Hundreds of apps doesn't make me buy. I buy because someone communicates to me the ONE (or few) apps
Everybody does. And everybody has different apps that matter to them. That's why having lots of apps matters.
And let me tell you - it's a vicious cycle. If the phone isn't attractive to mainstream, developers won't develop mainstream apps for it, and mainstream won't buy it.
Ah, yes, and Linux will never work because nobody will develop software for it, right? Current phones (including the iPhone) come with so little software that is so limited that the bar is really low. Most of the so-called mainstream developers are fixing bugs and omissions in the base OS, something OpenMoko doesn't need.
OpenMoko costs $450/$600. You can get a Symbian/WinMobile smart phone with open API for less than that.
OpenMoko costs $300 with a 640x480 screen and GPS (the $450 and $600 include development hardware, something that costs thousands of dollars from other vendors). There is no Symbian or WinMobile that comes even close. In fact, the only other 640x480 phone is a brick. $300 will barely get you the lowest end Symbian phone unlocked (the E50). And Symbian is not exactly open or standard and a pain to develop for (I've tried).
Well, if your view that it's all marketing is true, we might as well roll over and let Jobs and Gates and all the other non-innovative companies do it to us while they grab our wallets, which is just what those companies are trying to do.
OpenMoko is trying to compete, and I think they have a good chance. Apple's development speed seems like it's glacial, and the feature set on the iPhone will be fairly easy to replicate on the OpenMoko. It's the hundreds of additional apps that will make the phone attractive.
Of course, there is a good chance that OpenMoko and the Neo will fail, but it's certainly worth a try, given the kind of overpriced p.o.s. that the major phone vendors are putting out.
This smacks of the same sort of complaint-response attitude that drives the also-ran category in the music player market.
Possibly. Or possibly Apple got it wrong with the iPhone. Or possibly Apple got it wrong and they are still going to win through monopolistic practices and marketing. All one can do is try to develop a better product and see whether one can compete.
Wake me when it syncs with iTunes and automatically pulls my contacts, music, movies, TV shows, and calendar.
Why the hell would I want to sync with anything on my desktop? I want to sync with Yahoo! and Google and eMusic and Democracy and applications like those, over the air, without having to rely on a flaky and bulky desktop PC or Mac and without having a costly.Mac subscription. The iPhone view of the world is broken as far as I'm concerned.
Note that the screen is 640x480 pixels; this may be the first phone with good enough pixel density and resolution for decent handheld reading. And the fact that it's open source means that you aren't locked into an ebook reader.
I understand that a particular individual's experience in hiring people is extremely unlikely to meet the statistical criteria of a representative sample
First, what I was responding to was that you implied that it was necessary to actually interview a large fraction of the entire applicant pool in order to make statements about the applicant pool. I simply pointed out that a sample was sufficient.
Now, is the sample I interviewed representative? Well, it's representative for the kinds of companies I worked at, which tend to attract above average applicants. So, yes, it tells me a lot about the applicant pool that's out there.
and make no attempt to correlate that criteria with the actual performance of those they hire.
Of course, they do: I interview the guy, I have to live with him for at least the next 3 years. It doesn't require great amounts of bookkeeping to correlate my own impressions when I hired the guy with his performance. All those documents are also part of the personnel file, so one can look at them in a single place.
In practice, companies rarely have a well defined and consistently applied criteria to evaluate candidates [...] Thus the hiring process remains rather subjective.
All the companies I have worked at have had quite well-defined and consistently applied criteria: HR makes sure that the candidate's resume meets the formal job requirements, and after that, they hire if and only if several team members give the OK based on interviews.
Of course, the decision by each team member is a subjective one. That's because nobody has come up with better criteria. Many candidates that would have worked out are rejected that way. That's because accepting a bad candidate is at least as costly than rejecting a good one, and so people tend to set the cut-off for hiring at about where an applicant has a 50/50 chance of working out.
It will be really hard to implement this in a user friendly way
Well, AppArmor (available for Linux) has fairly straightforward configuration files for major apps. So it can, for example, ensure that your Firefox can only read/write your download directory and the system directories it needs to.
