If I could give you a mod point for the interesting pointer I would.
One slide indicating that Unicode encodings may, at some vague point in the future, be treated on a par with the obsolete national legacy encodings Matz prefers is hardly a bold step toward internationalization. It's more of Matz's "well, if you want to use Unicode, I won't stop you" level of commitment to internationalization.
I like a *lot* of what I see in Ruby, and I strongly agree with your criticisms of Python (though I'm still undecided about the whitepace issue.)
Even so, I think Matz's attitude about Unicode and internationalization rules Ruby out as a serious candidate for me.
All Windows and Macs machines being sold today already use Unicode. The Unix world is in last place but the conversion is happening almost everywhere. The very last holdout against Unicode will probably be the Unix-style OSes in Japan. While all other platforms are Unicode and most of the Unix world is converted, they will be the last of the legacy holdouts.
Japanese *nix is also the center of Matz's focus, and it shows. He has made it clear that the principal driving force behind Ruby's design is to help him do his own Japanese work in an encoding (EUC-JP) that he says is "good enough for me [him]". Since EUC-JP doesn't support anything well except for Japanese, he clearly doesn't internationalize his own work, and his comments make it pretty clear that anyone who does want to follow modern internationalization practices had better look elsewhere for a language.
Also, the libraries in Ruby are far less developed than in Python, and I don't see them getting "good enough" anytime soon. Matz's Japanese Unix-centric community just hasn't produced the libraries that the rest of us are beginning to demand from our production languages.
If I were just tinkering around with personal utilities and if legacy subset charsets were good enough for my text needs, then it might be all right. But I need lots of big, solid, production quality libraries for Unicode-based commercial systems, and Ruby isn't close. (Even Python is a stretch.)
China isn't just "some 3rd world nation". It's virtually the definition of 3rd world.
China shoots corrupt officials. America puts them in office.
What a pathetic joke this is. I've dealt with Chinese and Western officials for years. The level of corruption in China is in a completely different league from that in the US. China is solidly 3rd world in its corruption, like Indonesia, India, or Mexico. The law doesn't matter, just guanxi. The shooting of corrupt officials you refer to is just more corruption.
American police killed peace protestors, china killed human right protestors. Big frigging difference.
Ah, a favorite Chinese obfuscation tactic: one and a million are both positive integers, so they're both essentially the same.
We're not that stupid. While the Chinese lie with their words, trying to confuse the issue to "save face", we see them demonstrate with their actions that they know very well that there is a "big frigging difference".
I'd be willing to carry a "Time for a Different Party in Government" sign in front of the White House if I were in Washington, though I might have to wait my turn since apparently people in the US feel free to do that sort of thing.
Would you be willing to carry the same sign in Tiananmen Square?
If we were doing it for the scenery, we'd have aimed for Olympus Mons or one of the spectacular canyons.
This mission is designed to survive in order to dig into the dirt, drill into the rocks, and do quite a few other experiments that require being there and touching things. Of course, we're not going to do it with our eyes closed, so we take pictures to figure out where to go and what to try in addition to learning anything we can from the images.
I think you're dramatically underestimating the business logic involved in a tax program. How many rule sets can you think of that are as convoluted and byzantine as the US Federal Tax Code? You're dismissing the hard stuff with the wave of a hand and then focusing on the easy stuff. (Except for the liability part, which you correctly identify as a huge issue.)
And even with a small army of testers, so many issues still slip past QA that the first thing TurboTax does when you start it up is look for new patches. Yes, I've written custom spreadsheet functions that will apply the basic tax bracket ladder algorithm to your gross salary. But that isn't even scratching the surface.
And every time the tax laws change (e.g. at least every year), a lot of that logic has to be updated and tested some more.
Just consider the money TurboTax brings in. Half of the US buys it, and then buys it again *every year*. If it were as easy as you seem to believe, why wouldn't every company that could afford liability insurance get into that game?
You're erroneously presupposing that I'm interested in your approval.
No, I'm correctly supposing that regardless of your interests, you can't do what the poster (you?) claimed. If you're thinking that a simple filter and substitution approach will be good enough, you're mistaken, unless of course your definition of "good enough" is not good enough.
Why would I want non-ascii emails?
