Perens prefers not to include KDE because they have a lgpl strategy to make money,
KDE has no LGPL strategy. KDE uses the Qt library by TrollTech, and TrollTech allows use of Qt either a) for GPL projects, or b) if you pay TrollTech.
UserLinux will use GNOME because GNOME is LGPL licensed; so if someone wants to write good business software and sell it, they won't need to pay any money to anyone... in other words, the barriers to entry into the business software marketplace for UserLinux will be lower.
Note that this is not a strategy for UserLinux to make money directly; UserLinux is not charging anyone anything for LGPL code or for KDE.
yet userlinux will charge money to include it.
Actually, no. Bruce Perens's consulting business, which charges money for consulting, will be doing a consulting job for a customer. The job the customer has requested is help with setting up UserLinux with KDE.
The customer was free to install KDE themselves, to hire Bruce Perens, to hire someone else, or to ignore KDE and use something else.
I believe that both should be offered from the get-go.
KDE is offered from the (apt-)get go:
apt-get install kde
And then you will have KDE. Not hard.
The key point here is that if you want to be a certified UserLinux support guy, you will need to learn GNOME; you won't need to learn KDE. If you want to, on your own, you can learn KDE, and advertise that you are a certified UserLinux guy who also supports KDE.
In other words, KDE is not officially part of UserLinux, but it will be available for anyone who wants it, and any support organization that is willing to support it may do so.
Which is why this whole furor is silly.
So, in summary: UserLinux is based on Debian, and Debian already supports KDE. Using Debian's "apt-get" command, anyone who wants KDE can simply install it. Any support organizations that are willing to support it, can do so. UserLinux (and Bruce Perens) will take no action to prevent KDE from working with UserLinux. You, the user, still have complete freedom of choice for which DE you want.
Another reason: the GPL gives the ability to permit use of a patent in free software, without losing the ability to charge for the use of the patent for proprietary software.
A great example is IBM and the Read-Copy-Update patents. IBM was willing to donate free use of the RCU patents to Linux, but they would not be willing to donate them to BSD-licensed projects.
This is one way GPL is actually better for a business that wants to make money.
steveha
How does FreeBSD compare to Linux 2.6?
on
FreeBSD 5.2 Review
·
· Score: 2, Insightful
I remember a few years ago, I read that FreeBSD was far superior to the Linux kernel for a heavily-loaded server. Supposedly you could run a server at about 100% CPU load, for days, without any problem if you used FreeBSD, while a Linux kernel would have problems.
Now that Linux 2.6 is released, has Linux caught up with FreeBSD, or is FreeBSD still better?
(And play nice, folks, please. I'm not trying to start a flame war here.)
We came out last summer and put out some code that the Linux community on one hand said, preposterous, that's [Berkeley software]. On the other hand, some people in the Linux community said, hold on, you may have some copyright issues there.... There are 2.5 million servers out there today that have this code in it. When are Linux customers going to clean that stuff up? So that's one issue, Linux is tainted, even by their own admission.
Amazing. "some people in the Linux community said... you may have some copyright issues there..." Um, who, exactly, said this? And he leaps from that to "Linux is tainted, even by their own admission."
"When are Linux customers going to clean that stuff up?" Well, given that this code had already been removed from the Linux 2.5 kernel before SCO showed it in obfuscated form, and given that even the 2.4 kernels have had it removed now too, I'd say it has already been cleaned up.
"There are 2.5 million servers out there today that have this code in it." This code only ever existed in Itanium kernels; are there even 2.5 million Linux Itanium servers in the world? Of the Linux Itanium servers, how many are still running an old kernel with this code in it? (Not many, I should think, since there are some security holes that have been fixed in newer kernels.)
It's like studying a fractal. The more you look at the details of what he's saying, the more wrong stuff you find.
Athlon64 chips and Opteron chips use the same core, so they will run the same software. Some versions of Opteron support SMP, but no Athlon64 chips support it. The various chips have various amounts of cache, various memory configurations they will work with, etc. For example, the socket 754 chip works with plain old DDR400 RAM, but only has one memory channel.
If you want the specifics, I suggest you use Google. Here's a good article to start with:
Have you never built a computer for someone, and put some of your known-good components into the machine, whilst upgrading at the same time?
Of course I've sometimes swapped some parts around while upgrading. But since we usually don't upgrade our computers that often, by the time we do want a new computer it's easiest just to swap the whole computer.
I routinely e-mail people I've never e-mailed before, but not with spam, but I don't want to pay 5 cents a pop for the privilege.
If it would fix the spam problem, I'd cheerfully pay 5 cents to email strangers. It's cheaper than sending a letter.