So what percentage of total US applicants have you evaluated?
Do you understand the concept of a "representative sample"?
How do we know that you judgment is good in these matters?
If you're asking me "will I prove it to you", no I won't. I'm just sharing my observations. Whether you believe them or not is up to you.
Now, as for my observations, let me put it this way. The question isn't really whether I can correctly analyze whether a candidate is good; if a candidate's resume and interview doesn't clearly demonstrate that he is good, he is not going to get hired because the cost and risk of making a bad hire are just so high (and I know that from painful experience). Furthermore, nowhere I have worked have hiring decisions been the decision of a single person anyway. Usually, resumes and candidates get circulated and people get hired only if everybody on a team agrees, and there is usually pretty good consensus.
Incidentally, if you don't have an excellent resume and can't land a permanent job, one way of getting into a company is through internships or contracting, because that allows people to get to know you with low risk to themselves.
Right now, people can easily steal my identity or impersonate an American citizen because the national id system that we have (and make no mistake, we have one) is so poor: the id cards are easy to forge and hard to verify. That's why there are so many illegal migrants getting jobs in the US, for example.
Rather than helping preserve privacy rights, states like NH are, effectively, actively hostile to privacy, because they condemn us to sticking with the current, broken system. Real ID may have design flaws (I don't know the details), but efforts like those in NH were better spent on coming up on a better id system than to simply torpedo any legislation in this area.
You're not alone in thinking that 80 columns is not enough. You're also wrong. The 80 column convention isn't a result of hardware limitations, it derives from time-tested results on readability. And, in fact, 80 columns is basically readable text line length (around 60 characters) plus a handful of levels of indentation.
Almost any function/method that's more than an 80x24 screen full of text is too big. Occasionally, you'll need to write something bigger, but at least it shouldn't be convenient to do so.
Some modern languages have been designed to be positively hostile to good layout/presentation. For example, Java forces you to nest methods inside classes and to nest classes inside each other. But even with that kind of idiotic syntax, you still shouldn't need more than half a dozen levels of indentation at four spaces each, leaving plenty of room for actual code. Another problem with Java is that it doesn't give you the ability to create short aliases, meaning that you have to use long, descriptive names throughout your code, which tend to overflow lines. But bad language design doesn't alter the laws of readability, so you'll just have to figure out how to fit it in best. The 80 column limit is part of your brain's design, not part of the computer.
So your assumption is that all the US programmers who aren't hired are bad. What do you base this on?
I'm assuming nothing. I'm going by the applicants I have evaluated while working at different companies. There's a small percentage of good applicants that make the cut--nowhere near enough.
That would be quite remarkable given that those practices are radically different at different companies.
I haven't seen "radically different" hiring practices between companies.
There really aren't that many chord progressions that work, and many of them have been known for probably a century and are in public domain music. What exactly could be copyrighted there?
And how exactly are you going to encourage people to create new works?
Media companies don't want to encourage the creation of new works, they want artificial scarcity. They want to create the impression that making music, making pictures, writing stories, and making videos is some kind of black art that only they can do and that costs millions in investments. As far as they're concerned, if teaching music were outlawed, it would be all the better because they could just keep selling the crap they are selling right now.
In france and the united states, we can trivially see the cost of allowing large numbers of unskilled immigrants into the country.
The US doesn't allow "large numbers of unskilled immigrants into the country"; the US has a large illegal migrant population, which is silently tolerated by politicians because they are actually needed to keep economies like California alive.
There is no free lunch. You are giving away the store to these corporations.
So, you're saying that Microsoft is moving their R&D lab to Canada in order to hire unskilled immigrants that will be rioting in the streets and can't pay for their children's education? Get real.
You're confusing legal, skilled workers with illegal, unskilled migration. I have no idea what the benefits and costs of illegal, unskilled migration are (California seems to depend on it). But what people like you are doing is that you are criticizing and restricting the area of immigration that is clearly beneficial (skilled immigration and work visas) because it's easy to do, with the consequence that there will be fewer and fewer skilled immigrants and more and more unskilled migrants.
