Integration of higher and higher technology into daily life isn't one of my goals, bucko. Making each day better than the one before is, of course, and sometimes technology helps that. But the mere technology for its own sake doesn't improve anything, or the people in server rooms would be the happiest motherfuckers on the planet.
This goes double for advertising technology. The point of a billboard is to make you think about something other than what you're thinking about when you're near it. Improving the ability of people with money to distract me from my life might benefit somebody, but it sure as hell isn't me.
Could you point out this right wing media plz? It's been quite a while since there was a right wing centered media outlet, everything is liberal 24 hour news like CNN.
Intriguing. Conservatives scream about the liberal media bias. Liberals scream about the conservative media bias. Will y'all get back to me when you let me know who wins at claiming to be the bigger loser?
Personally, I'm upset about the dumb-ass media bias. There's a total of one high-quality US daily newspaper and zero high-quality US newsweeklies. I have to buy my newsweekly from fucking England.
Frankly, I liked Salon a lot, though I found some of their writers to be ridiculously left. However, with the bigger stories they always seemed to have multiple writers with opposing viewpoints writing about the same subject. I found Salon to be mostly objective.
Agreed! Overall, I'd say that Salon has a very lefty viewpoint, but relatively little lefty bias. Americans seem easily confused about the difference. Thus the slop you see in most newspapers, where "objective" means asking a person on either side of the story and printing their comments withought thought or analysis.
Let's face it, I must have screwed up real bad to come out of the dot-com boom with nothing but pocket change eh?
Living in San Francisco, my base assumption is that anybody who got rich during the boom did it by a) cynically cashing out, b) spewing bullshit, or c) screwing their colleagues as soon as they got the chance.
That's not true of everybody, of course. Some of the people who made bank were smart, honest, and worked their asses off. But the fact that you built a major website, tried selling out, and ended up with lint mainly tells me that you were honest, sincerely believed in what you did, and were too nice when negotiating with sharks.
All of the opinion mags above target roughly the same demographic as Salon (if not necessarily the same ideologies), and all have equivalent- or higher-quality writing, established reputations, and an existing subscriber base to draw from. The surprising thing is that anyone ever thought Salon's business model would surpass them.
Well, the notion that the web would be different was one a lot of people had. The theories behind it weren't bad: Distribution is much less of a headache; electrons are cheap and do what you tell 'em. You have much better information about your readers, so web ads should be more valuable. The Internet, even in English, has a much broader reach, so your potential readership is larger. And the web was going to be big, big, big!
Of course, it didn't pan out that way. A big part of the problem was that Salon was competing with people who had a big pile of VC money that they were determined to blow. As you point out, a lot of those labor-of-love magazines have the same problem: they're competing against people who don't have to break even.
And it doesn't help that everybody, as Slashdot demonstrates, seems to have gotten the idea that stuff on the web should be free.
They could have saved money in other ways. Why the office in San Francisco? Could they have not officed in, say, a small strip mall office in Lubbock TX and conducted the publisher/reporter/journalist interactions via e-mail and telephone? Being located in SFO really racks up the bills--everything is more expensive out there.
Hey, that's a great idea! Perhaps you should call up the Wall Street Journal and the New York Times and let them in on the secret. And now that you mention it, maybe you could get one of the movie studios to move to Wymore, Nebraska. I hear they only have one traffic light, so commutes should be much easier.
Personally, I doubt it would work. Smart creative people like to live in interesting places. They like to live near other smart, creative people. And they like to live in places where there are lots of companies who will hire them. If they're going to move somewhere for work, every journalist I know would rather move to Uzbekistan than Lubbock.
Note that at least two of the tech centers (SFO, BOS) grew up around major research universities. That's no accident.
With regards to your USA Today comment--no wonder it took them so long to be profitable. Have you seen their office building in Northern VA?
Their building had relatively little to do with it. Any new major magazine faces a long climb to profitability. The problem is much worse for a daily. It's the standard bootstrapping problem: until you reach a certain level of quality, you can't get the subscribers. But you can't afford to reach the level of quality until you get them.
Slashdot works fine when somebody is already writing about the topic of interest and is willing to give their material away for "free" (meaning free or with ads).
Salon (and every other decent magazine) pays people to write new material. Sure, they have stuff from an AP feed, but I can get an AP feed anywhere. What I'm buying with my subscription to Salon (or, say, The Economist) is that new material.
That material costs money to produce and more money to edit. That money has to get to the writers and editors somehow. How would you suggest?
Micropayments could work this way: You initally deposit $25 into your account and then you micropay... when your $25 gets low, you are automatically charged (on your credit card) for an additional $25.
However, this defeats what micropayments try to solve in the first place: paying for just what you want, like an individual story on Salon.
The theory works pretty well for long-distance phone cards, eh?