Because that's what everybody will be sending in the not-too-distant future, and it's what most of us are sending already. ASCII is pathetic for natural language and is steadily being replaced by Unicode, passing through various half-baked compromise charsets on the way. Email is following.
Are you foreign?
No, you are.;-)
It won't as I'd only be checking the subject line.
I take it back. If your standards are this low, whatever code you write you'll define as having solved the problem, so in a way, I'm now willing to bet that you actually could "solve" it in an afternoon.
You guys amaze me. I don't think you understand the problem at all, but maybe I'm wrong. Fine, take half a minute, write the Perl code, post it, and we'll see. (My guess is we won't be seeing any code from either you or the "halfway competent C programmer".)
As for "mark anything with a lot of high ASCII characters as spam" guy, is everything except English spam? Maybe to you, but I wouldn't call a solution that only works for you much of a solution.
Almost all of my outgoing email now is UTF-8, and I take advantage of the much wider range of characters it provides. Make sure your 30 second algorithm doesn't mistake a non-ASCII charset for spam.
And what about source code? Do you ever get source code snippets in the mail? Take a few seconds and make sure your algorithm doesn't mistake source code in any programming language or technical acronyms for spam.
Okay, get ready to write code. I'm looking at my watch. Go!
I've been using Linux for some time now, and there's little chance of my giving it up, but I still can't STOP using Windows. My taxes are complicated enough that there's no way I'd give up the benefits of TurboTax unless it's for something equivalent or better.
Where does Stallman think a free (as in yadda, yadda), reliable, continually updated TurboTax clone is going to come from? How does Stallman do *his* taxes (assuming he even does them)?
It's called making a contribution, and I'm with you. There are a lot of times when my friends and family have to make an extra effort to pull part of my load. I'm not good at everything.
I'm good at this, though, so this is a good way for me to be a net giver anyway, rather than a net taker.
It's refreshing to see a posting like this. There's nothing I wouldn't do to help my parents, if I could. I'll help my friends and extended family as much as I reasonably can, too. I expect them (except my parents) to pay for their own parts, but I can't imagine charging them for my time.
I won't spend an unlimited amount of time. I'm careful and I will sometimes have to explain that the job is going to take more time than I'm going to be able to put into it, but I try to do what I can to help them out with the skills that I'm fortunate enough to have. If they can help me out somehow in return by doing me some sort of return favor, that's great. If not, well, that's not why I do it.
Don't mind me, I'm just speaking for recorded history.
Well, you may be speaking for it, but little of what you say is true.
"Africa once enslaved their own people, but for far different terms...."
Africa had slavery before Europeans arrived, and slavery of Africans by Africans continues to this day.
"...far different...more of an indentured servitude, usually reserved for criminals."
Nonsense. Slavery was, and continues to be, something perpetrated for a wide variety of reasons: sex slavery, free labor, religious oppression, ethnic rivalry, etc.
"In American, conversely, only white slaves were ever granted such freedom."
More nonsense. There were a lot of blacks who were "freemen" under the law, having been granted their freedom legally for one reason or another. Such a legal category, and the law behind it, wouldn't make sense in a world such as you describe.
This isn't a defense of slavery, which was, and continues to be, an atrocity. It's just nonsense to label it as more of an atrocity if committed by whites than by blacks, or to try to whitewash the actions of the enslavers if they happen to be black.
I don't deny the caste consciousness still strongly present in many Indians, but I don't think it's relevant here. We Westerners are exempt from the system. I've worked in India and with Indians in the West, and among the many perplexing cultural differences I've run into, the inclusion of *me* in their caste system was never one of them.
I think that what you encountered was just an individual personality. I've had these experiences with Indians, too (especially bureaucrats who wanted to prove their importance), but I've had similar experiences with people everywhere. I've managed a tech support group in the US and some of my own people acted this way (until I either stopped it or got rid of them.) It wasn't correlated to their skill level either, just to the degree to which they seemed to feel the need to prove to others that they were smarter (which seems to afflict geniuses and idiots in roughly equal proportions.)
I want my CNET radio, but more hard core. The soft core fluff that dominates the broadcast "tech shows" is almost worthless. It's not technical enough to very useful to the only people interested enough in tech to tune in at all.