I'm not sure what you hope to gain by this. Isn't the loss of skilled jobs to Canada a clear enough signal to you? And that's only the tip of the iceberg.
I don't. There are good programmers in the US, just like there are in India and China. There are also a lot of bad programmers in all those countries. So, as a US company, you first hire all the good programmers you can get from the US and then you try to hire all the good programmers you need from India and China. The bad US programmers simply aren't worth hiring or training.
It's just that now the bad US programmers want to force companies to hire them instead of hiring good programmers from overseas. But that's not going to work: just because you have a BA in computer science and a MCSE (if that) doesn't mean you are useful or ever will be useful.
If the pressure to force US companies to hire bad US programmers doesn't let up, then two things are going to happen. First, a lot of software development labs will move out of the US. Second, there will be a push for a rigorous and expensive certification program, and you won't be able to get any programming job unless you pass. I think about 20% of US programmers would probably pass right now, and such a program would probably mean loans and expenses from students.
Microsoft is right that they aren't party to the license and that they aren't copying the distributed code. So, technically, "the license doesn't apply to them".
However, what does apply to them is the contract they have with their customers when they distribute vouchers, and the GPLv3 does apply to the recipient of those vouchers once they download the GPLv3 code. So, unless they become a party to the GPLv3 license, the vouchers are worthless and they aren't fulfilling their obligations.
This is really not so different from the kind of sublicensing agreements Microsoft makes frequently. I'm not a party to Microsoft's license with the MPEG-LA, but they are required by their license with the MPEG-LA to ensure that their license with me restricts my use of the software in the way that the MPEG-LA requires.
The canadians most likely did what some states were doing here in 2000. Giving so many inducements that it actually cost them money to attract the businesses (10 years no taxes-- AND we will pay all medical and social costs for your employees. Oh wait.. "we" the state are going broke attracting business in this fashion....)
The flaw with your reasoning is that immigrants are a net economic gain for the country, not a burden or cost. And that's true even moreso for skilled immigrants.
Now I call BS. The "normal" burden cost for a Seattle-area software engineer is ~$150K-200K.
Yes, that's what they pay for giving the guy an office and a desk, but that's not the figure that counts. What counts is what total expenses the company has for every employee as part of doing business, and those are about $400k. Those include a lot of other costs.
Companies like Microsoft seem to have developed the attitude that people shouldn't find their security holes at all, but if they do, they should be obligated to report them for free.
I think a free market approach like this is good.
As for vetting buyers and sellers, I don't think that's either necessary or desirable. If people find security holes through "illegal means" (whatever that means), it's a matter for the police and courts. And if the mafia outbids Microsoft, well, then Microsoft will have to live with the consequences or pay more next time. Companies like Microsoft should be exposed to the true costs of their security vulnerabilities, and they will be exposed to that only if the "bad guys" are in on the bidding, because vulnerabilities aren't worth a lot to the other "good guys".
If prices and damages get high enough, companies will invest enough in software development to stop creating security vulnerabilities in the first place.
They are having legal trouble finding immigrants. They don't want to raise salaries.
Oh, right, because we all know that paying a bad programmer an extra $100k/year suddenly makes him a good programmer.
Face it, many people who put "programmer" on their resume aren't worth hiring even if they work for free; you'd be willing to pay them in order to keep away from the source code.
Because their owners are benefiting from being in a safe, stable country where the government does not nationalize them and the people do not kidnap, torture, and murder them.
Yeah, they are moving to Canada! The horror of it! We all know what a dangerous, hostile place Canada is!
Slashdot Mirror
Look at the release dates. Also, many of those devices just can't be called "phones"; they are mini-laptops with SIM slots.
No, right now, .Mac isn't used--you're tethered to your desktop. But .Mac is the most likely future direction for OTA sync on the iPhone.
Hundreds of apps doesn't make me buy. I buy because someone communicates to me the ONE (or few) apps
Everybody does. And everybody has different apps that matter to them. That's why having lots of apps matters.