Heck, maybe that's the solution. Maybe I should just do a micropayment system based on charging to long-distance phone cards. Available everywhere, anonymous, and trusted. Anybody know how phone card settlement works?
I've been freelancing for six years now. I agree with many of the things others have said, especially about having a big savings cushion. But here's my big piece of advice: Getting work will be much harder than you think.
Once you go own your own, you go from being a technical person to a marketing person. Network relentlessly. Join user groups. Collect recommendations. Get and stay in touch with old colleagues. Write articles. Schedule plenty of time for keeping up on the latest skills and the latest buzzwords.
My tip: start out by getting part-time work. Don't quit your day job until you have so much outside work that you can't stand it anymore.
Also, I'd recommend that you read "The Computer Consultant's Handbook"; and Weinberg's "The Secrets of Consulting." The first one's a little old and the second one is focused more on consultants than contractors, but you will find both helpful.
You claim four or five years of qmail experience and still don't understand how qmail works? [...] If this is anything like your four to five years of qmail administration, then I wouldn't count it for much.
Cutely phrased, but in your eagerness to be snide, you missed my point.
Long ago, I was a full-time sysadmin, running large fancy networks of what then passed for cutting-edge unix systems. Eventually, I got sick of being a sysadmin for money; it was turning me into a grumpy BOFH. Today, I do other things for a living; I just maintain a few boxes for my own entertainment.
When I installed qmail five years ago, I figured out how it worked and knew it just fine. But I have to touch my mail server maybe every few months, which is plenty of time to forget all of the details.
Qmail is easy to use in the same way that a trumpet is easy to play: if you learn a lot about it and keep in practice, it's easy. But a novice will have little luck just picking it up and getting some music to come put.
I know a lot of geeks don't get this point about usability, but try reading Don Norman's "The Design of Everyday Things" and maybe that will help you get what I'm saying.
Personally, I wanted db-backed virtual domains, and pretty complex spam rules with exceptions for certain senders, certain recipients, and certain mail servers. For qmail, I just never bothered because it was too much of a pain.
Pure FUD. db-backed virtual domains? Use vpopmail or vmailmgr. Filtering? Use your favorite delivery agent, [...]
Actually, I wanted to do a lot of the rejection at the SMTP layer. Which I could also do for qmail with yet another program or patch, I know. But the point isn't whether it's possible, it's how easy it is. With postfix, it was all in the config files. And even better, the postfix author keeps improving things in response to user needs.
If somebody whose only job is maintaining a big ol' mail server asks me what to run, I'll certainly say "check out qmail". But when a non-expert asks me what they should run, I'll suggest postfix.
qmail follows the UNIX tradition: do one thing and do it well.
Great! I'm happy for it. I'm happy for you. It sounds like you should use qmail. But recognize that circa 99% of the computer-using population doesn't ever touch anything that follows that workbench-style approach.
For most purposes, people just want appliances. A master chef may have a fantastic commercial-grade kitchen but only use an iMac. A master Unix geek may know which four lowercase letters aren't valid arguments to ls, but max out his cooking skills heating frozen burritos in a microwave.
This doesn't mean that the geek or the cook (or the iMac or the microwave) are somehow intrinsically lame; having chosen one thing to master, they just want other parts of their lives to be as easy as possible.
Personally, I needed a mail server that was more an appliance than a workbench. Postfix fills that need for me.
How are qmail control files not readable? Files are named according to their function and only contain the necessary values that you place into them. They are fully documented in the man pages. The values of all the control files can be shown easily by running qmail-showctl. How is this difficult?
Have you run a Postfix system? I've run both Postfix and qmail systems. I used qmail for four or five years. I find Postfix easier to manage. You may like the qmail style better. That's fine: my getting chocolate ice cream is not some sort of implicit assault on your preference for strawberry ice cream.
For postfix, there is a main configuration file, called main.cf. The stock main.cf has extensive commenting, describing many of the common options. When I want to do something with Postfix, I just edit the config file and rummage around a bit. Then I type "sudo postfix reload".
With qmail, I always have to read a bunch of different man pages, tracking down the particular program that happens to handle whatever I need to change. Then I echo some value into some magically named file. Then I have to read the man pages again to figure out exactly which signal I send to which program and/or programs.
As I said, maybe this is just me. Maybe most other people find it easier to think like Dan Bernstein. Maybe I just don't have enough practice adminstering Unix boxes; I've only been doing it for 15 years or so.
What features do you need that are not available in stock qmail?
Go browse the qmail web site. Note the dozens and dozens of links to patches, add-on programs, and other cruft.
Personally, I wanted db-backed virtual domains, and pretty complex spam rules with exceptions for certain senders, certain recipients, and certain mail servers. For qmail, I just never bothered because it was too much of a pain.
Howdy, Russ. You've been a very helpful person on the qmail lists and int the qmail community and I appreciate that. I'm going to try to take your apparent excess of attitude in that context.
