Some "third-world" countries have difficulty keeping track of their population, in other words, some people simply are not registered on any lists. For those countries, using biometrics for voting actually makes sense, as it allows for "unregistered" people but disallows them from voting twice
I'd like to see a "biometrics" system that
a) never falsely identifies two different people as being the same, as this would result in turning away a legitimate voter, and
b) identifies the same voter differently sufficiently rarely that it is difficult for a voter to trick the machine into allowing him/her to vote twice, and
c) is cheaper than traditional voter registration systems.
This strikes me as, a best, a question for research.
If you simply install a firewall filter that blocks the outgoing spam mail, the spammers can never figure it out and you're making money for nothing.
It'd be trivial for them to detect this--seed the mailing lists they give you with a few addresses that forward to them. I believe the same thing is established practice in the world of mass (non-electronic) mailings.
Sales tax and property tax are the ones that have the most "positive" local impact - much of that money stays at least in your own county instead of being sent to Des Moines or Washington.
Though given that the places I've lived have been at the high end of the income scale, I'm not sure they really need my taxes as much either.
Having counties and states that consistently fail to, say, fund decent public schools, *does* end up hurting us all in the long run.
Your credit card statements wouldn't be enough. For example, IIRC
shipping charges are exempt (at least in some places). Also, some
on-line retailers already charge the appropriate sales tax (companies
that already have a presence in your state are supposed to track and
charge sales tax, and I know at least some do).
Yeah, fair enough. If I was really organized I'd calculate the tax in advance when I entered the transaction into GnuCash, which would make figuring stuff out at the end of the year trivial. As it is I think I get close enough not to care. (Tax on shipping isn't going to add up to much, and it seems few if any of the vendors I order from collect sales tax themselves.)
Not only that, being the organized person that I am, clearly I have kept an accurate record of every internet transaction I made in 2003.
Assuming you pay for them all by credit card, all you need is your credit card statements for 2003. I have 12 of those, so it probably takes me half an hour at most to go through them all.
(I do pay this every year--Michigan seems to have had this for a while. Taxes support some things I don't like, but they also support a bunch of things (like schools) that I think are more important than practically anything else I could spend money on, and more generally I believe in the idea that we should all chip in for things we decide need money, so I don't get so upset about paying taxes. I *do* curse a lot at having to the forms, but the use tax is one of the simpler things to figure out.)
Given time, people can become comfortable with anything. Who woulda thunk that we could sleep through traffic noise, normal rail noise, low-flying jet aircraft etc?
I understand your point, but at the same time I wonder if "resigned to" might be more appropriate than "comfortable with". Most of the time I never notice all the noise around me, but every now and then I stop and listen to the sound of the ventilation and the computers humming and the traffic outside and think how nice it would be to be able to just experience something like actual *silence* once in a while, and, for example, to not have to turn up my headphone volume to 11 to hear all the details in a classical recording with a lot of dynamic range[*].
If you stop to listen a moment (or if you think what it's like to be in a power-out), it's actually kind of shocking to realize the amount of noise we put up with on a regular basis.
--Bruce Fields
[*] It might make an interesting study, by the way, to look at the way music has evolved in relation to the ambient noise around us. Try listening to any modern pop album next time you're driving on the highway, and then try listening to a string quartet. The string quartet will be almost impossible to follow, but the pop songs will still make sense.
Another reason to buy gigabit is if you are planning ahead.
A cheap 100Mbit switch today can be as little as $20. And fine 100Mbit interfaces are built into even the lowest end machines.
In a two years, when I upgrade to machines fast enough to know the difference, the interest on the difference between the cost of my switch and yours might buy me a switch better than the gigabit switch you just bought. (OK, I'm a little optimistic there. But you get the point.)
The cheapest 100mbit switches are $20, and the cheapest 1000mbit switches are over $100. May be no big deal, depending on your budget, but "tiny price premium" is a bit of an exageration....
In all seriousness, you make an excellent point on a philosophical level. Can you cite any concrete examples?