And let me tell you - it's a vicious cycle. If the phone isn't attractive to mainstream, developers won't develop mainstream apps for it, and mainstream won't buy it.
Ah, yes, and Linux will never work because nobody will develop software for it, right? Current phones (including the iPhone) come with so little software that is so limited that the bar is really low. Most of the so-called mainstream developers are fixing bugs and omissions in the base OS, something OpenMoko doesn't need.
OpenMoko costs $450/$600. You can get a Symbian/WinMobile smart phone with open API for less than that.
OpenMoko costs $300 with a 640x480 screen and GPS (the $450 and $600 include development hardware, something that costs thousands of dollars from other vendors). There is no Symbian or WinMobile that comes even close. In fact, the only other 640x480 phone is a brick. $300 will barely get you the lowest end Symbian phone unlocked (the E50). And Symbian is not exactly open or standard and a pain to develop for (I've tried).
Well, if your view that it's all marketing is true, we might as well roll over and let Jobs and Gates and all the other non-innovative companies do it to us while they grab our wallets, which is just what those companies are trying to do.
OpenMoko is trying to compete, and I think they have a good chance. Apple's development speed seems like it's glacial, and the feature set on the iPhone will be fairly easy to replicate on the OpenMoko. It's the hundreds of additional apps that will make the phone attractive.
Of course, there is a good chance that OpenMoko and the Neo will fail, but it's certainly worth a try, given the kind of overpriced p.o.s. that the major phone vendors are putting out.
This smacks of the same sort of complaint-response attitude that drives the also-ran category in the music player market.
.Mac subscription. The iPhone view of the world is broken as far as I'm concerned.
Possibly. Or possibly Apple got it wrong with the iPhone. Or possibly Apple got it wrong and they are still going to win through monopolistic practices and marketing. All one can do is try to develop a better product and see whether one can compete.
Wake me when it syncs with iTunes and automatically pulls my contacts, music, movies, TV shows, and calendar.
Why the hell would I want to sync with anything on my desktop? I want to sync with Yahoo! and Google and eMusic and Democracy and applications like those, over the air, without having to rely on a flaky and bulky desktop PC or Mac and without having a costly
Note that the screen is 640x480 pixels; this may be the first phone with good enough pixel density and resolution for decent handheld reading. And the fact that it's open source means that you aren't locked into an ebook reader.
I understand that a particular individual's experience in hiring people is extremely unlikely to meet the statistical criteria of a representative sample
First, what I was responding to was that you implied that it was necessary to actually interview a large fraction of the entire applicant pool in order to make statements about the applicant pool. I simply pointed out that a sample was sufficient.
Now, is the sample I interviewed representative? Well, it's representative for the kinds of companies I worked at, which tend to attract above average applicants. So, yes, it tells me a lot about the applicant pool that's out there.
and make no attempt to correlate that criteria with the actual performance of those they hire.
Of course, they do: I interview the guy, I have to live with him for at least the next 3 years. It doesn't require great amounts of bookkeeping to correlate my own impressions when I hired the guy with his performance. All those documents are also part of the personnel file, so one can look at them in a single place.
In practice, companies rarely have a well defined and consistently applied criteria to evaluate candidates [...] Thus the hiring process remains rather subjective.
All the companies I have worked at have had quite well-defined and consistently applied criteria: HR makes sure that the candidate's resume meets the formal job requirements, and after that, they hire if and only if several team members give the OK based on interviews.
Of course, the decision by each team member is a subjective one. That's because nobody has come up with better criteria. Many candidates that would have worked out are rejected that way. That's because accepting a bad candidate is at least as costly than rejecting a good one, and so people tend to set the cut-off for hiring at about where an applicant has a 50/50 chance of working out.
It will be really hard to implement this in a user friendly way
Well, AppArmor (available for Linux) has fairly straightforward configuration files for major apps. So it can, for example, ensure that your Firefox can only read/write your download directory and the system directories it needs to.
So what percentage of total US applicants have you evaluated?
Do you understand the concept of a "representative sample"?
How do we know that you judgment is good in these matters?