You do not need a license to use qmail. Period. End of sentence. And your problem with this is?
My problem is that whatever you call it, Dan Bernstein has been careful to keep legal control of qmail in a way that prevents others from forking. I was very excited when it came out; qmail had a fresh approach that showed a lot of promise. And there it stopped. I decided not to.
Clue: qmail is at version 1.03. qmail has never had a security flaw. Now, if Dan was to change qmail in any way, do you think that would 1) increase qmail's security, 2) decrease qmail's security, or 3) have no effect on qmail's security?
If security were my only concern, I'd just unplug everything from the net and be done with it. Alas, I have other concerns, which I balance with security.
But let me ask you a related question: If people other than DJB are patching qmail to get features they want, do you think that would 1) increase qmail's security, 2) decrease qmail's security, or 3) have no effect on qmail's security versus funneling all the changes through an expert coder who knows the system better than anybody?
Configuration files are easy to read. cat/var/qmail/control/me. Works for me.
That's swell! You should probably use qmail, then. I find Postfix much more easily managed. Many other people do, too.
That, of course, doesn't mean that qmail should change a whit. I gather that DJB has consciously decided that qmail won't be all things to everyone. In which case, you guys should presumably be happy that other packages address other audiences, yes?
What features do you need that qmail does not have?
Honestly, will it do any good to say? If you really don't know of any feature differences between the two and would like to add some things to a qmail 2.0, I'm glad to make suggestions.
But what I suspect will happen is that you will tell me that I can get everything I want with add-on programs A, B, C, D, and E, along with patches X, Y, and Z, plus some jiggery-pokery with various dot files and magically named files. That's been my experience with qmail, and others can verify that by looking at the qmail home page for spam prevention, high-volume servers, and other add-ons.
Then I'll tell you that I'd rather just download the postfix RPM, set a few lines in a config file, and go. At which point I'll receive some more attitude for not being sufficiently smart, elite, tough, or whatever quality it is that makes people willing to deal with a mail server on its terms rather than their own.
But if you're really looking for help understanding the difference between the two mailers, drop by the postfix mailing lists. There are a number of former qmail users there, and I'm sure you can collect a wealth of feature suggestions there.
And before anyone complains about the license of QMail/ezmlm, yes, that sucks.
qmail does not have a license and does not need one[...]
Please don't be obtuse. The legal meaning is as you say, of course. But the common meaning is, "the terms under which the program is available." In the common sense, qmail has a license. And the OP is right, it's a pain.
DJB's insistence on keeping an iron grip on qmail and keeping it at v1.0 for five years is what made me finally change over to Postfix. It's in the same league as qmail regarding security, reliability, and having a good architecture. But unlike qmail, you don't have to apply patches to get needed features. And the configuration files are actually readable and helpful to admins.
Three months after conversion, my only regret is that I didn't switch from qmail to Postfix sooner.
But when you look at the numbers another way, calculating an individual person's likelihood of being involved in a fatal automobile accident in his or her lifetime, the percentages come out very close to zero.
I calculate it to be about 1%, which isn't very close to zero in my book. So figure circa a 10% chance that you'll lose a relative, and awfully good odds that you'll know somebody who was killed in a car wreck.
According to National Safety Council stats, deaths from motor vehicle accidents are more than 40% of all accidental deaths. Your chances of dying in a motor vehicle accident are
4 times your chance of dying from posoning
10 times your chance of drowning
13 times your chances of dying in a fire
59 times your chance of dying in a firearm accident
Given how much people worry about those sorts of accidents, it seems to me that they're inordinately cavalier about automobile accidents.
Well, that's a _good_ thing. A higher population spread will spread the pollution. In a country like Canada where about 5% of the land makes something like 95% of the pollution, this is exactly what's needed. We could be cleaner than Japan if everyone spread out.
At the cost of vast ecosystem destruction and large increases in resource consumption. Every envirogeek I know feels that if people are going to pollute, they should do it in cities, where at least the damage is contained.
And note that Japan, noted for its cleanness, is very dense. They learned how to be clean because of the density. Perhaps we could learn from that.
So, what's your solution in this case, where mass transit is a no go?
Mass transit is a no-go because people made decisions that caused it to end up that way. The question is whether to notice the problem and move to correct it or to continue to use government money to subsidize more bad decisions. There is no easy solution, but some solutions pay off in the long term as well as the short.
Personally, I make sure to live near where I work, and I moved to an urban center that invests in public transport. These days I don't even own a car; I just check them out when I need them. Compared to the typical commuter, I save a lot of time and money, and consume far less of our shared environmental resources than most.
In the long term, we need to charge people properly for the use of shared resources. Road pricing, pollution taxes, and carbon taxes would help the problem a lot. If you give people something for free, they'll just run it into the ground. Thus, your 20 minute wait in traffic and your asthma deaths. The full change we need will take decades, of course, but that's no excuse for not starting now.