And note that my system provides for refunds; good manners would be that when you receive an email that isn't spam, you should refund the 5 cents.
you're assuming most people will apply a charge, which then means most people won't be able to receive e-mail from people they've never heard of before
It's interesting that you equate a small charge with "won't be able to receive". I guess if you need to put a stamp on a letter, people won't be able to receive letters. Oh wait, people do receive letters. And money you spend on stamps is gone, unlike the spam-stopping fee that can be refunded under the proposed system.
you could adapt your system to charging a variable number of cents based on the 'spamminess' of the mail (for example, using the SpamAssassin score)
Sure. I didn't really go into the wrinkle of variable fees, but of course it's an option.
That article was comparing SPARC 64-bit binaries to SPARC 32-bit binaries.
The Athlon64 (and all AMD 64-bit processors) have additional features that are only available in 64-bit mode. In particular, there are extra general-purpose registers available in 64-bit mode. Since x86 is starved for registers, the extra registers help speed up your programs.
This isn't just theory; the benchmarks show that the AMD 64-bit processors are faster in 64-bit mode.
And I was actually looking into buying some oranges today, but I'm not seeing the price benefit compared to applesauce.
Or, to put it less obliquely, that's a strange comparison. A PowerMac G5 is for someone who wants a Mac. An Athlon64 motherboard is for... well, not someone who wants a mac.
Hope this helps.
P.S. The Athlon64 actually offers great price/performance in plain old 32-bit mode. It gets even better in 64-bit mode, but there's no reason to wait for ready availability of 64-bit software. Just as there's no reason to hold off on buying a G5 for a fully 64-bit MacOS.
Then I noticed that I never swap CPUs out anyway. Motherboards are cheap enough, I swap an entire motherboard with its CPU. In fact, usually I swap out entire computers.
Since we use all our computers, I usually build a complete new computer, get it working, swap it for the older one, and keep the older one handy for a while as a hot spare in case something goes wrong with the new one. Then later I find a good home for the older computer.
(Now that I'm buying Lian Li aluminum cases, I'll probably start swapping motherboards into cheaper steel cases, and putting new motherboards into the Lian Li case.)
But anyway, I might get a socket 754 motherboard and chip. It will outperform any computer I currently own, and it should have adequate horsepower to play Half-Life 2 and Doom 3.
Who sets the payments? That's right, you do, for your server. You have the option of setting the fee to zero if you want. How do you call this "mandatory"?
all us geeks will end up having our own 'free' e-mail system.
Why would you do that? If you just use digital signatures on your emails, all your geek friends will accept your emails for free because you will be on their white lists. And vice versa.
if MS does bring in micropayments for mail
What does that have to do with the idea I wrote about?
if the user set the threshold at higher amounts, which many companies would likely do, as spam affects the bottom line
Never happen. "Hey Joe, I was going to send a query to BlinkeyWidgets.com, but they want us to pay them a buck just for the email!" "Forget it, let's order from AcmeWidgets.com instead."
if I was as evil as a spammer, all I'd have to do is maybe pay $5 or 10 for an online classified, promising a lucrative job for the right geek, set the threshold at $0.20, wait for the resumes to come in... Hire nobody, and $$$ profit. 300 responses... $50 profit off a $10 investment.
Now that is a problem I hadn't thought of. I suspect that in the long run, it would work itself out: perhaps custom would require job postings to set the threshold at zero, and any job posting that requires an email fee would be suspect on its face. Until and unless custom and experience sort that out, some evil people could pull that scam.
And I do hate that scam. The amounts are small enough (they would be stealing just a little bit from many people) that no one person would be strongly motivated to check up on them to see whether there ever was a position. And any solution involving government to check up on want ads is not a welcome one.
I don't believe that commercial speech (ie, adverts) needs to be unregulated.
Yeah, but I want a system that works even against hijacked computers being used to send illegal messages. And I don't want a system that involves a government agency filtering all email.
Suppose that each spammer in your example above had to endure a 3 sec latency before the SMTP session would begin. That would add a net 300,000 seconds to their delivery time (using your numbers of 100,000 emails).
An interesting idea. How does this help if the spammers use hijacked computers? They are already using hijacked computers for DOS attacks and to get around SMTP blacklists.
And will the delay hurt legit mailing lists (like the Linux kernel hackers mailing list) more than it will hurt spammers?
I mean, what agencies / governments / banks would I need to contact when I wanted to setup a new mail server?
No one... except at least one micropayments company.
Who do I contact in case of dispute? Who arbitrates the money transfer? Even with International currency conversion? In Real Time?
Simple: the answer to all of those questions is "the micropayments company".
Still, you have put your finger on the major problem with the idea. The idea does assume a micropayments system that is reasonably universally accepted, can handle a high volume of transactions, does not charge a large fee per transaction... Does such a micropayments company exist yet?
I only know of one micropayments company right now: bitpass.com. I don't know if they meet the requirements for this idea, or not.
Whether bitpass.com will work or not, I am certain that micropayments will common in the future. The cost of running a database server capable of handling micropayments traffic isn't going up, it's going down. And micropayments will be useful.