Consider a hypothetical political activist that wants to tape broadcasts of the president saying contradictory things on two different occasions, and use these recordings in a documentary. He could just paraphrase the president, but it wouldn't be as effective as actually showing the clips side-by-side. He'd obviously like the largest possible distribution of the resulting documentary, without getting hung up on legal problems having to do with the tools used to capture the original broadcasts.
A court reviewing such cases after the fact has the chance to weight first amendment, fair use, and other concerns, to arrive at a balanced decision in a complicated case. A device plugged into your TV can't do that.
I hate to sound all Princess Leia, but they keep piling this nonsense on, and we keep ignoring it/circumventing it (and ignoring the laws against circumvention). At some point the whole thing becomes a joke and enforcement becomes impossible.
This is all well and good if you're a consumer who just wants to watch the stuff and maybe keep a personal recording or two.
What if you actually want to use outlawed tools for research or political activism or your own art? Then your violations are public knowledge, and you no longer depend on flying below the radar.
If you do a full-fledged security analysis, the system is "secure" if the cost of breaking the system is less then the value the system protects.
OK, let's take that definition.
DRM systems will be designed to protect billions of dollars of content. Billions of dollars will be able to crack any remotely feasible DRM scheme. Once cracked, the content can be freely copied by all.
The system isn't succesfully cracked, from the DRM proponents point of view, unless a significant portion of those devices are actually compromised. So you have to take into account the cost of distributing the crack. You gloss over that cost: "once cracked, the content can be freely copied by all"; but I think that cost may actually be significant, given the likely aggressive persuit of sites hosting such cracks, and given the likelihood that in the near future an actual *hardware* crack will be required.
Sure, you can cut down the number of crackers and you might even make it unfeasible, but you can never make it impossible.
Impossible? Who cares? Making a crack difficult is all that matters, both to the pro- and anti- DRM people.
Most of what we think of as our shared popular culture these days is, alas, centered around proprietary copyrighted works. In a vibrant, creative, democratic society, we need to be able to quote from, creatively reuse, and criticize copyrighted works. I don't much care whether such activities are still theoretically possible--if they become difficult and illegal then 90% of the battle is already lost.
It should be possible a music educator to use quotes from recordings to use as examples for students.
It should be possible for a political activist to record a video clip from his or her TV and rebroadcast it as a part of a critique of the original broadcast.
It should be possible for a South Park fan to put together their own episode entirely by pasting together existing episodes in a creative way. As long as they don't pass off as their own things that aren't, as long as they're adding ideas of their own, I think society only benefits.
It would be a shame if the above activities could only be carried out furtively and with difficulty, only using special equipment or only with special permission.
This is nonsense. Encryption systems may be practically uncrackable. Encryption systems that have to decrypt the "protected" contents for you so that you can listen to them will never be in the least bit secure. If you can hear it you can record it. There is no getting around this.
You can outlaw recordering equipment; or legislate that it include watermark-detection technology.
Printers are getting built-in currency detection, old-fashioned analog VCR's are getting their own copy protection--macrovision isn't uncrackable, but it's annoying at least--the DRM folks are talking about "closing the analog hole"--so that isn't farfetched.
Clearly DRM is hard, but with tamper-resistent security subsystems being built into new machines and such, it's not obvious to me who wins this arms race; certainly I wouldn't go so far to say that "The entire idea of DRM is, on the face of it, futile."
At the very least, DRM is likely to make things that should be easy very annoying, and force us to do things furtively that we should be able to do openly. So I think it's sensible to worry.
But if the book is still in copyright, I can't legally give those photocopies to someone else
That's an oversimplification, of course. To give just one example, I believe that photocopying one chapter from a book to distribute to students in a class for educational reasons, charging them no more than the cost of the coyping itself, has generally been held to be fair use.
IFF there is a clear need, build from source- a 5% speed optimization may not be worth it (that's the prof's call). A 50% speed improvement (unlikely, but possible)...
Could you give an example of a 50% speed improvement gained from tweaking compile options?
I loved being "taught" what the examples showed and given a graded homework assignment only to find that 90% of the problems could not be solved with the given examples.
Sorry. As a calculus teacher, my job isn't to teach you a step-by-step program for, say, maximizing a smooth function of two variables with a unique maximum on an open interval. You don't have to understand a darned thing to do that.