Man, this burns me up. A guy grows up in a mud hut, studies his ass off, wins a scholarship, and becomes a crackerjack programmer. Why in the world should somebody here get $100k for a job he'd happily do for a tenth of that?
Now really, I think most of the outsourcing to India is just dumb; most really good software requires a deep understanding of the user and the context. But for the stuff that really can be shipped off to India, why shouldn't they have a crack a it?
The net effect of free trade is that everybody is better off. (The reason why is one of the few non-obvious things in Economics: the principle of Comparative Advantage.)
I am quite concerned about being able to work as an engineer for 20 more years (I've got 11 years already). I think that the corporations will find ways to reduce our salaries. What will you do when your $100K/yr job is gone and the only things around are $30K work at Frys?
That won't happen; Moore's Law and distance society has to go to make good use of even current technology guarantee you a long career. At worst, you'll be doing the same sort of work for $80K/yr.
That is ok by me; the supply and demand imbalance has meant that a lot of complete idiots have been able to make bank. It would do the field some good to have the deadwood shaken out a little. And our salaries won't drop below the salaries for other people with equivalent amounts of education; geek skills apply across enough industries that demand is very broad.
The only difference between a cartel and a union is that unions are legal.
For the most part, I'd agree with your point. But there's a basic asymmetry between employers and employees; it's much harder for an employee to lose his job than it is for an employer to have one employee quit.
That's not a big problem for geeks; our labor market is very liquid. But in industries where it's hard to change jobs and where the employers are generally huge (e.g., the airlines, manufacturers that dominate a particular labor market) I think some collective action is worthwhile; it equalizes the imbalance in negotiating power.
But too often, unions just turn into a job protection racket. (Anybody who has ever tried to set up anything big in a convention hall in Chicago will discover how impressively a union can abuse its power.) Smart unions recognize that they are playing a positive-sum game with their employers; smart employers recognize that keeping their employees happy is in their long-term interest, and that a union can help with that.
For all the problems that labor unions may have, they are the only organizations that working class people have, and if they never existed, most Americans would be in the position that Chinese workers are in now, lots of work, little pay, and zero rights.
I think unions were an important part of American history, and I have a lot of respect for what they achieved. But you're overstating the case here.
US workers can compete with Chinese workers because we're much more productive than they are. This is mainly because we are much more highly educated. You can't whip somebody into high productivity, not on American levels; that requires the employees to be relatively happy. And our highly liquid labor market makes it easy for us to change jobs when we aren't.
There may be some benefit to unionizing geeks, but scaring people with spooky stories is a poor way to convince people of it.
Give me a good reason. Tell me why this golf course mentioned in the article shouldn't be able to dump a big fugly pile of boulders on their beach.
If it were their beach and theirs alone, you might have a point. It's a public beach.
Moreover, even if it were private, their creation of a seawall changes erosion patterns in the area, harming the property of other people. Your right to do as you please on your property ends when it harms your neighbors.
What good does this really do? Who are they actually protecting?
In Cali, the coast belongs to all of us. Ergo, they're protecting all of us.
These people shouldn't be hailed as heros, they haven't really done anything other then invade the privacy of land owners.
Taking pictures of a city from the air is not an invasion of privacy in this country. And suggesting that taking pictures of public land is somehow an invasion of privacy is just bizarre.
Yes, it's much better to deny people the right to build a seawall, and then spend billions on beach reclamation projects. Sheeeeerrrrr Genius (said in Wile E. Coyote voice)
This is wrong twice over:
The coast is a public resource. A private landowner who was dumb enough to build on eroding land doesn't have any right to build a seawall, anymore than a guy who lives next to a public park has a right to put in a vegetable garden.
Putting up seawalls will require you to have beach reclamation projects. Beaches are the result of erosion; if you stop erosion, the existing sand gets washed away.
That reminds me of the reason they blocked building a bypass route near my area. An environmental group said a bunch of frogs would be killed because of it.
Well, that's just great. Instead we have 20 minute idle times and the city's smog is so bad it kills many asthmatics per year.
Won't someone think of the poor people.
Has it occurred to you that maybe all those people shouldn't have moved to a place with crappy transportation infrastructure? Or that if driving their cars kills asthmatics, maybe the solution is to pollute less rather than destroying more (publicly owned, pollution-cleaning) ecosystem?
More to the point, in the long term, adding more suburban freeway capacity doesn't solve anything. In the studies I've looked at, commuters don't drive less after capacity is added, they just drive farther. They do this mainly by moving yet farther out from the cities where they work.
As far as I can tell, the people are doing a great job of thinking of themselves already. Which is sort of the problem, isn't it?
The beach was a public beach. Putting boulders on a public beach does not exactly cry out for privacy protection. And according to the California Costal Commission, the agency that grants the permits, the golf course didn't have a permit.