And the proposed solution requires you to trust no one. You either pocket five cents (or whatever you set as the fee), or else a friend of yours has digitally signed the message and you have verified the signature yourself (well, your mail server has).
Okay, for those who don't run their own mail servers, they need to trust their ISPs, I guess. Just like you currently have to trust your ISP to not monkey around with your email. Nothing new.
The bad thing about your proposal is that the sender has to in some way be "trusted" and traceable. Nobody can "just send messages"
You are mistaken. Anyone can send you an email, if they pay your threshold fee.
If your friends want to be on your white list, they will need to digitally sign their emails. That's easy to do. Your server would need to keep a copy of their public keys. With a public key and a digitally signed message, it's easy to figure out if the signed message matches the public key. Public key encryption is nifty.
Not least of which: what if somebody uses your account to send email and you get charged?
As a practical matter, I think most people would not put more than $20 at a time into their email pool. Then if you are careless with your password, you can lose up to $20. It's like losing your long-distance phone card: you lose the money.
The bigger worry for me is that Microsoft
Note that my solution does not depend on large companies doing things for us. We run our own servers and manage our own email. This system would work on any OS that can run a mail server.
Then there's a question of "who keeps the money?"
Well, I already said that: you do.
Writers of the software aren't going to want the users to keep it.
Tough for them. You can just use free mail servers if the proprietary ones will steal from you. It's hard to imagine this really being a problem.
Neither are ISPs
This is why competition is good. If one ISP wants to steal from you, go to another ISP. If you are locked in to one ISP and they want to steal from you, you could always sue them; presumably they are a regulated monopoly. Again, it's hard to imagine this really being a problem.
Problem: email is cheap, almost free, so a 0.00001% response rate on spam is still enough to make money.
Solution: make email cost something.
How?
Government? No no no no no. We want full control over our own email. Government should only be used to solve problems that only government can solve, and email doesn't rise to that level.
So, the solution:
A new protocol to replace SMTP. Someone sends you an email, and your server replies with the amount of the micropayment required for the email to go through. Then they can pay or decline. Most people would leve this set to a low amount (five cents sounds good to me), but famous people might set the bar higher to reduce the amount of email they get. The server has a "white list" of people you won't charge for email; this will use digital signatures, not an easily-forged header field.
Your email client has three toolbar buttons: refund the fee for this message and add the sender to the white list, refund the fee for this message, and delete message without refunding the fee.
We would have to run this in parallel with SMTP for a while, but it will be hugely popular. People using this will find no penis enlargement (excuse me, "pen1s en.la.rg.em.en.t") emails in their new inbox, even as their SMTP inbox gets worse and worse with spam. The word-of-mouth on this would be incredible: "I only check my spambox every other day or so, if you want to get in touch with me quickly you will need to use the new email format."
Quick numbers:
Let's assume some wild numbers (I have done no research, I just made these up). Suppose a typical spam run sends out 100,000 pieces of spam, and 30 people are dumb enough to bite (sounds high, but let's assume it) and each of those people sends $30 (hoping to "get bigger now"). That's $900, which is a clear profit if you are simply blasting emails over SMTP. But if the average person charges five cents to receive an email, it would cost 5,000 dollars to send out that spam run, for a net loss of $4,100. This is why spam would no longer work.
Note that you might receive ads in your inbox, but they would be ads where the sender is confident that the ad is worth five cents. If someone sent me a coupon good for $20 off something I actually want to buy, I'd even refund the five cents.
It dosen't take a computer expert to pull up a shell SU into root start up the Linux config tools and recompile the Linux kernel.
Most computer users would disagree with you.
When I used the phrase "Linux guru", I meant to include the people who know how to build their own kernels. This already leaves behind the vast majority of computer users.
It wasn't the best example, I guess. I'm running Debian Unstable, and I haven't really looked very hard into what it would take to get automatic hardware discovery and config.
There is a great discussion here and I might try some of the stuff discussed there.
I think my basic point stands, however. Maybe the specific case of plugging in a CompactFlash reader works automagically now. But what will happen when a user plugs an iPod in to a 1394 port? How about a USB audio module? How about a Palm PDA cradle? This "Project Utopia" looks like it will provide a clean framework to allow the system to gracefully handle just about any hardware.
Linux already works fine as a desktop; what most potential switchers need are a few good apps
Linux already works fine, as long as you are a Linux guru, or if you never need to change the hardware configuration at all.
Would you like to read CompactFlash cards? Okay, plug a CompactFlash reader in to a USB port. Let's say your kernel was set up correctly, and lucky you, you don't need to run modprobe or edit modules.conf. But you still need to (for KDE) manually create an icon for the desktop to let you mount the CompactFlash reader, or (for GNOME) edit/etc/fstab so that the reader will show up in the "Disks" submenu when you right-click the desktop. Then you need to mount the CompactFlash chip manually.