My job is to teach you some underlying concepts, and to give you practice using those concepts as tools to solve a variety of problems.
This means that while I will give students lots of examples, explain concepts as clearly as I possibly can, and do everything to help, I will *always* assign problems that require fundamentally different solutions from the solutions given in any of the examples.
I've seen a lot of frustrated freshman who've learned over the years to do homework by skimming a chapter quickly (if at all) before looking for the example that gives them a template to solve the particular problem. You have to get past that.
The "Conflict Manager," as they were called, actually followed a script for the meeting, from a paper in plain view of those in attendance (the two kids that were fighting).
In fairness to them, this may have been on purpose: they were probably clinging to the distant hope that, given enough repetition, eventually the kids in question would figure out how to go through the darned script on their own....
I'm upset at how little people bother to actually pay attention when driving, and relying on some device to warn you if your manuever could potentially kill someone or be safe is just insane!
Amen. Learn to look and pay attention. I've actually seen drivers that clearly *rely* on people honking when they change lanes, rather than checking first.
In general the well-meant trend towards "safer" vehicles has resulted in faster, less attentive drivers, the result being *less* safety for everyone that's not in a metal cage. But of course, nobody notices, because bike/ped accidents don't make a serious dent in the statistics, because--guess what--more people have just given up on getting around under their own power entirely....
There's obviously been an arms race developing for a long time between the people running queries and the people with sites that might be returned by queries. Has anyone thought about what the likely endgame is? To me it seems possible that good impartial search engines are just doomed. How can you write algorithms that automatically read pages and determine their relevance to particular subjects in the face of web-page creators who will do anything to get ranked highly?
And it's not enough for your ranking method to be a little bit obscure or hard-to-understand; any search engine now has to face the prospect that the economy is capable of supporting smart poeple to work full time on figuring out how to break your ranking algorithm.
It's not hard to imagine a future where any search engine is either manually maintained (like the various web directories) or completely advertiser-run.
--Bruce Fields
Re:A Few Quick Bits of Wisdom for Mr. Spolsky
on
Joel Rants About Resumes
·
· Score: 3, Insightful
There are a lot of problems that could be solved more easily if there wasn't always someone jumping out to call "racist" anytime someone points out a statistical truth about a demographic. Hey, is it racist to say that American Jews are better educated than the general population?No?
I think that one needs to be careful to be extremely precise when making such statements. A statement like, say, "men are better at math than women" is often taken to mean any of:
The distributions of men's scores on math tests has a higher mean|median|whatever than the distribution of women's scores
Every man is better at math than every woman.
Most men have inherent mathematical abilities that most women do not.
All men have inherent mathematical abilities that all women do not.
Most men have more mathematical education than most women.
Men are naturally better at math than women (so where there are female mathematicians, this is evidence of something "unnatural")
Men are on average more interested in math than women
Mathematical ability is a masculine trait (so men or women may be more or less good at math, but that makes them more or less "masculine")
Etc., etc.; some of these versions are more or less true (or, more importantly, more or less testable) than others. It's in the slipping between all of these than I think people often unknowningly cover up what is essentially sexist thinking. So I think it's important to make sure you say precisely what you mean.
Many applicants came to the job fair dressed in non-formal attire. This is not good. At least, wear a shirt and tie. Don't roll out of bed and throw on some jeans, take the time to look presentable.
This is good advice, and it's generally better to err on the side of dressing up rather than dressing down, but I'd add that you can overdo it. I've made the mistake of showing up in full business regalia to interview or just hang out in what turned out to be a very casual environment. I think this all comes under the category of doing your homework--as long as you're serious enough about the job to want to go talk to the potential employer, you should also do a minimum of research on the dress standards.
Again and again, I see job positions for which the applicant is asked to submit a resume via a textbox in a web form. Usually, no mention is given of what format is allowed (Plain ASCII? HTML? PDF? Tex?), so one pretty much has to assume least-common denominator, and submit in ASCII. Then, one has to pray to the line-width gods that the end product (printed out? online?) will not look too horrible compared to what you just put in.