Yes, but wouldn't the added benefit of not having your entire property slide into the ocean give you a legitimate claim against the frogs?
The no-seawall stuff isn't just for the little froggies, although destroying a public resource (the ecosystem) for private gain is generally a no-no. Other reasons include
California requires public access to the coast; some seawalls impede that, often intentionally
seawalls on one property can increase erosion on nearby properties
stopping erosion means that beaches aren't replenished, destroying them
So turn it around: Why would being dumb enough to build on an eroding piece of land give you a legitimate claim to build a seawall?
Slashdot Mirror
Integration of higher and higher technology into daily life isn't one of my goals, bucko. Making each day better than the one before is, of course, and sometimes technology helps that. But the mere technology for its own sake doesn't improve anything, or the people in server rooms would be the happiest motherfuckers on the planet.
This goes double for advertising technology. The point of a billboard is to make you think about something other than what you're thinking about when you're near it. Improving the ability of people with money to distract me from my life might benefit somebody, but it sure as hell isn't me.
Could you point out this right wing media plz? It's been quite a while since there was a right wing centered media outlet, everything is liberal 24 hour news like CNN.
Intriguing. Conservatives scream about the liberal media bias. Liberals scream about the conservative media bias. Will y'all get back to me when you let me know who wins at claiming to be the bigger loser?
Personally, I'm upset about the dumb-ass media bias. There's a total of one high-quality US daily newspaper and zero high-quality US newsweeklies. I have to buy my newsweekly from fucking England.
Frankly, I liked Salon a lot, though I found some of their writers to be ridiculously left. However, with the bigger stories they always seemed to have multiple writers with opposing viewpoints writing about the same subject. I found Salon to be mostly objective.
Agreed! Overall, I'd say that Salon has a very lefty viewpoint, but relatively little lefty bias. Americans seem easily confused about the difference. Thus the slop you see in most newspapers, where "objective" means asking a person on either side of the story and printing their comments withought thought or analysis.
Let's face it, I must have screwed up real bad to come out of the dot-com boom with nothing but pocket change eh?
Living in San Francisco, my base assumption is that anybody who got rich during the boom did it by a) cynically cashing out, b) spewing bullshit, or c) screwing their colleagues as soon as they got the chance.
That's not true of everybody, of course. Some of the people who made bank were smart, honest, and worked their asses off. But the fact that you built a major website, tried selling out, and ended up with lint mainly tells me that you were honest, sincerely believed in what you did, and were too nice when negotiating with sharks.
All of the opinion mags above target roughly the same demographic as Salon (if not necessarily the same ideologies), and all have equivalent- or higher-quality writing, established reputations, and an existing subscriber base to draw from. The surprising thing is that anyone ever thought Salon's business model would surpass them.
Well, the notion that the web would be different was one a lot of people had. The theories behind it weren't bad: Distribution is much less of a headache; electrons are cheap and do what you tell 'em. You have much better information about your readers, so web ads should be more valuable. The Internet, even in English, has a much broader reach, so your potential readership is larger. And the web was going to be big, big, big!
Of course, it didn't pan out that way. A big part of the problem was that Salon was competing with people who had a big pile of VC money that they were determined to blow. As you point out, a lot of those labor-of-love magazines have the same problem: they're competing against people who don't have to break even.
And it doesn't help that everybody, as Slashdot demonstrates, seems to have gotten the idea that stuff on the web should be free.
They could have saved money in other ways. Why the office in San Francisco? Could they have not officed in, say, a small strip mall office in Lubbock TX and conducted the publisher/reporter/journalist interactions via e-mail and telephone? Being located in SFO really racks up the bills--everything is more expensive out there.
Hey, that's a great idea! Perhaps you should call up the Wall Street Journal and the New York Times and let them in on the secret. And now that you mention it, maybe you could get one of the movie studios to move to Wymore, Nebraska. I hear they only have one traffic light, so commutes should be much easier.
Personally, I doubt it would work. Smart creative people like to live in interesting places. They like to live near other smart, creative people. And they like to live in places where there are lots of companies who will hire them. If they're going to move somewhere for work, every journalist I know would rather move to Uzbekistan than Lubbock.
Note that at least two of the tech centers (SFO, BOS) grew up around major research universities. That's no accident.
With regards to your USA Today comment--no wonder it took them so long to be profitable. Have you seen their office building in Northern VA?
Their building had relatively little to do with it. Any new major magazine faces a long climb to profitability. The problem is much worse for a daily. It's the standard bootstrapping problem: until you reach a certain level of quality, you can't get the subscribers. But you can't afford to reach the level of quality until you get them.
You're missing the point.
Slashdot works fine when somebody is already writing about the topic of interest and is willing to give their material away for "free" (meaning free or with ads).