These shows always end up gradually dumbing themselves down in an effort to increase ratings. They figure, correctly I think, that there are more people out there striving to be hiphoppy and sunglassy than technically savvy.
What they miss is that these people aren't going to tune in to a "tech station" at all, no matter how many of them there are. The only people who *are* interested in tech are too few, as a percentage, to support a traditional "broadcast" and usually too well informed already for occasional small bits of technical information floating in a watery gruel of hipness to hold their attention for long.
I think what we need is to have something with the reach of satellite radio that can carry thousands of channels, most of which are tailored to niche markets. Among them would be a handful of stations for those who are already well-informed in the field: science, technology, serious investing and finance, foreign language study, etc.
If something like satellite radio can't handle this, then nuts to 'em. Let's find a way to make access to the Internet packet stream as ubiquitous as mobile phone service is becoming, and start tuning in to URLs from our cars and pocket radios.
Look, nobody is overlooking anything here. DNA planting is already a staple of crime dramas. Regardless of who is in the database, you can easily obtain DNA from someone who is, or you can plant DNA from someone who isn't and just report that you think you may have seen them leaving the scene. Go ahead and report their license plate number, which is easier to get than their DNA. Then the police sample their DNA and it's just as if they'd been in the database all along.
If you can frame someone with DNA (and you certainly can), then you can frame them approximately as easily whether they're already in the database or not.
Same thing happened to me with an Icon computer backpack, but I was just standing there. The strap simply ripped off the body. (I had it over one arm, the way college students tend to carry a backpack.)
You ought to be able to run an obstacle course with a laptop backpack full of reference manuals, and the runner should fail before the backpack ever does.
Good point. Let's put mine in, too. I don't care if they *do* put us all in the database. I don't understand why I should have the right to be not knowable if my DNA shows up at a crime scene unless I have been previously *convicted* of a felony. By this argument, we want to make sure that those who are committing their first felony cannot be found via DNA. What is the advantage to society that we be unable to find first-time felons, or tenth-time felons who managed to plea bargain down to a non-felony each time they commit a felony?
I think people get confused about the laws regarding unreasonable search and seizure. Historically, many governments would go in and tear apart the homes and destroy or steal the belongings of anyone they didn't like. If they claimed it was a search for criminal evidence, they could get away with causing you almost unlimited harm at their whim. The law was to prevent them from harming you without demonstrating to an independent party that there was a reasonable necessity to conduct the search, despite the harm it might cause you.
If your DNA data is in a database, you're not having your home torn apart or your possessions ransacked each time someone does a pattern search. You're not being hurt at all, except for the risk that you might be identified, correctly or not. IANAL, but I don't know of any constitutional provision against being eligible for possible (mis)identification. If there were, they'd have to take our license plates off our cars, because they allow our vehicles to be uniquely identified as we drive away from a crime scene. And every time someone searches for a half-remembered license plate number, my number is a potential candidate with an enormously higher probability of misidentification than my DNA.
Most of us who live in major metro areas can already find sports stations, rap stations, left & right talk stations, Spanish language stations, etc.
Yes, it's nice to have it with somewhat fewer ads and somewhat more thinly sliced into genres, but unless I lived in a rural area or did a lot of cross-country driving, it just wouldn't be worth it to have just an incremental improvement in what I already get on any radio, for free.
What I'm really looking for is programming that I simply can't get at all because even in the largest metro area there's still not a critical mass of people who would want it.
I want the "scientific conference channel". I want the "computer science lecture" channel. I want the "science news and talk" channel. I want a channel devoted to world history lectures.
I want what I had in Japan on cable radio: a dedicated "learn (Chinese|French|Korean|...) language" channel for each of ten or so major languages (that's ten or so separate channels), with a daily half-hour lesson that repeats 48 times over 24 hours, so you can tune in at any time and hear the day's lesson. (Accompanying textbook available at every newsstand.)
I want the leading talk radio and local music stations in every country on earth with a radio station to be piped through on their own dedicated channels. On a cold winter day, stuck in traffic, I'd like to be able to tune in to a live radio broadcast from Tahiti, practice my French, and dream of the islands.