The stuff that RML is working on right now would make it work like this:
You plug in the CompactFlash reader to a USB port, and it appears on the desktop. If your system is set up for it, GNOME then asks you "Would you like to import these photographss into your photo album?"
This is the sort of ease-of-use that Apple brags about. And I think it's really cool.
Project Utopia is going to glue a whole bunch of stuff together. Meanwhile, some of the pieces look interesting.
Is udev ready for use by typical Linux users (as opposed to kernel hackers)? How about sysfs -- that is just part of 2.6 and is completely ready, right? How about D-BUS?
Meanwhile, on a flamefest^H^H^H^H^H^H^H^H^Hdiscussion about KDE and GNOME, I saw a claim made that "hardly any GNOME applications use Bonobo". Is that true? If it is true, is it changing? (Wasn't a Network Object Model one of the fundamental things about gNOMe?)
I browsed RML's blog, and some of the screenshots look really cool. I'm really looking forward to this stuff.
Good points re: licensing. Not sure they really affect me
In practice, they probably don't. But note that you benefit from Ogg Vorbis even if you don't use it: since Ogg Vorbis is a decent standard and free, the non-free standards can't tighten their grip too much. If licensing AAC becomes unreasonable, Ogg Vorbis will look much more attractive, so AAC licensing will probably always be reasonable.
here's listening tests
Sure, those are the ones I found. I read the results a little differently than you did. You read it as "AAC beats Vorbis", and I read it as "AAC and Vorbis about the same".
That's why I said I hadn't seen one that was a slam-dunk for AAC. At 128kbps AAC won, but it was close: 4.42 for AAC vs. 4.28 for Vorbis, on a 1 to 5 scale. That works out to a 2.8% advantage of AAC against Vorbis, and this on subjective listening tests involving human beings.
And anyway, if you are like me, you won't use either 64kbps or 128kbps. You'll set it for an extra-high-quality setting where you don't hear any problems. I use -q 6 on the Ogg encoder, which seems to work out to around 180kbps. And the Vorbis files I download from Magnatune.com are encoded aroung 192kbps!
All of which is a long way of saying that Ogg Vorbis meets the "Good Enough" standard for me, so I'm not interested in anything patent-encumbered. Since I'm already using Ogg Vorbis, I am not motivated to switch.
AAC has consistantly beat Vorbis in listening tests at all higher (64+) bitrates.
I haven't seen those tests. A quick Google for "AAC Vorbis listening tests" didn't turn up any that were slam-dunk victories of AAC over Vorbis. Could you please point me to some of the listening test results you have seen?
why OGG if you have AAC?
Because AAC is covered by patents, and Ogg isn't.
If the platform you use has a good AAC encoder, and the licensing requirements for AAC aren't a problem for you, then you might as well use AAC. I certainly won't tell you not to do so.
But keep in mind that a bunch of companies hold basic patents on AAC, and they can change the licensing deal anytime until the patents run out. In practice, you will probably never need to care about this, unless you want to distribute music commercially. But it means that if you use AAC, some companies are allowed to tell you what you can do with your AAC, at least some of the time.
With Ogg, I have my choice of several players, all free and all legal. I have a free, legal encoder.
There are now several good players that support Ogg, and in future most or all players will support Ogg. Ogg is a safe bet for the future, with excellent quality at reasonable bit rates. Since I'm already using Ogg, I'm unlikely to change for AAC.
If you are already using AAC, you probably aren't in a hurry to change for Ogg, either. That's fine with me.
What's your deal, man? "flac is whack"? Could you be, perhaps, a little more specific than that?
FLAC is an open, free standard. Shorten is owned by the guys who wrote it. Does "anybody" support Shorten? Talk to the Shorten guys.
But it's probably already too late. Several portable players already support FLAC. Magnatune.com offers music in FLAC format. The train already has left the station, and Shorten was not on board. Why would a company pay to license Shorten when they can use FLAC for free, and FLAC is better?
IBM pushes Linux, yet All of the Lotus remains on windows. Sad.
I think what is happening with Lotus stuff is that it's simply being allowed to age out. IBM is pushing free software solutions, and not trying very hard to drag the older proprietary solutions into the free software world.
The Lotus office applications never went anywhere against Microsoft Office. Why try to enter that fight again? The free applications are already good enough that many people don't need anything else, and they are getting better. By virtue of being free software, they have a chance of grabbing market share from Microsoft Office.
IBM should have some guys working on the free alternatives to all the Lotus apps, and just let the Lotus apps die.
Slashdot Mirror
Perens prefers not to include KDE because they have a lgpl strategy to make money,
KDE has no LGPL strategy. KDE uses the Qt library by TrollTech, and TrollTech allows use of Qt either a) for GPL projects, or b) if you pay TrollTech.