Salon (and every other decent magazine) pays people to write new material. Sure, they have stuff from an AP feed, but I can get an AP feed anywhere. What I'm buying with my subscription to Salon (or, say, The Economist) is that new material.
That material costs money to produce and more money to edit. That money has to get to the writers and editors somehow. How would you suggest?
The theory works pretty well for long-distance phone cards, eh?
Heck, maybe that's the solution. Maybe I should just do a micropayment system based on charging to long-distance phone cards. Available everywhere, anonymous, and trusted. Anybody know how phone card settlement works?
I've been freelancing for six years now. I agree with many of the things others have said, especially about having a big savings cushion. But here's my big piece of advice: Getting work will be much harder than you think.
Once you go own your own, you go from being a technical person to a marketing person. Network relentlessly. Join user groups. Collect recommendations. Get and stay in touch with old colleagues. Write articles. Schedule plenty of time for keeping up on the latest skills and the latest buzzwords.
My tip: start out by getting part-time work. Don't quit your day job until you have so much outside work that you can't stand it anymore.
Also, I'd recommend that you read "The Computer Consultant's Handbook"; and Weinberg's "The Secrets of Consulting." The first one's a little old and the second one is focused more on consultants than contractors, but you will find both helpful.
Cutely phrased, but in your eagerness to be snide, you missed my point.
Long ago, I was a full-time sysadmin, running large fancy networks of what then passed for cutting-edge unix systems. Eventually, I got sick of being a sysadmin for money; it was turning me into a grumpy BOFH. Today, I do other things for a living; I just maintain a few boxes for my own entertainment.
When I installed qmail five years ago, I figured out how it worked and knew it just fine. But I have to touch my mail server maybe every few months, which is plenty of time to forget all of the details.
Qmail is easy to use in the same way that a trumpet is easy to play: if you learn a lot about it and keep in practice, it's easy. But a novice will have little luck just picking it up and getting some music to come put.
I know a lot of geeks don't get this point about usability, but try reading Don Norman's "The Design of Everyday Things" and maybe that will help you get what I'm saying.
Pure FUD. db-backed virtual domains? Use vpopmail or vmailmgr. Filtering? Use your favorite delivery agent, [...]
Actually, I wanted to do a lot of the rejection at the SMTP layer. Which I could also do for qmail with yet another program or patch, I know. But the point isn't whether it's possible, it's how easy it is. With postfix, it was all in the config files. And even better, the postfix author keeps improving things in response to user needs.
If somebody whose only job is maintaining a big ol' mail server asks me what to run, I'll certainly say "check out qmail". But when a non-expert asks me what they should run, I'll suggest postfix.
qmail follows the UNIX tradition: do one thing and do it well.
Great! I'm happy for it. I'm happy for you. It sounds like you should use qmail. But recognize that circa 99% of the computer-using population doesn't ever touch anything that follows that workbench-style approach.
For most purposes, people just want appliances. A master chef may have a fantastic commercial-grade kitchen but only use an iMac. A master Unix geek may know which four lowercase letters aren't valid arguments to ls, but max out his cooking skills heating frozen burritos in a microwave.
This doesn't mean that the geek or the cook (or the iMac or the microwave) are somehow intrinsically lame; having chosen one thing to master, they just want other parts of their lives to be as easy as possible.
Personally, I needed a mail server that was more an appliance than a workbench. Postfix fills that need for me.
How are qmail control files not readable? Files are named according to their function and only contain the necessary values that you place into them. They are fully documented in the man pages. The values of all the control files can be shown easily by running qmail-showctl. How is this difficult?
Have you run a Postfix system? I've run both Postfix and qmail systems. I used qmail for four or five years. I find Postfix easier to manage. You may like the qmail style better. That's fine: my getting chocolate ice cream is not some sort of implicit assault on your preference for strawberry ice cream.
For postfix, there is a main configuration file, called main.cf. The stock main.cf has extensive commenting, describing many of the common options. When I want to do something with Postfix, I just edit the config file and rummage around a bit. Then I type "sudo postfix reload".
With qmail, I always have to read a bunch of different man pages, tracking down the particular program that happens to handle whatever I need to change. Then I echo some value into some magically named file. Then I have to read the man pages again to figure out exactly which signal I send to which program and/or programs.
As I said, maybe this is just me. Maybe most other people find it easier to think like Dan Bernstein. Maybe I just don't have enough practice adminstering Unix boxes; I've only been doing it for 15 years or so.
What features do you need that are not available in stock qmail?
Go browse the qmail web site. Note the dozens and dozens of links to patches, add-on programs, and other cruft.
Personally, I wanted db-backed virtual domains, and pretty complex spam rules with exceptions for certain senders, certain recipients, and certain mail servers. For qmail, I just never bothered because it was too much of a pain.
