you're firing someone, or even in the midst of a layoff, it might make sense. But if someone gives notice, that's a completely different situation.
Here's another vote for PC-Lint by http://www.gimpel.com/
You really can't beat it for the money, and it is probably as comprehensive as some of the other more expensive products for C and C++.
At any accredited university, you get a slew of general requirements. In 12 quarters, you had to finish 6 humanities (English, Languages, etc.) and 6 social sciences (History, Sociology, etc.) classes at my engineering college. So you will take a couple of upper level reading classes, some anthropology, and maybe a foreign language.
It is just that you aren't going to have tons of free elective hours, as they'll be full of math, science, and your major area work.
Bachelors in EE : Masters in Education :: Masters in Education : Working at McDonalds.
First, I assume that this person is talking about an engineering or science focused school, such as an MIT, Berkeley, Georgia Tech, NC State, etc. These are institutions that focus on research and granting advanced degrees. They are not "technical trade schools".
And maybe this is what differentiates real engineering schools from "liberal arts" schools. Most engineering programs, and most CS programs associated with engineering schools, have only suffered from a little bit of grade inflation over the last 20 years. And a BS from an MIT (or even a Georgia Tech or NC State) and a bit of experience/demonstrated clue is plenty to get you a decent job.
If you get a BS or BA in CS from a non-engineering school, you'll be under a bit more pressure to prove you actually learned something in your CS classes in an interview, and that you can do something.
If you get a degree from a "Vocational" tech school, I'd seriously quiz you to see if you actually understand the principles of CS. While some sharp people attend these schools, I generally assume that most top-tier people will be attending a "real college".
You will probably get a significantly better CS education from an Engineering school, and you'll probably have more options to specialize in what interests you. You'll also have a larger group of fellow students, which may help you in your future career.
And Engineering schools (MIT, Georgia Tech, UIUC, Texas A&M) do have a healthy dose of "liberal arts".
However, most Engineering schools have relatively limited alternatives for other degree programs. That means that you won't be able to easily transfer to another degree program if you don't want to do something technical.
When a High School honor student had bad grades at my Alma Mater (Georgia Tech) and wanted to transfer to a liberal arts school, one of the advisors had a letter that went something like this:
"This students was an excellent performer in high school and would have likely been accepted to your school. [Long section about how Georgia Tech gives lower grades than most schools and that it is a very intense program that just doesn't fit some students.] In summary, I feel that this student would perform better in a more humane university environment."
So, if you want the best CS degree possible, surrounded by the sharpest people doing leading edge research in the field, go to the engineering school. If you want a more humane university environment, go to the liberal arts school.
How about a bench next to the sidewalk in someone's front yard?
It is, certainly, private property, but it is not at all clear by the placement that owner does not intend the public to use it.
Further, the owner is harmed by the temporary use of a bench about as much as the use of her WAP access point.
Of course sleeping on the bench overnight is pretty rude, as is damaging the bench, but taking a rest in the middle of a walk around the neighborhood doesn't cause the least bit of harm.
Assuming you're not discounting the wholesale price to other retailers, and assuming that the preliminary buzz is reasonable, I suspect Gamestop senior management isn't very picky about when they get paid. (i.e. at the sale or in advance.)
Of course, you can't discount the price to other retailers, because then they could sell it for $2 over wholesale, and suddenly Gamestop won't sell any...
And you'd have to check with a lawyer to see if structuring the deal in any specific way causes collusion / anti-trust problems..
Speaking as a fellow Computer Science graduate, any field that has "Science" in the name isn't..
Actually dams generally sequester a great deal of carbon under the water, when they flood out the trees. There was a great article on a new robot that can harvest the underwater forests of the world for lumber.
Mind you, any tree doesn't really sequester carbon long term, because they WILL die, and then be burned or decompose.
I don't see how various people justify "carbon credits" for trees unless they can guarantee (through a trust, etc.) that the land will never be de-forested again.
If you go with a pure hand-counting system, you put each vote on it's own piece of paper.
You sort the ballots into races, and then sort each race into a stack for each candidate.
Then you count and re-count the stack. Some countries use bank tellers, who, in those countries, get the day off. And they are, of course, very experienced in counting the number of little bits of paper and don't make a ton of money.
The optical scan machines are probably worth it, although I'd resist the temptation to have them be programmed to understand which races are which, and to automatically report things. The more configuration you have, the more opportunities to corrupt the process.
But I recall a case where mis-calibration of the optical scan machines caused a number of SAT scores to be reported in error. I think I would be willing to accept a very small error rate by human counters (keeping in mind that the error rate on handling money is VERY low), to prevent occasional large-scale mechanical failures.
Let us say that a person making $10 / hour can count 1000 votes an hour. That's one cent per vote counted.
Let us assume that a person can enter one vote in 20 seconds on a voting machine. Let us assume that voting machines are busy 10 hours on voting day. Each voting machine will "count" 1800 votes in a day. So for $20, you can count more votes than the voting machine.
If each voting machine costs $400, it will take 20 elections to recoup your investment. And while there are multiple elections a year, you have to buy enough machines to handle the presidential elections every 4 years. Most elections (i.e. primaries, run-offs, etc.), the machines will be significantly underutilized, so so only register a few dozen or hundred votes on average.
It may not be quite as fast as a mechanical system, but it certainly would be a heck of a lot cheaper.
Optical scan, i.e. standardized test style, isn't a bad way to have a machine count ballots, and leaves a paper trail, and is cheaper than the video machines (since you can use one machine for all of the votes), but probably isn't really that much better.
If you're only making $5000 a year, you should qualify for the Earned Income Tax Credit, which will offset most of the social security taxes.
Google is set up so that only the founders have much control over the board and direction of the company. They don't report to the common shareholders, and they're all billionaires, so hopefully won't be primarily motivated by money.
No guarantees, but they have better odds than some other companies of maintaining 'do no evil'.
One more good article, with references, that talks about 100:1, 22:1, and the fact that 30% of the programmers on a typical product are "net negative producing programmers":
h told.html
http://www.stsc.hill.af.mil/crosstalk/2000/05/bec
Here's one source of data that shows a 5:1 to 10:1 difference in productivity for programmers, even after correcting for quality:
h tml
- productivity/
http://www.joelonsoftware.com/articles/HighNotes.
This guy has some data that show 2-3x difference, up to 22x depending on how you interpret the data.
http://www.webfoot.com/blog/2006/12/07/programmer
Steve McConnell's _Rapid_Development_, p 12, cites 7 academic studies claiming 10:1 productivity measures. Variations of teams vary by 3-5:1.
I've personally helped manage someone very senior who was consistently produced 1/5th of what I would. We'd schedule a problem for 2 days for me, or 2 weeks for him.
And as Joel pointed out in the first article, the problem isn't just that lesser programmers aren't as productive, but that there are some things that they just can't do.
I'm firmly convinced that many of the less productive programmers wash out relatively early in their career, before they start hitting system architecture and other challenges that greatly exceed their abilities.
In the programming business, the top 10% of the programmers are about 10x as productive as the middle 50%. The middle 50% are 10x as productive as the bottom 10%. Experience plays a big factor in productivity, but if you aren't in the top 50%, 12 years of experience won't make you as good as a person with 2 years of experience who's top 10%.
At some point, if your salary requirements increase much at all, you're priced out of the market.
And if you're at the lower end of the productivity spectrum, it can cost more to manage and support you than you produce.
However, you may just need a more challenging job, and an environment where you can advance. Getting too comfortable happens all too frequently.