UserLinux will use GNOME because GNOME is LGPL licensed; so if someone wants to write good business software and sell it, they won't need to pay any money to anyone... in other words, the barriers to entry into the business software marketplace for UserLinux will be lower.
Note that this is not a strategy for UserLinux to make money directly; UserLinux is not charging anyone anything for LGPL code or for KDE.
yet userlinux will charge money to include it.
Actually, no. Bruce Perens's consulting business, which charges money for consulting, will be doing a consulting job for a customer. The job the customer has requested is help with setting up UserLinux with KDE.
The customer was free to install KDE themselves, to hire Bruce Perens, to hire someone else, or to ignore KDE and use something else.
Hope this helps.
steveha
I believe that both should be offered from the get-go.
KDE is offered from the (apt-)get go:
apt-get install kde
And then you will have KDE. Not hard.
The key point here is that if you want to be a certified UserLinux support guy, you will need to learn GNOME; you won't need to learn KDE. If you want to, on your own, you can learn KDE, and advertise that you are a certified UserLinux guy who also supports KDE.
In other words, KDE is not officially part of UserLinux, but it will be available for anyone who wants it, and any support organization that is willing to support it may do so.
Which is why this whole furor is silly.
So, in summary: UserLinux is based on Debian, and Debian already supports KDE. Using Debian's "apt-get" command, anyone who wants KDE can simply install it. Any support organizations that are willing to support it, can do so. UserLinux (and Bruce Perens) will take no action to prevent KDE from working with UserLinux. You, the user, still have complete freedom of choice for which DE you want.
steveha
Why doesn't everyone use the BSD license?
Another reason: the GPL gives the ability to permit use of a patent in free software, without losing the ability to charge for the use of the patent for proprietary software.
A great example is IBM and the Read-Copy-Update patents. IBM was willing to donate free use of the RCU patents to Linux, but they would not be willing to donate them to BSD-licensed projects.
This is one way GPL is actually better for a business that wants to make money.
steveha
I remember a few years ago, I read that FreeBSD was far superior to the Linux kernel for a heavily-loaded server. Supposedly you could run a server at about 100% CPU load, for days, without any problem if you used FreeBSD, while a Linux kernel would have problems.
Now that Linux 2.6 is released, has Linux caught up with FreeBSD, or is FreeBSD still better?
(And play nice, folks, please. I'm not trying to start a flame war here.)
steveha
I think it's just that Linus uses shorthand abbreviations and slang.
"Linux doesn't have any SCO IP" would be quoted as "Linux doesn't have any [intellectual property belonging to SCO]".
"I couldn't automate a grep through my mail store" would be quoted as "I couldn't automate a [search through my saved email messages]".
steveha
Darl McBride speaks:
We came out last summer and put out some code that the Linux community on one hand said, preposterous, that's [Berkeley software]. On the other hand, some people in the Linux community said, hold on, you may have some copyright issues there.... There are 2.5 million servers out there today that have this code in it. When are Linux customers going to clean that stuff up? So that's one issue, Linux is tainted, even by their own admission.
Amazing. "some people in the Linux community said... you may have some copyright issues there..." Um, who, exactly, said this? And he leaps from that to "Linux is tainted, even by their own admission."
"When are Linux customers going to clean that stuff up?" Well, given that this code had already been removed from the Linux 2.5 kernel before SCO showed it in obfuscated form, and given that even the 2.4 kernels have had it removed now too, I'd say it has already been cleaned up.
"There are 2.5 million servers out there today that have this code in it." This code only ever existed in Itanium kernels; are there even 2.5 million Linux Itanium servers in the world? Of the Linux Itanium servers, how many are still running an old kernel with this code in it? (Not many, I should think, since there are some security holes that have been fixed in newer kernels.)
It's like studying a fractal. The more you look at the details of what he's saying, the more wrong stuff you find.
steveha
Athlon64 chips and Opteron chips use the same core, so they will run the same software. Some versions of Opteron support SMP, but no Athlon64 chips support it. The various chips have various amounts of cache, various memory configurations they will work with, etc. For example, the socket 754 chip works with plain old DDR400 RAM, but only has one memory channel.
If you want the specifics, I suggest you use Google. Here's a good article to start with:
http://www.theinquirer.net/?article=10954
steveha
Have you never built a computer for someone, and put some of your known-good components into the machine, whilst upgrading at the same time?
Of course I've sometimes swapped some parts around while upgrading. But since we usually don't upgrade our computers that often, by the time we do want a new computer it's easiest just to swap the whole computer.
steveha
I routinely e-mail people I've never e-mailed before, but not with spam, but I don't want to pay 5 cents a pop for the privilege.
If it would fix the spam problem, I'd cheerfully pay 5 cents to email strangers. It's cheaper than sending a letter.