I use a perl script that formats an ascii resume automatically from an xml-like source with a configurable line length. I usually break at about 70 characters and have made an effort to make it look fairly clear and well-organized, something which doesn't require fancy formatting--look at well-written email or usenet posts for ideas. I take the output from the script, copy it into the body of the email/web form/whatever, edit it a bit for the participient, then add a paragraph or so of introduction at the top. It's very plain but I like to think the result is pretty readable.
Maybe the problem is that after sending out 6000 resumes
Um, you're not applying for jobs, you're spamming, and it's obvious to everyone who reads your resumes; with 6000 resumes, you can't have taken the time to even figure out the most rudimentary facts--are you sending this resume to a valid address? Is the potential employer likely to have any interest in you, or you in them?
You need to send out 60 resumes (and maybe really only devote a full effort to 6 of them) and make sure you do some basic homework on each position; ideally you and the potential employer should already know a lot about each other before you get to any sort of formal hiring process.
and cover letters only to receive rejects letters from about 40% while being completely ignored by the rest has led me to believe that spending a great deal of time on each application/resume/cover letter I send out for Yet Another Job Opening would consume an amount of time equivelent to a full-time job?
Duh, yes, looking for work *is* a full-time job.... You need a trip to your local library; as the author of this rant says, this is all in chapter 1 of every job-search or resume book.
Galeon supports mouse gestures for this. Left-down is previous, right-down is next, and up-left-right-up is table of contents. These gestures use link rel tags, if present.
Hmm. And I just noticed mozilla will use the link rel tags too (view->show/hide->site navigation bar), though I can't seem to find keyboard shortcuts. Also I can't figure out whether lynx does this.
The info format was created a long time ago.
At that time, HTML didn't yet exist (or, at least, wasn't ubiquitous as it is now), so info made at least some sense (although I've always preferred man pages and n/troff docs myself).
Nowadays, however, it makes no sense at all to continue with info when HTML/XML is so common.
All of the info docs should be translated to HTML or XML and the old, obsolete info format should be abandoned.
There are still some things I miss when using html documentation that I appreciate in info:
structured navigation: u, n, p take me, respectively, up a level, to the next section, and to the previous section. My understanding is that html gives the ability to express that kind of structure but that most browsers don't currently provide a convenient user interface for it; so maybe that'll change some day.
easy searching: ^S and / are extremely helpful; no browser I use seems to provide similar functionality for documents split between multiple pages.
Slashdot Mirror
I'd like to see a "biometrics" system that
This strikes me as, a best, a question for research.
--Bruce Fields
It'd be trivial for them to detect this--seed the mailing lists they give you with a few addresses that forward to them. I believe the same thing is established practice in the world of mass (non-electronic) mailings.
--Bruce Fields
Though given that the places I've lived have been at the high end of the income scale, I'm not sure they really need my taxes as much either.
Having counties and states that consistently fail to, say, fund decent public schools, *does* end up hurting us all in the long run.
--Bruce Fields
Yeah, fair enough. If I was really organized I'd calculate the tax in advance when I entered the transaction into GnuCash, which would make figuring stuff out at the end of the year trivial. As it is I think I get close enough not to care. (Tax on shipping isn't going to add up to much, and it seems few if any of the vendors I order from collect sales tax themselves.)
--Bruce Fields
Assuming you pay for them all by credit card, all you need is your credit card statements for 2003. I have 12 of those, so it probably takes me half an hour at most to go through them all.
(I do pay this every year--Michigan seems to have had this for a while. Taxes support some things I don't like, but they also support a bunch of things (like schools) that I think are more important than practically anything else I could spend money on, and more generally I believe in the idea that we should all chip in for things we decide need money, so I don't get so upset about paying taxes. I *do* curse a lot at having to the forms, but the use tax is one of the simpler things to figure out.)
--Bruce Fields
I understand your point, but at the same time I wonder if "resigned to" might be more appropriate than "comfortable with". Most of the time I never notice all the noise around me, but every now and then I stop and listen to the sound of the ventilation and the computers humming and the traffic outside and think how nice it would be to be able to just experience something like actual *silence* once in a while, and, for example, to not have to turn up my headphone volume to 11 to hear all the details in a classical recording with a lot of dynamic range[*].
