I thought Paul Allen (cofounder of MS with Gates) was one of the principals in Transmeta (where Linus works). Not that strange to see them still working together on things (in fact, doesn't Allen still own a decent stake in MS?).
It's more like someone tells you that they found out that all the locks from some manufacturer use the same key--are you then liable if you did not get around to changing the locks before someone uses their key to open the door and steal all your stuff? Does time make a difference? What if you find out after the theft, at the moment of the theft, seconds prior to the theft, an hour, a week, a year? What if you there is only one lock manufacturer (although a door manufacturer includes its own locks with its products and there are instructions on the internet to make your own locks) and the last time they had to replace locks with a problem it turned out that they didn't actually lock? What then?
To get back to reality, there are plenty of reasons not to patch servers. Notice that the Slammer crashed parts of MS's network. Further, note that the patch you mention was cancelled out by a later patch so that people actually had to apply a third patch to eliminate the vulnerability. Again. Does it make a difference to you on which patch they were? If the patch came out yesterday, last week, last month...
MS patches are just as buggy as the original software (in fact, they might be even more buggy, since they don't go through the same review process as the original release does). On a production server can you afford to take the risk that MS's patch of the week won't cause data loss or introduce a new security vulnerability? Is it reasonable to expect small installations (with only one SQL Server, maybe as an SBS 2000 box) to be able to keep up with the massive amounts of information that come out and choose the correct patch schedule?
To get back to the Pinto comparison, what if you did look but you couldn't see because there was a tree? So you inch out a bit; then a bit more; then a car coming around the blind curve hits you because the driver happened to glance away at the critical moment? Both you and the other driver are badly burned in the explosion. You were both wearing your seat belts and would have been uninjured if not for the faulty gas tank. Who's responsible for the burns? It's not at all unreasonable to claim that the *burns* are at least partially Ford's fault. This is the most analagous to the MS situation. The maintainers followed a reasonable, albeit unagressive, upgrade path (the same one that many MS admins followed) and got burnt.
In American law (and presumably that of Korea), there are certain things that one cannot sign away. For example, river boats always use to make passengers sign a waiver (essentially an EULA) saying that they were not responsible for luggage lost if the boat sank. However, it was decided in a famous court case that this was an unreasonable requirement. Ever since that case, all river boat owners were responsible for recompensing passengers for lost belongings. Depending on Korean law, MS may or may not be responsible for damages caused by its products performance, regardless of what the EULA says.
The argument in your second paragraph is easier. Car manufacturers *are* required to recompense people for damages suffered when their car failed to operate properly. (Or McDonalds paying $8 million for someone who spilled coffee on herself.) There is always plenty of blame to spread around. The question is if there is enough blame due MS to convince a Korean court to make them liable.
Regardless of whether it is possible to make software that is 100% bug free, it is certainly possible to make a greater effort to remove bugs than MS makes: a longer beta period (and cut it out with these unrealistic dates for next release), more quality assurance efforts, hiring Kevin Mitnick to try to crack it, etc.
Even if MS loses this lawsuit, it won't end software releases. They'll buy liability insurance and do more testing (as mandated by their insurer). Compare this to the medical field: not only must all steps be taken to fix damage caused by negligent behavior, but compensatory damages are paid as well--thus the high price of malpractice insurance.
Open source software is both more and less vulnerable to this. More because it is transferred without charge and cannot simply increase its price to include the cost of insurance. Less, because there is no charge and thus no claim against earnings or property (and there may not be any property of the programmers in the country that is using the software); further, since the source is available, it is more reasonable to claim that due diligence involves checking for bugs (and fixing them--something MS software users cannot do). Note that distributing binaries weakens this somewhat over source only distributions.
If you really want to improve the trade balance, then increase foreign aid. This seems counter intuitive but is a simple result of looking at the flow of money. We (in the US) would like to do three things (and can choose two, currently 1 and 3):
1. Keep foreign aid low to save money. 2. Keep imports lower than exports, so that we produce more stuff and have more jobs. 3. Act as the Reserve currency for the world. This allows us to largely ignore currency fluctuations. It also means that when we print money that we can buy foreign goods with it.
To provide the Reserve currency, the US has to export dollars (for other countries to use in their transactions). To export dollars, we either have to buy stuff (i.e. import) or give them away (i.e. foreign aid). We choose to maintain a trade deficit. This results in our receiving goods in exchange for a product with a very high profit margin, money (even if we had to print paper currency, the profit on printing money is over 90%; in point of fact though, the relevant transactions are generally electronic; electronic creation of money is close to 100% profitable; the only danger is that eventually we might have to provide goods for the money).
Tariffs won't reduce the trade deficit (unless they lower world demand for dollars). Instead, tariffs reduce *exports*. Why? Because they reduce imports and take away foreign countries ability to buy our goods (since they don't provide the world's reserve currency, foreign countries need dollars to buy our goods; unless we send them some, they can't buy). Since exports are five to ten times the size of the trade deficit, we would stand to lose more than we could possibly gain.
Don't believe me? Perhaps you should take a little look at history. What bill's passage in 1929 led to Black Monday's crash in October? Yes, that's it, the Smoot Hawley tariff act. The domino effect (as other countries responded with their own tariffs) cracked open the world economic system. This was followed by a power struggle between the Fed's Board of Governors (Greenspan is the current Chairman) and the Fed. Reserve Bank of NY, which led to the US money supply being cut in half.