And note that my system provides for refunds; good manners would be that when you receive an email that isn't spam, you should refund the 5 cents.
you're assuming most people will apply a charge, which then means most people won't be able to receive e-mail from people they've never heard of before
It's interesting that you equate a small charge with "won't be able to receive". I guess if you need to put a stamp on a letter, people won't be able to receive letters. Oh wait, people do receive letters. And money you spend on stamps is gone, unlike the spam-stopping fee that can be refunded under the proposed system.
you could adapt your system to charging a variable number of cents based on the 'spamminess' of the mail (for example, using the SpamAssassin score)
Sure. I didn't really go into the wrinkle of variable fees, but of course it's an option.
steveha
That article was comparing SPARC 64-bit binaries to SPARC 32-bit binaries.
The Athlon64 (and all AMD 64-bit processors) have additional features that are only available in 64-bit mode. In particular, there are extra general-purpose registers available in 64-bit mode. Since x86 is starved for registers, the extra registers help speed up your programs.
This isn't just theory; the benchmarks show that the AMD 64-bit processors are faster in 64-bit mode.
steveha
And I was actually looking into buying some oranges today, but I'm not seeing the price benefit compared to applesauce.
Or, to put it less obliquely, that's a strange comparison. A PowerMac G5 is for someone who wants a Mac. An Athlon64 motherboard is for... well, not someone who wants a mac.
Hope this helps.
P.S. The Athlon64 actually offers great price/performance in plain old 32-bit mode. It gets even better in 64-bit mode, but there's no reason to wait for ready availability of 64-bit software. Just as there's no reason to hold off on buying a G5 for a fully 64-bit MacOS.
steveha
I used to worry about sockets and the future.
Then I noticed that I never swap CPUs out anyway. Motherboards are cheap enough, I swap an entire motherboard with its CPU. In fact, usually I swap out entire computers.
Since we use all our computers, I usually build a complete new computer, get it working, swap it for the older one, and keep the older one handy for a while as a hot spare in case something goes wrong with the new one. Then later I find a good home for the older computer.
(Now that I'm buying Lian Li aluminum cases, I'll probably start swapping motherboards into cheaper steel cases, and putting new motherboards into the Lian Li case.)
But anyway, I might get a socket 754 motherboard and chip. It will outperform any computer I currently own, and it should have adequate horsepower to play Half-Life 2 and Doom 3.
steveha
If payment becomes mandatory,
Okay, go back and read it again. I'll wait.
Who sets the payments? That's right, you do, for your server. You have the option of setting the fee to zero if you want. How do you call this "mandatory"?
all us geeks will end up having our own 'free' e-mail system.
Why would you do that? If you just use digital signatures on your emails, all your geek friends will accept your emails for free because you will be on their white lists. And vice versa.
if MS does bring in micropayments for mail
What does that have to do with the idea I wrote about?
steveha
if the user set the threshold at higher amounts, which many companies would likely do, as spam affects the bottom line
Never happen. "Hey Joe, I was going to send a query to BlinkeyWidgets.com, but they want us to pay them a buck just for the email!" "Forget it, let's order from AcmeWidgets.com instead."
if I was as evil as a spammer, all I'd have to do is maybe pay $5 or 10 for an online classified, promising a lucrative job for the right geek, set the threshold at $0.20, wait for the resumes to come in... Hire nobody, and $$$ profit. 300 responses... $50 profit off a $10 investment.
Now that is a problem I hadn't thought of. I suspect that in the long run, it would work itself out: perhaps custom would require job postings to set the threshold at zero, and any job posting that requires an email fee would be suspect on its face. Until and unless custom and experience sort that out, some evil people could pull that scam.
And I do hate that scam. The amounts are small enough (they would be stealing just a little bit from many people) that no one person would be strongly motivated to check up on them to see whether there ever was a position. And any solution involving government to check up on want ads is not a welcome one.
I don't believe that commercial speech (ie, adverts) needs to be unregulated.
Yeah, but I want a system that works even against hijacked computers being used to send illegal messages. And I don't want a system that involves a government agency filtering all email.
steveha
Suppose that each spammer in your example above had to endure a 3 sec latency before the SMTP session would begin. That would add a net 300,000 seconds to their delivery time (using your numbers of 100,000 emails).
An interesting idea. How does this help if the spammers use hijacked computers? They are already using hijacked computers for DOS attacks and to get around SMTP blacklists.
And will the delay hurt legit mailing lists (like the Linux kernel hackers mailing list) more than it will hurt spammers?
I mean, what agencies / governments / banks would I need to contact when I wanted to setup a new mail server?
No one... except at least one micropayments company.
Who do I contact in case of dispute? Who arbitrates the money transfer? Even with International currency conversion? In Real Time?
Simple: the answer to all of those questions is "the micropayments company".
Still, you have put your finger on the major problem with the idea. The idea does assume a micropayments system that is reasonably universally accepted, can handle a high volume of transactions, does not charge a large fee per transaction... Does such a micropayments company exist yet?
I only know of one micropayments company right now: bitpass.com. I don't know if they meet the requirements for this idea, or not.