Since I don't see that happening soon, I think I'll just hope for the infrastructure for an always-on Internet connection from the car and I'll hope streaming audio on the Net will grow into what I'm looking for.
In the meantime, any suggestions for a good way to record streaming RealAudio into MP3 files that I can then load into a portable MP3 player? I think that's going to be my "satellite radio" if I can figure out a convenient, automated way to do it.
You just have to look at Pearl Harbour and 9/11 to see how badly the US take to threats to the 'homeland'.
I agree. Look at Pearl Harbor and notice how the US treated the defeated Japanese as well as they treated the defeated Germans. You're right that the US takes threats to the homeland badly, if you mean very seriously. You're clearly wrong about it having anything to do with US magnanimity in victory.
I do think C is past its prime. As machines get more speed and memory, and as more and more code is written to run on a server, the percentage of apps that have to make speed or memory size the first priority is steadily decreasing. That means C's market share is steadily decreasing.
Even so, I expect C to far outlast Perl, and not because it's a better language. It's because it is an extreme language: the portable language closest to the metal. Though the need for such a thing continues to diminish as a percentage, it will still be the best tool for certain jobs for a long time to come.
Yes, thanks, Larry. I think Perl was the right tool in the right place at the right time: the duct tape of the Web gold rush.
From what I can tell, though, it appears to have peaked and is now in relative decline. Python is gaining rapidly on Perl in the "scripting language" space. Java, and now PHP, have eroded Perl's popularity in an area it once almost monopolized: Web apps. And its drive to evolve its way from being a useful merger of sed and awk to a full-blown object-oriented programming language may be dragging too much legacy syntax to go much farther.
I'm not trying to insult Perl. It has been enormously helpful to me for years. I'm just seeing signs that at 16, it's probably past its prime.
I don't want to put too fine a point on this, but your concept of "quality per dollar" is not quite what I'm describing. "Since dollars are limited and we want as much quality as possible" sort of implies a situation where we have a fixed budget (hold price constant) and shop for the highest (try to maximize) quality obtainable at that price.
That's opposite from the events I've been involved with or have discussed with colleagues. In our case, we were holding quality constant in the sense that it had to meet certain standards, and we were trying to minimize cost. "What's the cheapest way to get the required quality?" instead of your "What's the highest quality we can get with our limited dollars?"
As I said, it's not really one extreme or the other. People really are aware of the added costs that can come from lower quality, but the emphasis really does tend to be on cost savings.
I've been a part of this outsourcing decision-making process on a few occasions and discussed it with others like me who were making their own decisions.
I'm not personally aware of a single case where it was thought that work being outsourced to India would be higher in quality. Just that it would meet the required quality threshold at a dramatically lower price.
And these decisions are thumbs down as often as thumbs up. (Actually, far more often, but that's must my personal experience. I don't know how representative it is.) At times, we conclude that the quality isn't likely to be good enough, or that the savings isn't likely to be large enough, after factoring in the logistical overhead, to be worth doing.
And companies large enough to open their own offices in India may do critical work there, but I'm not personally aware of any decisions where critical code was outsourced. The ones I've seen have always been the sort of "well, if that system has some problems for a while it won't kill us" type of utility projects.
Sure. And if that's a good argument for you, then you can enjoy the benefits of that system by moving someplace like Afghanistan or Bolivia.
Is it better to have every single employee be in a constant state of entry-level skill when there are no entry-level careers?
"Entry-level" is an interesting notion. By your definition, all Java programmers in 1995 were "entry-level Java programmers", but what does that mean? Being new to a new specialty is not the same as being new to an old one. The former includes people who are quite successful, while the latter is just an ordinary newbie.
Every Windows laptop has two buttons built in and the most popular models even have an equivalent of the scroll wheel. On ThinkPads, Dells, and others, I can scroll a window without moving my fingers off the center of the keyboard. It's so convenient that I never use an external mouse and almost never have any use for a scrollbar. I can scroll any window in two dimensions by simply pointing anywhere in the window and moving only my right index finger off the home keys.
None of this is possible on any Macintosh laptop. Apple's primitive mouse standard is a real problem.
Slashdot Mirror
If I could give you a mod point for the interesting pointer I would.