There is only one way to get out of a recession and that is to increase demand. To increase demand, you need to increase the money supply. To increase the money supply, you need the Fed to release more money. To get the Fed to release more money, you need to convince them that inflation is under control. Since tariffs exist to allow domestic products to be priced higher, they cause inflation. Thus, tariffs will make it harder to get the Fed to increase the money supply and lengthen the recession.
If you dist-upgrade to Sid or Sarge, it can break LindowsOS. CNR runs perfectly fine on Woody AFAIK (yes, there are a couple people running CNR on Debian boxes just to prove that they can). You can also mess things up by adding Sid or Sarge sources to your sources.list (LindowsOS defaults to Woody sources).
Note: you don't have to use CNR after the dist-upgrade to mess things up. You can also have problems using a program that accesses glibc (or another library that is radically different from LindowsOS to Sid/Sarge) that is not affected by the dist-upgrade (something third party like StarOffice or MoneyDance).
I think that you will find that you have many of the same issues when you dist-upgrade from Woody to Sid or Sarge. Maybe that's why a dist-upgrade is not usually recommended on a production box (in fact, you probably shouldn't be running a distribution called testing or unstable on a production box anyway).
In general, the focus in LindowsOS is to make the GUI simpler and safer to use. If you are using the command line, it is assumed that you know what you are doing well enough to recover from your mistakes. If not...LindowsOS reinstalls really quickly:)
Check out TMDA.net. Essentially what it does is require people who have never emailed you previously to respond to a challenge email. If you respond to the challenge, you get whitelisted and your email goes through henceforth. This is discussed more lower in the page (look for TMDA).
I don't think that this is *the* answer to spam, but it is an answer that improves the current situation.
TMDA supports this by allowing you to create temporary email addresses and list-specific addresses. For example, say my email address is me@whitelist.com. I sign up for an account at Amazon. Instead of putting in an email address of me@whitelist.com, I put in me-amazon.whitelist.com. Mail sent to this address is automatically passed through. If Amazon sells my email address, I will be able to tell since I will get spam to me-amazon. If so, I can remove this address from my pass through list and complain to Amazon.
Alternately, if I want to put out my email for a reply that will come within a week, TMDA can create an expirable email of the form me-A283653DE@whitelist.com (the A283653DE is an encrypted date/time string). I can then happily send that email out to anyone with the assurance that after a week, messages sent there will bounce.
I agree that this is not an absolute solution to the problem since not everyone will use whitelists. However, it can be a very good solution to the personal problem of trying to manage one's own receipt of spam.
Other than one graphic designer (Mac OSX), everyone at Lindows.com runs LindowsOS at work. This was announced by Kevin Carmony in response to a question on the forums.
It is a lot easier to use mirrors if you tell us what they are. As stands, to find a mirror, I would look it up on your site, which rather defeats the purpose.
If you want a Linux distribution based around WINE, you should get Xandros (Lindows.com's partner at the time of the marketing you mention). Suse is also moving in this direction.
Click-N-Run was the original idea behind Lindows.com (think of it as mp3.com for software). It wasn't the main marketing point initially, since MS Windows compatibility sounds sexier and that's where they were paying the big bucks.
Lindows.com put half a million dollars into improving WINE. At the end of that, WINE runs a subset of Office 2000 apps (Word, Excel, PowerPoint). However, these apps all have good Linux alternatives (e.g. OpenOffice and StarOffice).
Every Linux distribution has access to WINE and the related commercial apps from www.transgaming.com and www.codeweavers.com. LindowsOS is the only one that has Click-N-Run (Lycoris has a product with similar purpose that installs a limited set of software). It also helps them compete with MS Windows, by offering a simple solution available at lower cost.
Lindows.com will be introducing a $49 version of their Click-N-Run subscription that does not come with any commercial software (i.e. it includes OpenOffice but not StarOffice; Gimp but not Photogenics--commercial software will still be available at a member's discount--like a warehouse club).
Lindows.com is still in investment phase. They pay more for development and marketing than they get back in revenue (i.e. they lose money). Hopefully this will change as they start collecting commissions off commercial software that is added to the warehouse. However, they will have to get more commercial software first. It might also help that CNR subscriptions will begin to expire in about five to six months. If they get a good renewal rate, it will help supplement their new sign ups.
The LGPL exists to allow people to use *static* libraries in their code. Shared libraries (.so) are another story. Unsurprisingly, LindowsOS includes very few static (.a) libraries.
Under your system, vendors could not issue binary only drivers for their hardware without providing the source code, since they *must* link to the GPLed kernel code to be useful.
One can avoid learning how to learn if left with slower classmates. A good memory and a certain ability to match up useful information can obtain decent grades. If you are never pushed, you will never have to learn how to learn. Of course, pushing up grades won't necessarily help either, since the student is still with classmates without the innate ability, just more experience.
Her choice may have been correct (if I understand you correctly), keep him with other students for the social interaction but allow him to interact intellectually with an adult with a thorough understanding of the subject. Depending on how much of the tutoring involved independent study, this actually could help teach him how to learn.