If you stop to listen a moment (or if you think what it's like to be in a power-out), it's actually kind of shocking to realize the amount of noise we put up with on a regular basis.
--Bruce Fields
[*] It might make an interesting study, by the way, to look at the way music has evolved in relation to the ambient noise around us. Try listening to any modern pop album next time you're driving on the highway, and then try listening to a string quartet. The string quartet will be almost impossible to follow, but the pop songs will still make sense.
A cheap 100Mbit switch today can be as little as $20. And fine 100Mbit interfaces are built into even the lowest end machines.
In a two years, when I upgrade to machines fast enough to know the difference, the interest on the difference between the cost of my switch and yours might buy me a switch better than the gigabit switch you just bought. (OK, I'm a little optimistic there. But you get the point.)
Anyway, that's what I call thinking ahead....
The cheapest 100mbit switches are $20, and the cheapest 1000mbit switches are over $100. May be no big deal, depending on your budget, but "tiny price premium" is a bit of an exageration....
Nothing that halves in price every year or two is a smart investment unless you can really benefit from it *now*.
Consider a hypothetical political activist that wants to tape broadcasts of the president saying contradictory things on two different occasions, and use these recordings in a documentary. He could just paraphrase the president, but it wouldn't be as effective as actually showing the clips side-by-side. He'd obviously like the largest possible distribution of the resulting documentary, without getting hung up on legal problems having to do with the tools used to capture the original broadcasts.
A court reviewing such cases after the fact has the chance to weight first amendment, fair use, and other concerns, to arrive at a balanced decision in a complicated case. A device plugged into your TV can't do that.
--Bruce Fields
This is all well and good if you're a consumer who just wants to watch the stuff and maybe keep a personal recording or two.
What if you actually want to use outlawed tools for research or political activism or your own art? Then your violations are public knowledge, and you no longer depend on flying below the radar.
--Bruce Fields
OK, let's take that definition.
The system isn't succesfully cracked, from the DRM proponents point of view, unless a significant portion of those devices are actually compromised. So you have to take into account the cost of distributing the crack. You gloss over that cost: "once cracked, the content can be freely copied by all"; but I think that cost may actually be significant, given the likely aggressive persuit of sites hosting such cracks, and given the likelihood that in the near future an actual *hardware* crack will be required.
Impossible? Who cares? Making a crack difficult is all that matters, both to the pro- and anti- DRM people.
Most of what we think of as our shared popular culture these days is, alas, centered around proprietary copyrighted works. In a vibrant, creative, democratic society, we need to be able to quote from, creatively reuse, and criticize copyrighted works. I don't much care whether such activities are still theoretically possible--if they become difficult and illegal then 90% of the battle is already lost.
It would be a shame if the above activities could only be carried out furtively and with difficulty, only using special equipment or only with special permission.
--Bruce Fields
You can outlaw recordering equipment; or legislate that it include watermark-detection technology.
Printers are getting built-in currency detection, old-fashioned analog VCR's are getting their own copy protection--macrovision isn't uncrackable, but it's annoying at least--the DRM folks are talking about "closing the analog hole"--so that isn't farfetched.
Clearly DRM is hard, but with tamper-resistent security subsystems being built into new machines and such, it's not obvious to me who wins this arms race; certainly I wouldn't go so far to say that "The entire idea of DRM is, on the face of it, futile."
At the very least, DRM is likely to make things that should be easy very annoying, and force us to do things furtively that we should be able to do openly. So I think it's sensible to worry.
--Bruce Fields
That's an oversimplification, of course. To give just one example, I believe that photocopying one chapter from a book to distribute to students in a class for educational reasons, charging them no more than the cost of the coyping itself, has generally been held to be fair use.
Here's a reference with some further details on copying for educational purposes. (Not that educational justifies any copying, or that it is the only such justification. But it's one good source of examples.)
--Bruce Fields
Could you give an example of a 50% speed improvement gained from tweaking compile options?