If you're asking me "will I prove it to you", no I won't. I'm just sharing my observations. Whether you believe them or not is up to you.
Now, as for my observations, let me put it this way. The question isn't really whether I can correctly analyze whether a candidate is good; if a candidate's resume and interview doesn't clearly demonstrate that he is good, he is not going to get hired because the cost and risk of making a bad hire are just so high (and I know that from painful experience). Furthermore, nowhere I have worked have hiring decisions been the decision of a single person anyway. Usually, resumes and candidates get circulated and people get hired only if everybody on a team agrees, and there is usually pretty good consensus.
Incidentally, if you don't have an excellent resume and can't land a permanent job, one way of getting into a company is through internships or contracting, because that allows people to get to know you with low risk to themselves.
Right now, people can easily steal my identity or impersonate an American citizen because the national id system that we have (and make no mistake, we have one) is so poor: the id cards are easy to forge and hard to verify. That's why there are so many illegal migrants getting jobs in the US, for example.
Rather than helping preserve privacy rights, states like NH are, effectively, actively hostile to privacy, because they condemn us to sticking with the current, broken system. Real ID may have design flaws (I don't know the details), but efforts like those in NH were better spent on coming up on a better id system than to simply torpedo any legislation in this area.
You're not alone in thinking that 80 columns is not enough. You're also wrong. The 80 column convention isn't a result of hardware limitations, it derives from time-tested results on readability. And, in fact, 80 columns is basically readable text line length (around 60 characters) plus a handful of levels of indentation.
Almost any function/method that's more than an 80x24 screen full of text is too big. Occasionally, you'll need to write something bigger, but at least it shouldn't be convenient to do so.
Some modern languages have been designed to be positively hostile to good layout/presentation. For example, Java forces you to nest methods inside classes and to nest classes inside each other. But even with that kind of idiotic syntax, you still shouldn't need more than half a dozen levels of indentation at four spaces each, leaving plenty of room for actual code. Another problem with Java is that it doesn't give you the ability to create short aliases, meaning that you have to use long, descriptive names throughout your code, which tend to overflow lines. But bad language design doesn't alter the laws of readability, so you'll just have to figure out how to fit it in best. The 80 column limit is part of your brain's design, not part of the computer.
So your assumption is that all the US programmers who aren't hired are bad. What do you base this on?
I'm assuming nothing. I'm going by the applicants I have evaluated while working at different companies. There's a small percentage of good applicants that make the cut--nowhere near enough.
That would be quite remarkable given that those practices are radically different at different companies.
I haven't seen "radically different" hiring practices between companies.
There really aren't that many chord progressions that work, and many of them have been known for probably a century and are in public domain music. What exactly could be copyrighted there?
I think that's called the Neo1973. It's cheaper and has a better screen than the iPhone.
And how exactly are you going to encourage people to create new works?
Media companies don't want to encourage the creation of new works, they want artificial scarcity. They want to create the impression that making music, making pictures, writing stories, and making videos is some kind of black art that only they can do and that costs millions in investments. As far as they're concerned, if teaching music were outlawed, it would be all the better because they could just keep selling the crap they are selling right now.
Oh yea, free market always works!
No, free market approaches don't always work. But in this case, I think it would.
Very select groups of immigrants provide a net economic gain.
0 5787940-tA5PP0Ya_9U0AlXBQQhnaDyMIYc_20060725.html? mod=tff_main_tff_top
5 .htm
That's wrong. Legal immigrants are a net gain for the economy overall:
http://online.wsj.com/public/article/SB1151009483
http://www.ailf.org/ipc/policy_reports_1997_pr973
In france and the united states, we can trivially see the cost of allowing large numbers of unskilled immigrants into the country.
The US doesn't allow "large numbers of unskilled immigrants into the country"; the US has a large illegal migrant population, which is silently tolerated by politicians because they are actually needed to keep economies like California alive.
There is no free lunch. You are giving away the store to these corporations.
So, you're saying that Microsoft is moving their R&D lab to Canada in order to hire unskilled immigrants that will be rioting in the streets and can't pay for their children's education? Get real.