I'm not sure that traditional schools really teach one how to learn. What they teach you is that there is always someone who will assign you the next challenge. That there is always someone to help if you get stuck. This isn't true in the real world. Sometimes, you are on the cutting edge and there is no knowledge of how to proceed. Sometimes, that knowledge is not available to you. A situation where one can learn from an expert is the exception in small businesses, not the rule.
Actually, the video game response discussed may be more useful in the real world. Look around; try things; if they don't work, try something else. This is more similar to the way graduate programs work in college. Sure, you have advisers, but to actually do the work involves you poking around at things.
Lindows.com returns all its source code modifications now. They never get credit for this. The biggest area was the WINE project, but they also hire open source developers to make improvements they require in other packages as well (e.g. Everaldo did a new set of icons that will appear in LindowsOS 4.0).
It is impossible to "get ahead at the expense of others" with GPLed software (at least within the law). At worst, a company could bundle and sell a bunch of GPLed binaries without contributing improvements. However, they would have to make the source code available to those buyers who want it; plus, if they do make changes, they have to GPL the changes to the source code as well. That's the great strength of the GPL in the OSS movement, once something is in the GPL, no one can take derived work out (except an agreement among *all* the contributors to that work).
The other thing that a lot of people miss is that Lindows.com is providing what the Linux community has lacked: marketing and sales savvy. Traditionally, the Gnu/Linux community has been very developer/techie oriented. This is good; it is why I like Linux. However, developers and techies will always be only a small portion of the population.
I would like to be able to run Linux 100% of the time, but I can't. Why not? Because some of the games^H^H^H^H^Happlications that I run require MS. Why? Because MS has over 95% of the desktop consumer market. Would many of that 95% be perfectly content running a linux distro? Probably. Why don't they? Because Gnu/Linux is lacking some critical features that desktop consumers want.
What features do deskop consumers want that Gnu/Linux does not have? The most prominent is niceness. MS applications are nice to the point of being cloying (e.g. Office assistant--Hi, I'm a staple; look at me; aren't I cute? Ask me a question, so I can tell you ten things you don't want to know; maybe the eleventh will be what you want). LindowsOs is working on these kinds of features.
It all begins with the OS installer. LindowsOS asks a few simple questions and a few minutes later it's installed. I spent over an hour just configuring a Red Hat install last week (yes, I'm picky). Plus, LindowsOS comes with a nice limited default install. Basically, the only apps that come with the base install are the web browser, email, and networking.
Want additional apps? That's what their flagship product, Click-N-Run does. One can use a web interface to find the application desired and click the Install button. CNR downloads and installs the application without further intervention from the user. No compile from source worries.
Is this appropriate for everyone? No. The existing market for Gnu/Linux distros is primarily aimed at people who want to be able to compile from source (which one can still do in LindowsOS). However, the consumer desktop market does not want to know as much about their computer as one needs to know to compile from source. They want it to work like a VCR. Press the power button and it turns on. Press the internet button and you're browsing the web. Press the email button and you're reading email. That's the direction LindowsOS is going.
The problem is that this kind of intuitive simplicity is not really supported by OS developers. Why? Two reasons: one, they don't need it--if you know enough to write the app, you know enough not to need training wheels to run it; two, it's better for developers if the users have to learn how the app works well enough to add their own bug fixes--that's how open source works--one writes, many improve. Expecting the 95% of users who are not professional or amateur programmers to learn how to program just to run apps is unrealistic. It's easier for them to just pay MS or Lindows.com to do their thinking for them.
This is just the same as someone hiring a doctor to diagnose their illness or a lawyer to write their will. They train for ten years. We pay for one hour (at rates high enough to justi
I must have posted the fact that a virus scanner is well nigh useless in Linux at least 10 times before they introduced it. It was still the most requested new app (beating out requests for a good html composer to compete with FrontPage or an easy to use GUI database front end like Access--my personal vote).
Also, if you read the page to which you link, you will notice that they spend as much time on talking about keeping from forwarding viruses (presumably bundled in with another email) as they do about protecting LindowsOS itself. I think that the phrase "to make sure your computer is ready for anything" pretty much sums up the Lindows.com attitude towards virus protection...
It's hard to educate people who don't want to be educated. It's easier to just sell them the product they want. Vexira works (it will happily scan your Fat32 drives for viruses as well). It's a real product, albeit unneeded for most. Also, the kind of people who don't understand why Linux is not as vulnerable as MS to viruses are the most likely to do something that will get past natural security precautions; in other words, launch an application they downloaded from a warez site as root. Maybe they do need an antivirus after all.
Yes, forging headers is fraud. You can sue based on damage to reputation and cost of processing the bounce messages. However, you will first need to determine who is sending the spam. To try to do that, you will have to do some research. If possible, obtain full headers for each type of email, so that you can see who is sending them. The other way you can determine identity is to follow the money. How is the spammer going to make a profit? Selling porn sites? Referrals? Just getting paid to send the emails by someone else?
If you could use help interpreting headers, send me an email at my Slashdot userid at yahoo dot com. Please include Otter from Slashdot in the subject if you send me email; otherwise, I might think it is spam...