One slide indicating that Unicode encodings may, at some vague point in the future, be treated on a par with the obsolete national legacy encodings Matz prefers is hardly a bold step toward internationalization. It's more of Matz's "well, if you want to use Unicode, I won't stop you" level of commitment to internationalization.
I like a *lot* of what I see in Ruby, and I strongly agree with your criticisms of Python (though I'm still undecided about the whitepace issue.)
Even so, I think Matz's attitude about Unicode and internationalization rules Ruby out as a serious candidate for me.
All Windows and Macs machines being sold today already use Unicode. The Unix world is in last place but the conversion is happening almost everywhere. The very last holdout against Unicode will probably be the Unix-style OSes in Japan. While all other platforms are Unicode and most of the Unix world is converted, they will be the last of the legacy holdouts.
Japanese *nix is also the center of Matz's focus, and it shows. He has made it clear that the principal driving force behind Ruby's design is to help him do his own Japanese work in an encoding (EUC-JP) that he says is "good enough for me [him]". Since EUC-JP doesn't support anything well except for Japanese, he clearly doesn't internationalize his own work, and his comments make it pretty clear that anyone who does want to follow modern internationalization practices had better look elsewhere for a language.
Also, the libraries in Ruby are far less developed than in Python, and I don't see them getting "good enough" anytime soon. Matz's Japanese Unix-centric community just hasn't produced the libraries that the rest of us are beginning to demand from our production languages.
If I were just tinkering around with personal utilities and if legacy subset charsets were good enough for my text needs, then it might be all right. But I need lots of big, solid, production quality libraries for Unicode-based commercial systems, and Ruby isn't close. (Even Python is a stretch.)
Stop think china is some 3rd world nation.
China isn't just "some 3rd world nation". It's virtually the definition of 3rd world.
China shoots corrupt officials. America puts them in office.
What a pathetic joke this is. I've dealt with Chinese and Western officials for years. The level of corruption in China is in a completely different league from that in the US. China is solidly 3rd world in its corruption, like Indonesia, India, or Mexico. The law doesn't matter, just guanxi. The shooting of corrupt officials you refer to is just more corruption.
American police killed peace protestors, china killed human right protestors. Big frigging difference.
Ah, a favorite Chinese obfuscation tactic: one and a million are both positive integers, so they're both essentially the same.
We're not that stupid. While the Chinese lie with their words, trying to confuse the issue to "save face", we see them demonstrate with their actions that they know very well that there is a "big frigging difference".
I'd be willing to carry a "Time for a Different Party in Government" sign in front of the White House if I were in Washington, though I might have to wait my turn since apparently people in the US feel free to do that sort of thing.
Would you be willing to carry the same sign in Tiananmen Square?
If we were doing it for the scenery, we'd have aimed for Olympus Mons or one of the spectacular canyons.
This mission is designed to survive in order to dig into the dirt, drill into the rocks, and do quite a few other experiments that require being there and touching things. Of course, we're not going to do it with our eyes closed, so we take pictures to figure out where to go and what to try in addition to learning anything we can from the images.
I think you're dramatically underestimating the business logic involved in a tax program. How many rule sets can you think of that are as convoluted and byzantine as the US Federal Tax Code? You're dismissing the hard stuff with the wave of a hand and then focusing on the easy stuff. (Except for the liability part, which you correctly identify as a huge issue.)
And even with a small army of testers, so many issues still slip past QA that the first thing TurboTax does when you start it up is look for new patches. Yes, I've written custom spreadsheet functions that will apply the basic tax bracket ladder algorithm to your gross salary. But that isn't even scratching the surface.
And every time the tax laws change (e.g. at least every year), a lot of that logic has to be updated and tested some more.
Just consider the money TurboTax brings in. Half of the US buys it, and then buys it again *every year*. If it were as easy as you seem to believe, why wouldn't every company that could afford liability insurance get into that game?
You're erroneously presupposing that I'm interested in your approval.
;-)
No, I'm correctly supposing that regardless of your interests, you can't do what the poster (you?) claimed. If you're thinking that a simple filter and substitution approach will be good enough, you're mistaken, unless of course your definition of "good enough" is not good enough.
Why would I want non-ascii emails?