You're confusing legal, skilled workers with illegal, unskilled migration. I have no idea what the benefits and costs of illegal, unskilled migration are (California seems to depend on it). But what people like you are doing is that you are criticizing and restricting the area of immigration that is clearly beneficial (skilled immigration and work visas) because it's easy to do, with the consequence that there will be fewer and fewer skilled immigrants and more and more unskilled migrants.
I'm not sure what you hope to gain by this. Isn't the loss of skilled jobs to Canada a clear enough signal to you? And that's only the tip of the iceberg.
I don't. There are good programmers in the US, just like there are in India and China. There are also a lot of bad programmers in all those countries. So, as a US company, you first hire all the good programmers you can get from the US and then you try to hire all the good programmers you need from India and China. The bad US programmers simply aren't worth hiring or training.
It's just that now the bad US programmers want to force companies to hire them instead of hiring good programmers from overseas. But that's not going to work: just because you have a BA in computer science and a MCSE (if that) doesn't mean you are useful or ever will be useful.
If the pressure to force US companies to hire bad US programmers doesn't let up, then two things are going to happen. First, a lot of software development labs will move out of the US. Second, there will be a push for a rigorous and expensive certification program, and you won't be able to get any programming job unless you pass. I think about 20% of US programmers would probably pass right now, and such a program would probably mean loans and expenses from students.
Microsoft is right that they aren't party to the license and that they aren't copying the distributed code. So, technically, "the license doesn't apply to them".
However, what does apply to them is the contract they have with their customers when they distribute vouchers, and the GPLv3 does apply to the recipient of those vouchers once they download the GPLv3 code. So, unless they become a party to the GPLv3 license, the vouchers are worthless and they aren't fulfilling their obligations.
This is really not so different from the kind of sublicensing agreements Microsoft makes frequently. I'm not a party to Microsoft's license with the MPEG-LA, but they are required by their license with the MPEG-LA to ensure that their license with me restricts my use of the software in the way that the MPEG-LA requires.
What makes you think those foreign programmers are bad? Companies like Microsoft can hire the best of the best from India and China, and they do.
The canadians most likely did what some states were doing here in 2000. Giving so many inducements that it actually cost them money to attract the businesses (10 years no taxes-- AND we will pay all medical and social costs for your employees. Oh wait.. "we" the state are going broke attracting business in this fashion....)
The flaw with your reasoning is that immigrants are a net economic gain for the country, not a burden or cost. And that's true even moreso for skilled immigrants.
Now I call BS. The "normal" burden cost for a Seattle-area software engineer is ~$150K-200K.
Yes, that's what they pay for giving the guy an office and a desk, but that's not the figure that counts. What counts is what total expenses the company has for every employee as part of doing business, and those are about $400k. Those include a lot of other costs.
Companies like Microsoft seem to have developed the attitude that people shouldn't find their security holes at all, but if they do, they should be obligated to report them for free.
I think a free market approach like this is good.
As for vetting buyers and sellers, I don't think that's either necessary or desirable. If people find security holes through "illegal means" (whatever that means), it's a matter for the police and courts. And if the mafia outbids Microsoft, well, then Microsoft will have to live with the consequences or pay more next time. Companies like Microsoft should be exposed to the true costs of their security vulnerabilities, and they will be exposed to that only if the "bad guys" are in on the bidding, because vulnerabilities aren't worth a lot to the other "good guys".
If prices and damages get high enough, companies will invest enough in software development to stop creating security vulnerabilities in the first place.
They are having legal trouble finding immigrants. They don't want to raise salaries.
Oh, right, because we all know that paying a bad programmer an extra $100k/year suddenly makes him a good programmer.
Face it, many people who put "programmer" on their resume aren't worth hiring even if they work for free; you'd be willing to pay them in order to keep away from the source code.
Because their owners are benefiting from being in a safe, stable country where the government does not nationalize them and the people do not kidnap, torture, and murder them.
Yeah, they are moving to Canada! The horror of it! We all know what a dangerous, hostile place Canada is!