The way the ionic air purifiers work, you should move it *away* from the computer to get maximum effect. The way that you are doing it, when the device pulls dust back towards it, some (most?) of the dust will get diverted by the air flow of the fans. In fact, I would suspect that by putting the device *next* to the computer, you are actually increasing the dust flow to the computer.
An ionic air purifier has two parts. One sends out charged ions which collide with particles of dust and impart a charge to them. The other is an area on the unit itself that has the opposite type of charge. I've never used one, so I don't know how effective they are. However, given the way it works, it seems logical to me that the best place for it would be away from the computer. The desired effect of the device is to pull dust into its area. Instead, it might be better to have it so that it pulls dust away from the computer. Besides, I would think that throwing electric charges (which is the basis of the ionic effect) around near a computer would be undesireable.
eBay => large, high revenue company that is afraid of being sued. If you send them an email telling them that this guy is stealing your work, I'm sure that they will cancel the auction. They have far more to lose than he does.
I would also agree that it is likely that the Free Software Foundation could direct you to someone who could help you go after him directly. You could send them an email at the same time as you send the one to eBay.
Go to tmda.net and add a whitelist to your email account. If an unauthorized sender tries to send you an email, you send back a challenge. If they respond to the challenge, you know that a real person sent you the email. Then you can accept the email add them to the whitelist. If you later decide that you don't want them on your whitelist, you can move them to your blacklist and email from them will be rejected.
Alternately, you can change SMTP. I offered a suggestion in another post entitled "How to do this for real."
P.S. IANAL, but if you fill out the order form with relevant but useless information, I don't think you can be prosecuted for fraud. Try something like this: First name: Hello Last name: I Address: was City: unable State: to Zip: respo (or 73560 if you need to be 1337 to make it numeric) Comment: nd to your email address, so I am submitting an order instead. Please stop sending me emails. I do not want your products. I will never buy your products. I think that your marketing tactics are immoral.
Since this proposal seems aimed at making it more difficult to become a spammer (you have to get certification to bulk mail) but protects "authorized" spammers, I would say that it is a spam promotion mechanism. The largest spammers will be able to send spam, because they can afford to buy the certification (which also protects them from ISP blacklists, etc.). Smaller spammers will no longer be allowed to compete. Further, it would also eliminate the ability of *legitimate* mailing lists to send newsletters, etc. without certification.
I think that this is way overcomplicating the situation. What is the number one reason why spammers can't be detected? They use relays and proxies to hide their identity. How to fix this? Only accept SMTP mail from servers authorized to send email for that domain. This would require a new DNS record (call it an SMTP record for now). If a server does not have authorization to send email for a domain (say yahoo.com), then when the receiving server looks up the SMTP record, it won't find it and will reject the email.
If the server is authorized, then the email will go through. If it turns out to be spam, then the sending server can either stop sending email for the spammer or be blacklisted. This will make relaying much harder. Also, it makes it useful for SMTP servers to authenticate users. Currently, this is meaningless in the fight against spam, since one could just use an open relay instead. However, if only authorized smtp servers were allowed to relay mail, then requiring authorization prevents spammers from sending mail under a false email address.
This would create a traceable system and allow spammers to be identified without forcing client software changes (might have to change SMTP configuration). Spammers would have to own an account or a domain name in order to send spam. Either requires payment and contact info. Faking the contact info would be fraud which would be prosecutable by tracing the payment.
It's true that Red Hat does not have the size to convince the insurance industry to support its products, but other major players in linux do; IBM and HP come to mind immediately.
Is voter apathy not voting? Even ignoring the potential increase in votes by dead people, this proposal would make it easier to vote and thus increase the percentage of people voting. I'm not sure that this is a good thing.
Many of the people who vote now do so without taking the time to understand the issues and the candidates' stands on the issues. Decreasing the barriers to voting will only increase the amount of stupid voting. I would rather have fewer voters who take more time to study the likely effects of their votes.
I encourage everyone to exhibit that kind of apathy. If you don't know what's going on, don't vote. I've done this selectively. If I am voting and have no real clue why one choice might be better than the other, I skip it and move on. Otoh, if you do want to vote, take the time to understand what's happening, look at the candidates and determine why they pick their positions.
Yes, but you are comparing apples to oranges. A hit is not a response. This is the equivalent of someone opening the envelope before tossing it (or maybe looking at the envelope before tossing it unopened).
I'm not sure what the click through rate is after people look at it, but I would be very surprised to find out that it is more than 10%. That gets you down to.2%. We still haven't gotten a response. Figure that maybe 1% of those who click through might buy or send feedback, and you have.002%.
This is what I mean by people getting fooled by the claims. You are an educated user (I'm basing this on your working in a marketing company), but you still did not look into the numbers enough to be able to truly compare them.
Internet advertising in general will be able to generate more stats for the price than traditional advertising. However, one must remember to compare equivalent situations. In your snail mail promotions, I suspect that your 1.5% to 5% response range involves actual sales. The equivalent internet number would be same visit purchases.
Perhaps I should get some numbers from Google or fark on same visit purchases to throw them out the next time that this comes along.
Slashdot Mirror
I thought Paul Allen (cofounder of MS with Gates) was one of the principals in Transmeta (where Linus works). Not that strange to see them still working together on things (in fact, doesn't Allen still own a decent stake in MS?).