Because that's what everybody will be sending in the not-too-distant future, and it's what most of us are sending already. ASCII is pathetic for natural language and is steadily being replaced by Unicode, passing through various half-baked compromise charsets on the way. Email is following.
Are you foreign?
No, you are.
It won't as I'd only be checking the subject line.
I take it back. If your standards are this low, whatever code you write you'll define as having solved the problem, so in a way, I'm now willing to bet that you actually could "solve" it in an afternoon.
[I'm replying to several posts in this subthread]
You guys amaze me. I don't think you understand the problem at all, but maybe I'm wrong. Fine, take half a minute, write the Perl code, post it, and we'll see. (My guess is we won't be seeing any code from either you or the "halfway competent C programmer".)
As for "mark anything with a lot of high ASCII characters as spam" guy, is everything except English spam? Maybe to you, but I wouldn't call a solution that only works for you much of a solution.
Almost all of my outgoing email now is UTF-8, and I take advantage of the much wider range of characters it provides. Make sure your 30 second algorithm doesn't mistake a non-ASCII charset for spam.
And what about source code? Do you ever get source code snippets in the mail? Take a few seconds and make sure your algorithm doesn't mistake source code in any programming language or technical acronyms for spam.
Okay, get ready to write code. I'm looking at my watch. Go!
I've been using Linux for some time now, and there's little chance of my giving it up, but I still can't STOP using Windows. My taxes are complicated enough that there's no way I'd give up the benefits of TurboTax unless it's for something equivalent or better.
Where does Stallman think a free (as in yadda, yadda), reliable, continually updated TurboTax clone is going to come from? How does Stallman do *his* taxes (assuming he even does them)?
It's called making a contribution, and I'm with you. There are a lot of times when my friends and family have to make an extra effort to pull part of my load. I'm not good at everything.
I'm good at this, though, so this is a good way for me to be a net giver anyway, rather than a net taker.
It's refreshing to see a posting like this. There's nothing I wouldn't do to help my parents, if I could. I'll help my friends and extended family as much as I reasonably can, too. I expect them (except my parents) to pay for their own parts, but I can't imagine charging them for my time.
I won't spend an unlimited amount of time. I'm careful and I will sometimes have to explain that the job is going to take more time than I'm going to be able to put into it, but I try to do what I can to help them out with the skills that I'm fortunate enough to have. If they can help me out somehow in return by doing me some sort of return favor, that's great. If not, well, that's not why I do it.
Don't mind me, I'm just speaking for recorded history.
Well, you may be speaking for it, but little of what you say is true.
"Africa once enslaved their own people, but for far different terms...."
Africa had slavery before Europeans arrived, and slavery of Africans by Africans continues to this day.
"...far different...more of an indentured servitude, usually reserved for criminals."
Nonsense. Slavery was, and continues to be, something perpetrated for a wide variety of reasons: sex slavery, free labor, religious oppression, ethnic rivalry, etc.
"In American, conversely, only white slaves were ever granted such freedom."
More nonsense. There were a lot of blacks who were "freemen" under the law, having been granted their freedom legally for one reason or another. Such a legal category, and the law behind it, wouldn't make sense in a world such as you describe.
This isn't a defense of slavery, which was, and continues to be, an atrocity. It's just nonsense to label it as more of an atrocity if committed by whites than by blacks, or to try to whitewash the actions of the enslavers if they happen to be black.
I don't deny the caste consciousness still strongly present in many Indians, but I don't think it's relevant here. We Westerners are exempt from the system. I've worked in India and with Indians in the West, and among the many perplexing cultural differences I've run into, the inclusion of *me* in their caste system was never one of them.
I think that what you encountered was just an individual personality. I've had these experiences with Indians, too (especially bureaucrats who wanted to prove their importance), but I've had similar experiences with people everywhere. I've managed a tech support group in the US and some of my own people acted this way (until I either stopped it or got rid of them.) It wasn't correlated to their skill level either, just to the degree to which they seemed to feel the need to prove to others that they were smarter (which seems to afflict geniuses and idiots in roughly equal proportions.)
I want my CNET radio, but more hard core. The soft core fluff that dominates the broadcast "tech shows" is almost worthless. It's not technical enough to very useful to the only people interested enough in tech to tune in at all.