Whether bitpass.com will work or not, I am certain that micropayments will common in the future. The cost of running a database server capable of handling micropayments traffic isn't going up, it's going down. And micropayments will be useful.
steveha
The problem comes down to this: Who do you trust?
And the proposed solution requires you to trust no one. You either pocket five cents (or whatever you set as the fee), or else a friend of yours has digitally signed the message and you have verified the signature yourself (well, your mail server has).
Okay, for those who don't run their own mail servers, they need to trust their ISPs, I guess. Just like you currently have to trust your ISP to not monkey around with your email. Nothing new.
The bad thing about your proposal is that the sender has to in some way be "trusted" and traceable. Nobody can "just send messages"
You are mistaken. Anyone can send you an email, if they pay your threshold fee.
If your friends want to be on your white list, they will need to digitally sign their emails. That's easy to do. Your server would need to keep a copy of their public keys. With a public key and a digitally signed message, it's easy to figure out if the signed message matches the public key. Public key encryption is nifty.
Not least of which: what if somebody uses your account to send email and you get charged?
As a practical matter, I think most people would not put more than $20 at a time into their email pool. Then if you are careless with your password, you can lose up to $20. It's like losing your long-distance phone card: you lose the money.
The bigger worry for me is that Microsoft
Note that my solution does not depend on large companies doing things for us. We run our own servers and manage our own email. This system would work on any OS that can run a mail server.
Then there's a question of "who keeps the money?"
Well, I already said that: you do.
Writers of the software aren't going to want the users to keep it.
Tough for them. You can just use free mail servers if the proprietary ones will steal from you. It's hard to imagine this really being a problem.
Neither are ISPs
This is why competition is good. If one ISP wants to steal from you, go to another ISP. If you are locked in to one ISP and they want to steal from you, you could always sue them; presumably they are a regulated monopoly. Again, it's hard to imagine this really being a problem.
steveha
Problem: email is cheap, almost free, so a 0.00001% response rate on spam is still enough to make money.
Solution: make email cost something.
How?
Government? No no no no no. We want full control over our own email. Government should only be used to solve problems that only government can solve, and email doesn't rise to that level.
So, the solution:
A new protocol to replace SMTP. Someone sends you an email, and your server replies with the amount of the micropayment required for the email to go through. Then they can pay or decline. Most people would leve this set to a low amount (five cents sounds good to me), but famous people might set the bar higher to reduce the amount of email they get. The server has a "white list" of people you won't charge for email; this will use digital signatures, not an easily-forged header field.
Your email client has three toolbar buttons: refund the fee for this message and add the sender to the white list, refund the fee for this message, and delete message without refunding the fee.
We would have to run this in parallel with SMTP for a while, but it will be hugely popular. People using this will find no penis enlargement (excuse me, "pen1s en.la.rg.em.en.t") emails in their new inbox, even as their SMTP inbox gets worse and worse with spam. The word-of-mouth on this would be incredible: "I only check my spambox every other day or so, if you want to get in touch with me quickly you will need to use the new email format."
Quick numbers:
Let's assume some wild numbers (I have done no research, I just made these up). Suppose a typical spam run sends out 100,000 pieces of spam, and 30 people are dumb enough to bite (sounds high, but let's assume it) and each of those people sends $30 (hoping to "get bigger now"). That's $900, which is a clear profit if you are simply blasting emails over SMTP. But if the average person charges five cents to receive an email, it would cost 5,000 dollars to send out that spam run, for a net loss of $4,100. This is why spam would no longer work.
Note that you might receive ads in your inbox, but they would be ads where the sender is confident that the ad is worth five cents. If someone sent me a coupon good for $20 off something I actually want to buy, I'd even refund the five cents.
steveha
It dosen't take a computer expert to pull up a shell SU into root start up the Linux config tools and recompile the Linux kernel.
Most computer users would disagree with you.
When I used the phrase "Linux guru", I meant to include the people who know how to build their own kernels. This already leaves behind the vast majority of computer users.
steveha
It wasn't the best example, I guess. I'm running Debian Unstable, and I haven't really looked very hard into what it would take to get automatic hardware discovery and config.
There is a great discussion here and I might try some of the stuff discussed there.
I think my basic point stands, however. Maybe the specific case of plugging in a CompactFlash reader works automagically now. But what will happen when a user plugs an iPod in to a 1394 port? How about a USB audio module? How about a Palm PDA cradle? This "Project Utopia" looks like it will provide a clean framework to allow the system to gracefully handle just about any hardware.
steveha
Linux already works fine as a desktop; what most potential switchers need are a few good apps
/etc/fstab so that the reader will show up in the "Disks" submenu when you right-click the desktop. Then you need to mount the CompactFlash chip manually.
Linux already works fine, as long as you are a Linux guru, or if you never need to change the hardware configuration at all.