It's more like someone tells you that they found out that all the locks from some manufacturer use the same key--are you then liable if you did not get around to changing the locks before someone uses their key to open the door and steal all your stuff? Does time make a difference? What if you find out after the theft, at the moment of the theft, seconds prior to the theft, an hour, a week, a year? What if you there is only one lock manufacturer (although a door manufacturer includes its own locks with its products and there are instructions on the internet to make your own locks) and the last time they had to replace locks with a problem it turned out that they didn't actually lock? What then?
To get back to reality, there are plenty of reasons not to patch servers. Notice that the Slammer crashed parts of MS's network. Further, note that the patch you mention was cancelled out by a later patch so that people actually had to apply a third patch to eliminate the vulnerability. Again. Does it make a difference to you on which patch they were? If the patch came out yesterday, last week, last month...
MS patches are just as buggy as the original software (in fact, they might be even more buggy, since they don't go through the same review process as the original release does). On a production server can you afford to take the risk that MS's patch of the week won't cause data loss or introduce a new security vulnerability? Is it reasonable to expect small installations (with only one SQL Server, maybe as an SBS 2000 box) to be able to keep up with the massive amounts of information that come out and choose the correct patch schedule?
To get back to the Pinto comparison, what if you did look but you couldn't see because there was a tree? So you inch out a bit; then a bit more; then a car coming around the blind curve hits you because the driver happened to glance away at the critical moment? Both you and the other driver are badly burned in the explosion. You were both wearing your seat belts and would have been uninjured if not for the faulty gas tank. Who's responsible for the burns? It's not at all unreasonable to claim that the *burns* are at least partially Ford's fault. This is the most analagous to the MS situation. The maintainers followed a reasonable, albeit unagressive, upgrade path (the same one that many MS admins followed) and got burnt.
In American law (and presumably that of Korea), there are certain things that one cannot sign away. For example, river boats always use to make passengers sign a waiver (essentially an EULA) saying that they were not responsible for luggage lost if the boat sank. However, it was decided in a famous court case that this was an unreasonable requirement. Ever since that case, all river boat owners were responsible for recompensing passengers for lost belongings. Depending on Korean law, MS may or may not be responsible for damages caused by its products performance, regardless of what the EULA says.
The argument in your second paragraph is easier. Car manufacturers *are* required to recompense people for damages suffered when their car failed to operate properly. (Or McDonalds paying $8 million for someone who spilled coffee on herself.) There is always plenty of blame to spread around. The question is if there is enough blame due MS to convince a Korean court to make them liable.
Regardless of whether it is possible to make software that is 100% bug free, it is certainly possible to make a greater effort to remove bugs than MS makes: a longer beta period (and cut it out with these unrealistic dates for next release), more quality assurance efforts, hiring Kevin Mitnick to try to crack it, etc.
Even if MS loses this lawsuit, it won't end software releases. They'll buy liability insurance and do more testing (as mandated by their insurer). Compare this to the medical field: not only must all steps be taken to fix damage caused by negligent behavior, but compensatory damages are paid as well--thus the high price of malpractice insurance.
Open source software is both more and less vulnerable to this. More because it is transferred without charge and cannot simply increase its price to include the cost of insurance. Less, because there is no charge and thus no claim against earnings or property (and there may not be any property of the programmers in the country that is using the software); further, since the source is available, it is more reasonable to claim that due diligence involves checking for bugs (and fixing them--something MS software users cannot do). Note that distributing binaries weakens this somewhat over source only distributions.
foobar/foobar worked for me
If you really want to improve the trade balance, then increase foreign aid. This seems counter intuitive but is a simple result of looking at the flow of money. We (in the US) would like to do three things (and can choose two, currently 1 and 3):
1. Keep foreign aid low to save money.
2. Keep imports lower than exports, so that we produce more stuff and have more jobs.
3. Act as the Reserve currency for the world. This allows us to largely ignore currency fluctuations. It also means that when we print money that we can buy foreign goods with it.
To provide the Reserve currency, the US has to export dollars (for other countries to use in their transactions). To export dollars, we either have to buy stuff (i.e. import) or give them away (i.e. foreign aid). We choose to maintain a trade deficit. This results in our receiving goods in exchange for a product with a very high profit margin, money (even if we had to print paper currency, the profit on printing money is over 90%; in point of fact though, the relevant transactions are generally electronic; electronic creation of money is close to 100% profitable; the only danger is that eventually we might have to provide goods for the money).
Tariffs won't reduce the trade deficit (unless they lower world demand for dollars). Instead, tariffs reduce *exports*. Why? Because they reduce imports and take away foreign countries ability to buy our goods (since they don't provide the world's reserve currency, foreign countries need dollars to buy our goods; unless we send them some, they can't buy). Since exports are five to ten times the size of the trade deficit, we would stand to lose more than we could possibly gain.
Don't believe me? Perhaps you should take a little look at history. What bill's passage in 1929 led to Black Monday's crash in October? Yes, that's it, the Smoot Hawley tariff act. The domino effect (as other countries responded with their own tariffs) cracked open the world economic system. This was followed by a power struggle between the Fed's Board of Governors (Greenspan is the current Chairman) and the Fed. Reserve Bank of NY, which led to the US money supply being cut in half.