--Bruce Fields
Sorry. As a calculus teacher, my job isn't to teach you a step-by-step program for, say, maximizing a smooth function of two variables with a unique maximum on an open interval. You don't have to understand a darned thing to do that.
My job is to teach you some underlying concepts, and to give you practice using those concepts as tools to solve a variety of problems.
This means that while I will give students lots of examples, explain concepts as clearly as I possibly can, and do everything to help, I will *always* assign problems that require fundamentally different solutions from the solutions given in any of the examples.
I've seen a lot of frustrated freshman who've learned over the years to do homework by skimming a chapter quickly (if at all) before looking for the example that gives them a template to solve the particular problem. You have to get past that.
--Bruce Fields
In fairness to them, this may have been on purpose: they were probably clinging to the distant hope that, given enough repetition, eventually the kids in question would figure out how to go through the darned script on their own....
--Bruce Fields
Amen. Learn to look and pay attention. I've actually seen drivers that clearly *rely* on people honking when they change lanes, rather than checking first.
In general the well-meant trend towards "safer" vehicles has resulted in faster, less attentive drivers, the result being *less* safety for everyone that's not in a metal cage. But of course, nobody notices, because bike/ped accidents don't make a serious dent in the statistics, because--guess what--more people have just given up on getting around under their own power entirely....
--Bruce Fields
There's obviously been an arms race developing for a long time between the people running queries and the people with sites that might be returned by queries. Has anyone thought about what the likely endgame is? To me it seems possible that good impartial search engines are just doomed. How can you write algorithms that automatically read pages and determine their relevance to particular subjects in the face of web-page creators who will do anything to get ranked highly?
And it's not enough for your ranking method to be a little bit obscure or hard-to-understand; any search engine now has to face the prospect that the economy is capable of supporting smart poeple to work full time on figuring out how to break your ranking algorithm.
It's not hard to imagine a future where any search engine is either manually maintained (like the various web directories) or completely advertiser-run.
--Bruce Fields
I think that one needs to be careful to be extremely precise when making such statements. A statement like, say, "men are better at math than women" is often taken to mean any of:
Etc., etc.; some of these versions are more or less true (or, more importantly, more or less testable) than others. It's in the slipping between all of these than I think people often unknowningly cover up what is essentially sexist thinking. So I think it's important to make sure you say precisely what you mean.
--Bruce Fields
This is good advice, and it's generally better to err on the side of dressing up rather than dressing down, but I'd add that you can overdo it. I've made the mistake of showing up in full business regalia to interview or just hang out in what turned out to be a very casual environment. I think this all comes under the category of doing your homework--as long as you're serious enough about the job to want to go talk to the potential employer, you should also do a minimum of research on the dress standards.
--Bruce Fields
I use a perl script that formats an ascii resume automatically from an xml-like source with a configurable line length. I usually break at about 70 characters and have made an effort to make it look fairly clear and well-organized, something which doesn't require fancy formatting--look at well-written email or usenet posts for ideas. I take the output from the script, copy it into the body of the email/web form/whatever, edit it a bit for the participient, then add a paragraph or so of introduction at the top. It's very plain but I like to think the result is pretty readable.
--Bruce Fields
Um, you're not applying for jobs, you're spamming, and it's obvious to everyone who reads your resumes; with 6000 resumes, you can't have taken the time to even figure out the most rudimentary facts--are you sending this resume to a valid address? Is the potential employer likely to have any interest in you, or you in them?
You need to send out 60 resumes (and maybe really only devote a full effort to 6 of them) and make sure you do some basic homework on each position; ideally you and the potential employer should already know a lot about each other before you get to any sort of formal hiring process.
Duh, yes, looking for work *is* a full-time job.... You need a trip to your local library; as the author of this rant says, this is all in chapter 1 of every job-search or resume book.
--Bruce Fields
Hmm. And I just noticed mozilla will use the link rel tags too (view->show/hide->site navigation bar), though I can't seem to find keyboard shortcuts. Also I can't figure out whether lynx does this.
--Bruce Fields
There are still some things I miss when using html documentation that I appreciate in info:
--Bruce Fields