Would you like to read CompactFlash cards? Okay, plug a CompactFlash reader in to a USB port. Let's say your kernel was set up correctly, and lucky you, you don't need to run modprobe or edit modules.conf. But you still need to (for KDE) manually create an icon for the desktop to let you mount the CompactFlash reader, or (for GNOME) edit
The stuff that RML is working on right now would make it work like this:
You plug in the CompactFlash reader to a USB port, and it appears on the desktop. If your system is set up for it, GNOME then asks you "Would you like to import these photographss into your photo album?"
This is the sort of ease-of-use that Apple brags about. And I think it's really cool.
steveha
Project Utopia is going to glue a whole bunch of stuff together. Meanwhile, some of the pieces look interesting.
Is udev ready for use by typical Linux users (as opposed to kernel hackers)? How about sysfs -- that is just part of 2.6 and is completely ready, right? How about D-BUS?
Meanwhile, on a flamefest^H^H^H^H^H^H^H^H^Hdiscussion about KDE and GNOME, I saw a claim made that "hardly any GNOME applications use Bonobo". Is that true? If it is true, is it changing? (Wasn't a Network Object Model one of the fundamental things about gNOMe?)
I browsed RML's blog, and some of the screenshots look really cool. I'm really looking forward to this stuff.
steveha
Good points re: licensing. Not sure they really affect me
In practice, they probably don't. But note that you benefit from Ogg Vorbis even if you don't use it: since Ogg Vorbis is a decent standard and free, the non-free standards can't tighten their grip too much. If licensing AAC becomes unreasonable, Ogg Vorbis will look much more attractive, so AAC licensing will probably always be reasonable.
here's listening tests
Sure, those are the ones I found. I read the results a little differently than you did. You read it as "AAC beats Vorbis", and I read it as "AAC and Vorbis about the same".
That's why I said I hadn't seen one that was a slam-dunk for AAC. At 128kbps AAC won, but it was close: 4.42 for AAC vs. 4.28 for Vorbis, on a 1 to 5 scale. That works out to a 2.8% advantage of AAC against Vorbis, and this on subjective listening tests involving human beings.
And anyway, if you are like me, you won't use either 64kbps or 128kbps. You'll set it for an extra-high-quality setting where you don't hear any problems. I use -q 6 on the Ogg encoder, which seems to work out to around 180kbps. And the Vorbis files I download from Magnatune.com are encoded aroung 192kbps!
All of which is a long way of saying that Ogg Vorbis meets the "Good Enough" standard for me, so I'm not interested in anything patent-encumbered. Since I'm already using Ogg Vorbis, I am not motivated to switch.
steveha
AAC has consistantly beat Vorbis in listening tests at all higher (64+) bitrates.
I haven't seen those tests. A quick Google for "AAC Vorbis listening tests" didn't turn up any that were slam-dunk victories of AAC over Vorbis. Could you please point me to some of the listening test results you have seen?
why OGG if you have AAC?
Because AAC is covered by patents, and Ogg isn't.
If the platform you use has a good AAC encoder, and the licensing requirements for AAC aren't a problem for you, then you might as well use AAC. I certainly won't tell you not to do so.
But keep in mind that a bunch of companies hold basic patents on AAC, and they can change the licensing deal anytime until the patents run out. In practice, you will probably never need to care about this, unless you want to distribute music commercially. But it means that if you use AAC, some companies are allowed to tell you what you can do with your AAC, at least some of the time.
With Ogg, I have my choice of several players, all free and all legal. I have a free, legal encoder.
There are now several good players that support Ogg, and in future most or all players will support Ogg. Ogg is a safe bet for the future, with excellent quality at reasonable bit rates. Since I'm already using Ogg, I'm unlikely to change for AAC.
If you are already using AAC, you probably aren't in a hurry to change for Ogg, either. That's fine with me.
steveha
What's your deal, man? "flac is whack"? Could you be, perhaps, a little more specific than that?
FLAC is an open, free standard. Shorten is owned by the guys who wrote it. Does "anybody" support Shorten? Talk to the Shorten guys.
But it's probably already too late. Several portable players already support FLAC. Magnatune.com offers music in FLAC format. The train already has left the station, and Shorten was not on board. Why would a company pay to license Shorten when they can use FLAC for free, and FLAC is better?
steveha
IBM pushes Linux, yet All of the Lotus remains on windows. Sad.
I think what is happening with Lotus stuff is that it's simply being allowed to age out. IBM is pushing free software solutions, and not trying very hard to drag the older proprietary solutions into the free software world.
The Lotus office applications never went anywhere against Microsoft Office. Why try to enter that fight again? The free applications are already good enough that many people don't need anything else, and they are getting better. By virtue of being free software, they have a chance of grabbing market share from Microsoft Office.
IBM should have some guys working on the free alternatives to all the Lotus apps, and just let the Lotus apps die.
steveha