There is only one way to get out of a recession and that is to increase demand. To increase demand, you need to increase the money supply. To increase the money supply, you need the Fed to release more money. To get the Fed to release more money, you need to convince them that inflation is under control. Since tariffs exist to allow domestic products to be priced higher, they cause inflation. Thus, tariffs will make it harder to get the Fed to increase the money supply and lengthen the recession.
If you dist-upgrade to Sid or Sarge, it can break LindowsOS. CNR runs perfectly fine on Woody AFAIK (yes, there are a couple people running CNR on Debian boxes just to prove that they can). You can also mess things up by adding Sid or Sarge sources to your sources.list (LindowsOS defaults to Woody sources).
:)
Note: you don't have to use CNR after the dist-upgrade to mess things up. You can also have problems using a program that accesses glibc (or another library that is radically different from LindowsOS to Sid/Sarge) that is not affected by the dist-upgrade (something third party like StarOffice or MoneyDance).
I think that you will find that you have many of the same issues when you dist-upgrade from Woody to Sid or Sarge. Maybe that's why a dist-upgrade is not usually recommended on a production box (in fact, you probably shouldn't be running a distribution called testing or unstable on a production box anyway).
In general, the focus in LindowsOS is to make the GUI simpler and safer to use. If you are using the command line, it is assumed that you know what you are doing well enough to recover from your mistakes. If not...LindowsOS reinstalls really quickly
Hope this helps answer your question.
Check out TMDA.net. Essentially what it does is require people who have never emailed you previously to respond to a challenge email. If you respond to the challenge, you get whitelisted and your email goes through henceforth. This is discussed more lower in the page (look for TMDA).
I don't think that this is *the* answer to spam, but it is an answer that improves the current situation.
TMDA supports this by allowing you to create temporary email addresses and list-specific addresses. For example, say my email address is me@whitelist.com. I sign up for an account at Amazon. Instead of putting in an email address of me@whitelist.com, I put in me-amazon.whitelist.com. Mail sent to this address is automatically passed through. If Amazon sells my email address, I will be able to tell since I will get spam to me-amazon. If so, I can remove this address from my pass through list and complain to Amazon.
Alternately, if I want to put out my email for a reply that will come within a week, TMDA can create an expirable email of the form me-A283653DE@whitelist.com (the A283653DE is an encrypted date/time string). I can then happily send that email out to anyone with the assurance that after a week, messages sent there will bounce.
I agree that this is not an absolute solution to the problem since not everyone will use whitelists. However, it can be a very good solution to the personal problem of trying to manage one's own receipt of spam.
Other than one graphic designer (Mac OSX), everyone at Lindows.com runs LindowsOS at work. This was announced by Kevin Carmony in response to a question on the forums.
It is a lot easier to use mirrors if you tell us what they are. As stands, to find a mirror, I would look it up on your site, which rather defeats the purpose.
If you want a Linux distribution based around WINE, you should get Xandros (Lindows.com's partner at the time of the marketing you mention). Suse is also moving in this direction.
Click-N-Run was the original idea behind Lindows.com (think of it as mp3.com for software). It wasn't the main marketing point initially, since MS Windows compatibility sounds sexier and that's where they were paying the big bucks.
Lindows.com put half a million dollars into improving WINE. At the end of that, WINE runs a subset of Office 2000 apps (Word, Excel, PowerPoint). However, these apps all have good Linux alternatives (e.g. OpenOffice and StarOffice).
Every Linux distribution has access to WINE and the related commercial apps from www.transgaming.com and www.codeweavers.com. LindowsOS is the only one that has Click-N-Run (Lycoris has a product with similar purpose that installs a limited set of software). It also helps them compete with MS Windows, by offering a simple solution available at lower cost.
Lindows.com will be introducing a $49 version of their Click-N-Run subscription that does not come with any commercial software (i.e. it includes OpenOffice but not StarOffice; Gimp but not Photogenics--commercial software will still be available at a member's discount--like a warehouse club).
Lindows.com is still in investment phase. They pay more for development and marketing than they get back in revenue (i.e. they lose money). Hopefully this will change as they start collecting commissions off commercial software that is added to the warehouse. However, they will have to get more commercial software first. It might also help that CNR subscriptions will begin to expire in about five to six months. If they get a good renewal rate, it will help supplement their new sign ups.
The LGPL exists to allow people to use *static* libraries in their code. Shared libraries (.so) are another story. Unsurprisingly, LindowsOS includes very few static (.a) libraries.
Under your system, vendors could not issue binary only drivers for their hardware without providing the source code, since they *must* link to the GPLed kernel code to be useful.
One can avoid learning how to learn if left with slower classmates. A good memory and a certain ability to match up useful information can obtain decent grades. If you are never pushed, you will never have to learn how to learn. Of course, pushing up grades won't necessarily help either, since the student is still with classmates without the innate ability, just more experience.
Her choice may have been correct (if I understand you correctly), keep him with other students for the social interaction but allow him to interact intellectually with an adult with a thorough understanding of the subject. Depending on how much of the tutoring involved independent study, this actually could help teach him how to learn.