Howdy, Russ. You've been a very helpful person on the qmail lists and int the qmail community and I appreciate that. I'm going to try to take your apparent excess of attitude in that context.
/var/qmail/control/me. Works for me.
You do not need a license to use qmail. Period. End of sentence. And your problem with this is?
My problem is that whatever you call it, Dan Bernstein has been careful to keep legal control of qmail in a way that prevents others from forking. I was very excited when it came out; qmail had a fresh approach that showed a lot of promise. And there it stopped. I decided not to.
Clue: qmail is at version 1.03. qmail has never had a security flaw. Now, if Dan was to change qmail in any way, do you think that would 1) increase qmail's security, 2) decrease qmail's security, or 3) have no effect on qmail's security?
If security were my only concern, I'd just unplug everything from the net and be done with it. Alas, I have other concerns, which I balance with security.
But let me ask you a related question: If people other than DJB are patching qmail to get features they want, do you think that would 1) increase qmail's security, 2) decrease qmail's security, or 3) have no effect on qmail's security versus funneling all the changes through an expert coder who knows the system better than anybody?
Configuration files are easy to read. cat
That's swell! You should probably use qmail, then. I find Postfix much more easily managed. Many other people do, too.
That, of course, doesn't mean that qmail should change a whit. I gather that DJB has consciously decided that qmail won't be all things to everyone. In which case, you guys should presumably be happy that other packages address other audiences, yes?
What features do you need that qmail does not have?
Honestly, will it do any good to say? If you really don't know of any feature differences between the two and would like to add some things to a qmail 2.0, I'm glad to make suggestions.
But what I suspect will happen is that you will tell me that I can get everything I want with add-on programs A, B, C, D, and E, along with patches X, Y, and Z, plus some jiggery-pokery with various dot files and magically named files. That's been my experience with qmail, and others can verify that by looking at the qmail home page for spam prevention, high-volume servers, and other add-ons.
Then I'll tell you that I'd rather just download the postfix RPM, set a few lines in a config file, and go. At which point I'll receive some more attitude for not being sufficiently smart, elite, tough, or whatever quality it is that makes people willing to deal with a mail server on its terms rather than their own.
But if you're really looking for help understanding the difference between the two mailers, drop by the postfix mailing lists. There are a number of former qmail users there, and I'm sure you can collect a wealth of feature suggestions there.
Please don't be obtuse. The legal meaning is as you say, of course. But the common meaning is, "the terms under which the program is available." In the common sense, qmail has a license. And the OP is right, it's a pain.
DJB's insistence on keeping an iron grip on qmail and keeping it at v1.0 for five years is what made me finally change over to Postfix. It's in the same league as qmail regarding security, reliability, and having a good architecture. But unlike qmail, you don't have to apply patches to get needed features. And the configuration files are actually readable and helpful to admins.
Three months after conversion, my only regret is that I didn't switch from qmail to Postfix sooner.
I calculate it to be about 1%, which isn't very close to zero in my book. So figure circa a 10% chance that you'll lose a relative, and awfully good odds that you'll know somebody who was killed in a car wreck.
According to National Safety Council stats, deaths from motor vehicle accidents are more than 40% of all accidental deaths. Your chances of dying in a motor vehicle accident are
- 4 times your chance of dying from posoning
- 10 times your chance of drowning
- 13 times your chances of dying in a fire
- 59 times your chance of dying in a firearm accident
Given how much people worry about those sorts of accidents, it seems to me that they're inordinately cavalier about automobile accidents.Well, that's a _good_ thing. A higher population spread will spread the pollution. In a country like Canada where about 5% of the land makes something like 95% of the pollution, this is exactly what's needed. We could be cleaner than Japan if everyone spread out.
At the cost of vast ecosystem destruction and large increases in resource consumption. Every envirogeek I know feels that if people are going to pollute, they should do it in cities, where at least the damage is contained.
And note that Japan, noted for its cleanness, is very dense. They learned how to be clean because of the density. Perhaps we could learn from that.
So, what's your solution in this case, where mass transit is a no go?
Mass transit is a no-go because people made decisions that caused it to end up that way. The question is whether to notice the problem and move to correct it or to continue to use government money to subsidize more bad decisions. There is no easy solution, but some solutions pay off in the long term as well as the short.
Personally, I make sure to live near where I work, and I moved to an urban center that invests in public transport. These days I don't even own a car; I just check them out when I need them. Compared to the typical commuter, I save a lot of time and money, and consume far less of our shared environmental resources than most.
In the long term, we need to charge people properly for the use of shared resources. Road pricing, pollution taxes, and carbon taxes would help the problem a lot. If you give people something for free, they'll just run it into the ground. Thus, your 20 minute wait in traffic and your asthma deaths. The full change we need will take decades, of course, but that's no excuse for not starting now.
Exporting IT and programming jobs overseas.