These shows always end up gradually dumbing themselves down in an effort to increase ratings. They figure, correctly I think, that there are more people out there striving to be hiphoppy and sunglassy than technically savvy.
What they miss is that these people aren't going to tune in to a "tech station" at all, no matter how many of them there are. The only people who *are* interested in tech are too few, as a percentage, to support a traditional "broadcast" and usually too well informed already for occasional small bits of technical information floating in a watery gruel of hipness to hold their attention for long.
I think what we need is to have something with the reach of satellite radio that can carry thousands of channels, most of which are tailored to niche markets. Among them would be a handful of stations for those who are already well-informed in the field: science, technology, serious investing and finance, foreign language study, etc.
If something like satellite radio can't handle this, then nuts to 'em. Let's find a way to make access to the Internet packet stream as ubiquitous as mobile phone service is becoming, and start tuning in to URLs from our cars and pocket radios.
Look, nobody is overlooking anything here. DNA planting is already a staple of crime dramas. Regardless of who is in the database, you can easily obtain DNA from someone who is, or you can plant DNA from someone who isn't and just report that you think you may have seen them leaving the scene. Go ahead and report their license plate number, which is easier to get than their DNA. Then the police sample their DNA and it's just as if they'd been in the database all along.
If you can frame someone with DNA (and you certainly can), then you can frame them approximately as easily whether they're already in the database or not.
Same thing happened to me with an Icon computer backpack, but I was just standing there. The strap simply ripped off the body. (I had it over one arm, the way college students tend to carry a backpack.)
You ought to be able to run an obstacle course with a laptop backpack full of reference manuals, and the runner should fail before the backpack ever does.
Ditto, if sharing your silly feelings is your idea of a rebuttal.
Good point. Let's put mine in, too. I don't care if they *do* put us all in the database. I don't understand why I should have the right to be not knowable if my DNA shows up at a crime scene unless I have been previously *convicted* of a felony. By this argument, we want to make sure that those who are committing their first felony cannot be found via DNA. What is the advantage to society that we be unable to find first-time felons, or tenth-time felons who managed to plea bargain down to a non-felony each time they commit a felony?
I think people get confused about the laws regarding unreasonable search and seizure. Historically, many governments would go in and tear apart the homes and destroy or steal the belongings of anyone they didn't like. If they claimed it was a search for criminal evidence, they could get away with causing you almost unlimited harm at their whim. The law was to prevent them from harming you without demonstrating to an independent party that there was a reasonable necessity to conduct the search, despite the harm it might cause you.
If your DNA data is in a database, you're not having your home torn apart or your possessions ransacked each time someone does a pattern search. You're not being hurt at all, except for the risk that you might be identified, correctly or not. IANAL, but I don't know of any constitutional provision against being eligible for possible (mis)identification. If there were, they'd have to take our license plates off our cars, because they allow our vehicles to be uniquely identified as we drive away from a crime scene. And every time someone searches for a half-remembered license plate number, my number is a potential candidate with an enormously higher probability of misidentification than my DNA.
Most of us who live in major metro areas can already find sports stations, rap stations, left & right talk stations, Spanish language stations, etc.
Yes, it's nice to have it with somewhat fewer ads and somewhat more thinly sliced into genres, but unless I lived in a rural area or did a lot of cross-country driving, it just wouldn't be worth it to have just an incremental improvement in what I already get on any radio, for free.
What I'm really looking for is programming that I simply can't get at all because even in the largest metro area there's still not a critical mass of people who would want it.
I want the "scientific conference channel". I want the "computer science lecture" channel. I want the "science news and talk" channel. I want a channel devoted to world history lectures.
I want what I had in Japan on cable radio: a dedicated "learn (Chinese|French|Korean|...) language" channel for each of ten or so major languages (that's ten or so separate channels), with a daily half-hour lesson that repeats 48 times over 24 hours, so you can tune in at any time and hear the day's lesson. (Accompanying textbook available at every newsstand.)
I want the leading talk radio and local music stations in every country on earth with a radio station to be piped through on their own dedicated channels. On a cold winter day, stuck in traffic, I'd like to be able to tune in to a live radio broadcast from Tahiti, practice my French, and dream of the islands.