I'm not sure that traditional schools really teach one how to learn. What they teach you is that there is always someone who will assign you the next challenge. That there is always someone to help if you get stuck. This isn't true in the real world. Sometimes, you are on the cutting edge and there is no knowledge of how to proceed. Sometimes, that knowledge is not available to you. A situation where one can learn from an expert is the exception in small businesses, not the rule.
Actually, the video game response discussed may be more useful in the real world. Look around; try things; if they don't work, try something else. This is more similar to the way graduate programs work in college. Sure, you have advisers, but to actually do the work involves you poking around at things.
At a penny a hit, your script nets the spammer an extra $20.
Lindows.com returns all its source code modifications now. They never get credit for this. The biggest area was the WINE project, but they also hire open source developers to make improvements they require in other packages as well (e.g. Everaldo did a new set of icons that will appear in LindowsOS 4.0).
It is impossible to "get ahead at the expense of others" with GPLed software (at least within the law). At worst, a company could bundle and sell a bunch of GPLed binaries without contributing improvements. However, they would have to make the source code available to those buyers who want it; plus, if they do make changes, they have to GPL the changes to the source code as well. That's the great strength of the GPL in the OSS movement, once something is in the GPL, no one can take derived work out (except an agreement among *all* the contributors to that work).
The other thing that a lot of people miss is that Lindows.com is providing what the Linux community has lacked: marketing and sales savvy. Traditionally, the Gnu/Linux community has been very developer/techie oriented. This is good; it is why I like Linux. However, developers and techies will always be only a small portion of the population.
I would like to be able to run Linux 100% of the time, but I can't. Why not? Because some of the games^H^H^H^H^Happlications that I run require MS. Why? Because MS has over 95% of the desktop consumer market. Would many of that 95% be perfectly content running a linux distro? Probably. Why don't they? Because Gnu/Linux is lacking some critical features that desktop consumers want.
What features do deskop consumers want that Gnu/Linux does not have? The most prominent is niceness. MS applications are nice to the point of being cloying (e.g. Office assistant--Hi, I'm a staple; look at me; aren't I cute? Ask me a question, so I can tell you ten things you don't want to know; maybe the eleventh will be what you want). LindowsOs is working on these kinds of features.
It all begins with the OS installer. LindowsOS asks a few simple questions and a few minutes later it's installed. I spent over an hour just configuring a Red Hat install last week (yes, I'm picky). Plus, LindowsOS comes with a nice limited default install. Basically, the only apps that come with the base install are the web browser, email, and networking.
Want additional apps? That's what their flagship product, Click-N-Run does. One can use a web interface to find the application desired and click the Install button. CNR downloads and installs the application without further intervention from the user. No compile from source worries.
Is this appropriate for everyone? No. The existing market for Gnu/Linux distros is primarily aimed at people who want to be able to compile from source (which one can still do in LindowsOS). However, the consumer desktop market does not want to know as much about their computer as one needs to know to compile from source. They want it to work like a VCR. Press the power button and it turns on. Press the internet button and you're browsing the web. Press the email button and you're reading email. That's the direction LindowsOS is going.
The problem is that this kind of intuitive simplicity is not really supported by OS developers. Why? Two reasons: one, they don't need it--if you know enough to write the app, you know enough not to need training wheels to run it; two, it's better for developers if the users have to learn how the app works well enough to add their own bug fixes--that's how open source works--one writes, many improve. Expecting the 95% of users who are not professional or amateur programmers to learn how to program just to run apps is unrealistic. It's easier for them to just pay MS or Lindows.com to do their thinking for them.
This is just the same as someone hiring a doctor to diagnose their illness or a lawyer to write their will. They train for ten years. We pay for one hour (at rates high enough to justi
I must have posted the fact that a virus scanner is well nigh useless in Linux at least 10 times before they introduced it. It was still the most requested new app (beating out requests for a good html composer to compete with FrontPage or an easy to use GUI database front end like Access--my personal vote).
Also, if you read the page to which you link, you will notice that they spend as much time on talking about keeping from forwarding viruses (presumably bundled in with another email) as they do about protecting LindowsOS itself. I think that the phrase "to make sure your computer is ready for anything" pretty much sums up the Lindows.com attitude towards virus protection...
It's hard to educate people who don't want to be educated. It's easier to just sell them the product they want. Vexira works (it will happily scan your Fat32 drives for viruses as well). It's a real product, albeit unneeded for most. Also, the kind of people who don't understand why Linux is not as vulnerable as MS to viruses are the most likely to do something that will get past natural security precautions; in other words, launch an application they downloaded from a warez site as root. Maybe they do need an antivirus after all.
Yes, forging headers is fraud. You can sue based on damage to reputation and cost of processing the bounce messages. However, you will first need to determine who is sending the spam. To try to do that, you will have to do some research. If possible, obtain full headers for each type of email, so that you can see who is sending them. The other way you can determine identity is to follow the money. How is the spammer going to make a profit? Selling porn sites? Referrals? Just getting paid to send the emails by someone else?
If you could use help interpreting headers, send me an email at my Slashdot userid at yahoo dot com. Please include Otter from Slashdot in the subject if you send me email; otherwise, I might think it is spam...