Man, this burns me up. A guy grows up in a mud hut, studies his ass off, wins a scholarship, and becomes a crackerjack programmer. Why in the world should somebody here get $100k for a job he'd happily do for a tenth of that?
Now really, I think most of the outsourcing to India is just dumb; most really good software requires a deep understanding of the user and the context. But for the stuff that really can be shipped off to India, why shouldn't they have a crack a it?
The net effect of free trade is that everybody is better off. (The reason why is one of the few non-obvious things in Economics: the principle of Comparative Advantage.)
I am quite concerned about being able to work as an engineer for 20 more years (I've got 11 years already). I think that the corporations will find ways to reduce our salaries. What will you do when your $100K/yr job is gone and the only things around are $30K work at Frys?
That won't happen; Moore's Law and distance society has to go to make good use of even current technology guarantee you a long career. At worst, you'll be doing the same sort of work for $80K/yr.
That is ok by me; the supply and demand imbalance has meant that a lot of complete idiots have been able to make bank. It would do the field some good to have the deadwood shaken out a little. And our salaries won't drop below the salaries for other people with equivalent amounts of education; geek skills apply across enough industries that demand is very broad.
The only difference between a cartel and a union is that unions are legal.
For the most part, I'd agree with your point. But there's a basic asymmetry between employers and employees; it's much harder for an employee to lose his job than it is for an employer to have one employee quit.
That's not a big problem for geeks; our labor market is very liquid. But in industries where it's hard to change jobs and where the employers are generally huge (e.g., the airlines, manufacturers that dominate a particular labor market) I think some collective action is worthwhile; it equalizes the imbalance in negotiating power.
But too often, unions just turn into a job protection racket. (Anybody who has ever tried to set up anything big in a convention hall in Chicago will discover how impressively a union can abuse its power.) Smart unions recognize that they are playing a positive-sum game with their employers; smart employers recognize that keeping their employees happy is in their long-term interest, and that a union can help with that.
For all the problems that labor unions may have, they are the only organizations that working class people have, and if they never existed, most Americans would be in the position that Chinese workers are in now, lots of work, little pay, and zero rights.
I think unions were an important part of American history, and I have a lot of respect for what they achieved. But you're overstating the case here.
US workers can compete with Chinese workers because we're much more productive than they are. This is mainly because we are much more highly educated. You can't whip somebody into high productivity, not on American levels; that requires the employees to be relatively happy. And our highly liquid labor market makes it easy for us to change jobs when we aren't.
There may be some benefit to unionizing geeks, but scaring people with spooky stories is a poor way to convince people of it.
Check out the California Costal Commission, including their page on Legal and Legislative Information and the list of their permanent responsibilities
Give me a good reason. Tell me why this golf course mentioned in the article shouldn't be able to dump a big fugly pile of boulders on their beach.
If it were their beach and theirs alone, you might have a point. It's a public beach.
Moreover, even if it were private, their creation of a seawall changes erosion patterns in the area, harming the property of other people. Your right to do as you please on your property ends when it harms your neighbors.
What good does this really do? Who are they actually protecting?
In Cali, the coast belongs to all of us. Ergo, they're protecting all of us.
These people shouldn't be hailed as heros, they haven't really done anything other then invade the privacy of land owners.
Taking pictures of a city from the air is not an invasion of privacy in this country. And suggesting that taking pictures of public land is somehow an invasion of privacy is just bizarre.
This is wrong twice over:
That reminds me of the reason they blocked building a bypass route near my area. An environmental group said a bunch of frogs would be killed because of it.
Well, that's just great. Instead we have 20 minute idle times and the city's smog is so bad it kills many asthmatics per year.
Won't someone think of the poor people.
Has it occurred to you that maybe all those people shouldn't have moved to a place with crappy transportation infrastructure? Or that if driving their cars kills asthmatics, maybe the solution is to pollute less rather than destroying more (publicly owned, pollution-cleaning) ecosystem?
More to the point, in the long term, adding more suburban freeway capacity doesn't solve anything. In the studies I've looked at, commuters don't drive less after capacity is added, they just drive farther. They do this mainly by moving yet farther out from the cities where they work.
As far as I can tell, the people are doing a great job of thinking of themselves already. Which is sort of the problem, isn't it?
Did you even read the article?
The beach was a public beach. Putting boulders on a public beach does not exactly cry out for privacy protection. And according to the California Costal Commission, the agency that grants the permits, the golf course didn't have a permit.
The no-seawall stuff isn't just for the little froggies, although destroying a public resource (the ecosystem) for private gain is generally a no-no. Other reasons include
- California requires public access to the coast; some seawalls impede that, often intentionally
- seawalls on one property can increase erosion on nearby properties
- stopping erosion means that beaches aren't replenished, destroying them
So turn it around: Why would being dumb enough to build on an eroding piece of land give you a legitimate claim to build a seawall?