Since I don't see that happening soon, I think I'll just hope for the infrastructure for an always-on Internet connection from the car and I'll hope streaming audio on the Net will grow into what I'm looking for.
In the meantime, any suggestions for a good way to record streaming RealAudio into MP3 files that I can then load into a portable MP3 player? I think that's going to be my "satellite radio" if I can figure out a convenient, automated way to do it.
You just have to look at Pearl Harbour and 9/11 to see how badly the US take to threats to the 'homeland'.
I agree. Look at Pearl Harbor and notice how the US treated the defeated Japanese as well as they treated the defeated Germans. You're right that the US takes threats to the homeland badly, if you mean very seriously. You're clearly wrong about it having anything to do with US magnanimity in victory.
I do think C is past its prime. As machines get more speed and memory, and as more and more code is written to run on a server, the percentage of apps that have to make speed or memory size the first priority is steadily decreasing. That means C's market share is steadily decreasing.
Even so, I expect C to far outlast Perl, and not because it's a better language. It's because it is an extreme language: the portable language closest to the metal. Though the need for such a thing continues to diminish as a percentage, it will still be the best tool for certain jobs for a long time to come.
Yes, thanks, Larry. I think Perl was the right tool in the right place at the right time: the duct tape of the Web gold rush.
From what I can tell, though, it appears to have peaked and is now in relative decline. Python is gaining rapidly on Perl in the "scripting language" space. Java, and now PHP, have eroded Perl's popularity in an area it once almost monopolized: Web apps. And its drive to evolve its way from being a useful merger of sed and awk to a full-blown object-oriented programming language may be dragging too much legacy syntax to go much farther.
I'm not trying to insult Perl. It has been enormously helpful to me for years. I'm just seeing signs that at 16, it's probably past its prime.
I don't want to put too fine a point on this, but your concept of "quality per dollar" is not quite what I'm describing. "Since dollars are limited and we want as much quality as possible" sort of implies a situation where we have a fixed budget (hold price constant) and shop for the highest (try to maximize) quality obtainable at that price.
That's opposite from the events I've been involved with or have discussed with colleagues. In our case, we were holding quality constant in the sense that it had to meet certain standards, and we were trying to minimize cost. "What's the cheapest way to get the required quality?" instead of your "What's the highest quality we can get with our limited dollars?"
As I said, it's not really one extreme or the other. People really are aware of the added costs that can come from lower quality, but the emphasis really does tend to be on cost savings.
I've been a part of this outsourcing decision-making process on a few occasions and discussed it with others like me who were making their own decisions.
I'm not personally aware of a single case where it was thought that work being outsourced to India would be higher in quality. Just that it would meet the required quality threshold at a dramatically lower price.
And these decisions are thumbs down as often as thumbs up. (Actually, far more often, but that's must my personal experience. I don't know how representative it is.) At times, we conclude that the quality isn't likely to be good enough, or that the savings isn't likely to be large enough, after factoring in the logistical overhead, to be worth doing.
And companies large enough to open their own offices in India may do critical work there, but I'm not personally aware of any decisions where critical code was outsourced. The ones I've seen have always been the sort of "well, if that system has some problems for a while it won't kill us" type of utility projects.
Worked fine for thousands of years.
Sure. And if that's a good argument for you, then you can enjoy the benefits of that system by moving someplace like Afghanistan or Bolivia.
Is it better to have every single employee be in a constant state of entry-level skill when there are no entry-level careers?
"Entry-level" is an interesting notion. By your definition, all Java programmers in 1995 were "entry-level Java programmers", but what does that mean? Being new to a new specialty is not the same as being new to an old one. The former includes people who are quite successful, while the latter is just an ordinary newbie.
Every Windows laptop has two buttons built in and the most popular models even have an equivalent of the scroll wheel. On ThinkPads, Dells, and others, I can scroll a window without moving my fingers off the center of the keyboard. It's so convenient that I never use an external mouse and almost never have any use for a scrollbar. I can scroll any window in two dimensions by simply pointing anywhere in the window and moving only my right index finger off the home keys.
None of this is possible on any Macintosh laptop. Apple's primitive mouse standard is a real problem.