The way the ionic air purifiers work, you should move it *away* from the computer to get maximum effect. The way that you are doing it, when the device pulls dust back towards it, some (most?) of the dust will get diverted by the air flow of the fans. In fact, I would suspect that by putting the device *next* to the computer, you are actually increasing the dust flow to the computer.
An ionic air purifier has two parts. One sends out charged ions which collide with particles of dust and impart a charge to them. The other is an area on the unit itself that has the opposite type of charge. I've never used one, so I don't know how effective they are. However, given the way it works, it seems logical to me that the best place for it would be away from the computer. The desired effect of the device is to pull dust into its area. Instead, it might be better to have it so that it pulls dust away from the computer. Besides, I would think that throwing electric charges (which is the basis of the ionic effect) around near a computer would be undesireable.
eBay => large, high revenue company that is afraid of being sued. If you send them an email telling them that this guy is stealing your work, I'm sure that they will cancel the auction. They have far more to lose than he does.
I would also agree that it is likely that the Free Software Foundation could direct you to someone who could help you go after him directly. You could send them an email at the same time as you send the one to eBay.
Good luck.
Go to tmda.net and add a whitelist to your email account. If an unauthorized sender tries to send you an email, you send back a challenge. If they respond to the challenge, you know that a real person sent you the email. Then you can accept the email add them to the whitelist. If you later decide that you don't want them on your whitelist, you can move them to your blacklist and email from them will be rejected.
Alternately, you can change SMTP. I offered a suggestion in another post entitled "How to do this for real."
P.S. IANAL, but if you fill out the order form with relevant but useless information, I don't think you can be prosecuted for fraud. Try something like this:
First name: Hello
Last name: I
Address: was
City: unable
State: to
Zip: respo (or 73560 if you need to be 1337 to make it numeric)
Comment: nd to your email address, so I am submitting an order instead. Please stop sending me emails. I do not want your products. I will never buy your products. I think that your marketing tactics are immoral.
Should be legal enough.
Since this proposal seems aimed at making it more difficult to become a spammer (you have to get certification to bulk mail) but protects "authorized" spammers, I would say that it is a spam promotion mechanism. The largest spammers will be able to send spam, because they can afford to buy the certification (which also protects them from ISP blacklists, etc.). Smaller spammers will no longer be allowed to compete. Further, it would also eliminate the ability of *legitimate* mailing lists to send newsletters, etc. without certification.
I think that this is way overcomplicating the situation. What is the number one reason why spammers can't be detected? They use relays and proxies to hide their identity. How to fix this? Only accept SMTP mail from servers authorized to send email for that domain. This would require a new DNS record (call it an SMTP record for now). If a server does not have authorization to send email for a domain (say yahoo.com), then when the receiving server looks up the SMTP record, it won't find it and will reject the email.
If the server is authorized, then the email will go through. If it turns out to be spam, then the sending server can either stop sending email for the spammer or be blacklisted. This will make relaying much harder. Also, it makes it useful for SMTP servers to authenticate users. Currently, this is meaningless in the fight against spam, since one could just use an open relay instead. However, if only authorized smtp servers were allowed to relay mail, then requiring authorization prevents spammers from sending mail under a false email address.
This would create a traceable system and allow spammers to be identified without forcing client software changes (might have to change SMTP configuration). Spammers would have to own an account or a domain name in order to send spam. Either requires payment and contact info. Faking the contact info would be fraud which would be prosecutable by tracing the payment.
Leverage existing law with proper infrastructure.
It's true that Red Hat does not have the size to convince the insurance industry to support its products, but other major players in linux do; IBM and HP come to mind immediately.
Is voter apathy not voting? Even ignoring the potential increase in votes by dead people, this proposal would make it easier to vote and thus increase the percentage of people voting. I'm not sure that this is a good thing.
Many of the people who vote now do so without taking the time to understand the issues and the candidates' stands on the issues. Decreasing the barriers to voting will only increase the amount of stupid voting. I would rather have fewer voters who take more time to study the likely effects of their votes.
I encourage everyone to exhibit that kind of apathy. If you don't know what's going on, don't vote. I've done this selectively. If I am voting and have no real clue why one choice might be better than the other, I skip it and move on. Otoh, if you do want to vote, take the time to understand what's happening, look at the candidates and determine why they pick their positions.
Support democracy; vote with intelligence.
Yes, but you are comparing apples to oranges. A hit is not a response. This is the equivalent of someone opening the envelope before tossing it (or maybe looking at the envelope before tossing it unopened).
.2%. We still haven't gotten a response. Figure that maybe 1% of those who click through might buy or send feedback, and you have .002%.
I'm not sure what the click through rate is after people look at it, but I would be very surprised to find out that it is more than 10%. That gets you down to
This is what I mean by people getting fooled by the claims. You are an educated user (I'm basing this on your working in a marketing company), but you still did not look into the numbers enough to be able to truly compare them.
Internet advertising in general will be able to generate more stats for the price than traditional advertising. However, one must remember to compare equivalent situations. In your snail mail promotions, I suspect that your 1.5% to 5% response range involves actual sales. The equivalent internet number would be same visit purchases.
Perhaps I should get some numbers from Google or fark on same visit purchases to throw them out the next time that this comes along.