[snip]
What would have been better would be to register foogiston.org yourself,
but if you got beat to the punch, you should still be able to tell
someone not to use your trademark in their name.
No, this would NOT have been better.
If I have a trademark on Foogiston,
I should not have to pay Network Solutions to defend it.
IPs don't have the same functionality that domain names do.
Instead of IP addresses, we could use secure hashes of public
keys (like a P.G.P. finger print) and IPs which are signed by the public key.
Since it's cryptographically hard to forge, no one would have to trust anybody.
These would be hard for a human to remember,
(20-30 random characters) but they have the other advantages of DNS, and a few extra;
They are guaranteed to be unique,
A single computer (IP address) can support multiple names
- and -
Hi-jacking a domain is nearly impossible,
Anyone who wanted to could become a listing service (competition)
ICANN drags it's feet over creating TLDs, and then complains when someone else does the work for them?
If this were a software project, we wouldn't listen to anyone who didn't contribute code.
Show me the patches ICANN.
What TLDs have you created?
What has been done to promote competition?
What have you done for the public?
Instead of wasting time tearing down the others,
make them all moot.
If ICANN got off it's ass and approved a few hundred new TLDs,
then no one would care a whit about new.net, alternic, or any of the others.
What influence can the European Union have on an american company like microsoft?
Consider that they can do any or all of the following;
Seize all property, cash and other holdings, Microsoft has in Europe.
Legalize the copying and distribution of Microsoft software.
Deport, detain, or even kill Microsoft employees in Europe.
Demand the extradition of Microsoft executives for trail in Europe.
Sure, none of these things are likely, but they do point out that
you shouldn't ever treat an organization as large as a government lightly.
Governments have used their right of eminent domain to seize real property.
Most wouldn't even think twice about seizing the intellectual property of a
convicted criminal.
With over 5 million computers connected to the internet, even if 99.99% were clueful, then that's still over 500 idiots.
Welcome to the internet.
The idiots who complain of probes on 192.168.1.10 are annoying (and yes, I have gotten such a complaint) but the ones I hate are those who think attempting to connect to a service on their machine is an attack.
"Yes, I did attempt to connect to bar.foo's DNS server while trying to resolve bletch.bar.foo... Yes, I did do it intentionally... No, I will not pay your company for damages, or fix your software for free... No, I've never met an FBI agent before... "
I wonder if these people call the phone company to complain when someone dials their phone number by mistake.
Having a computer answer multiple choice questions with no wrong answers is hardly a meaningful "test." When it can get over 30% on the S.A.T., I'll be impressed.
Re:Gilmore doesn't get it
on
ORBS Forks
·
· Score: 1
Spare me
Getting spam through an open relay doesn't consume any more of your resources, than getting it direct.
If the operator of an open relay wants to spend the money to carry the traffic, that should be his decision, not ORBS.
Further, the costs are minuscule. If one assumes the incremental cost of a gigabyte of data at about $4, then you can relay over 25,000 messages for under a buck. Even if you recieve 100 spams a day, your part of the delivery cost is less than a dime a month. It costs the average user more to throw away the junk snail mail he receives than to download spam.
Along with the "fan in another room" idea is the "power supply in another room" idea. This would also allow the box to be considerably smaller.
You can also use a networked drive, and probably other components can be pushed into another room as well.
... with future software (including SSE-2 optimizations) it will outclass the competition...
Future software is better known as vaporware.
A year from now, we should be seeing 2 gigahertz stuff at the same price.
The only important benchmarks are the ones we can run today.
The transistor uses far less than a milliamp.
If the pentium III used transistors that drew
an entire "measly" milliamp, it would suck down over 100,000 watts.
Yes, we pay for receiving spam, but face facts:
Bandwidth is cheap.
If you pay $20 a month for a 56K dial up, and use it only 10 minutes a day,
then it's still less than a 1/10 of a cent to download most spam. And the true cost of delievery is a lot less, and falling every day. It is the time wasted reading enough of the spam to know it should be deleted that really costs.
Charging for UBE is an interesting idea, but until there's a practical suggestion of how to do so, it's just a fantasy.
Assuming the end point of the connection is a human, there's a real, practical limit to the amount of bandwidth needed. Most people can not watch more than one video stream at a time. A hi-res video stream is less than 100Mbps, so a gigabit should be enough for a 10 person household.
First we should distinguish between consensual and nonconsensual communications. Obviously, if you ask someone to connect to you on a port, then it's not a "scan" no matter what the port number is. Likewise, I think we can rule attempts to connect to standard public ports, like ftp, http, smtp, nntp, dns, whois, etc. as "white."
Then there are the "gray" ports, like the non-standard Gnutella ports. Attempting to connect to 6347 after failing to connect to 6346 might be an attempt to ask for permission to speak to a Gnutella client, though it's unlikely.
That still leaves a lot of ports, and attempting to connect to any of them is suspect.
If it is only once though, then we're still in the "gray" area. However, attempting to connect to hundreds of ports sequentially, even as slow as once a minute, puts us firmly in the "black" area.
Ok, now let me go back to your questions:
How many TCP ports need to be probed in how much time to be defined as port scanning?
More than 1 and less than 200. The total time for the scan needs to be less than the average time for a system upgrade.
Does a program have to used?
No. If you can whistle into the phone to do a port scan, then it's still a port scan.
If I send 50 http GET requests to a computer within one second, is that port scanning?
No, because it involves a generally public method of accessing the computer. If the requests were made on 50 different connections at the same time, then it
might be construed as a denial of service attack, but if it's a single connection, it's probably reasonable. Note that most browsers can actually make more than 50 requests in a second if both machines are connected via a fast pipe. This happens on web pages with lots of little graphics.
What about 50 TCP requests to a computer to 50 different ports in one second?
This is gray, but it's a pretty dark shade of gray. Note that in general speed is less of an issue than spread. 50 is a large spread, but if it were say, a DNS, followed by an ftp, then 16 at the Hotline section, followed by 16 in the Gnutella section, and finally 16 of the usual unusual ftp ports, then I'd say it's not.
Note that most illegal port scans brute force all the ports sequentially as fast as they can. They expect to be stopped, and they don't even attempt to hide there actions. The faster they look, the more they can see. But a thief who smashes a store window, grabs what he can and walks away, is just as guilty as the thief that does a smash and grab, then runs. He's just more likely to get caught.
What's so unamerican about a company having the freedom to make and sell products as they see fit?
Corporate entities are not freed from moral or legal responsibility, just because they are corporate entities.
Businesses can't kill people for money anymore than private citizens can.
The goal of business is to make money, and in the US, corporations have an obligation to do so.
But that doesn't mean they can break the law just to make a buck. Microsoft has already been convicted of abusing monopoly power.
In other words, Microsoft has already proven more than once that they can not be trusted with monopoly power, and their recent "innovations" are little more than a ruthless attempt to extend that abusive power.
It may be that/.ers have a knee jerk reaction to everything Microsoft does, but that's a learned reaction, based on long exposure to numerous abuses.
Ok, I'll bite:
"the users pay you and for the bandwidth to share the songs they already bought"
Paraphased:
"The users pay Napster. They also pay their ISP's for the bandwith used to share music with other users. And they are sharing songs they've already paid for."
the real problem is the lack of really good hooks for commonly used e-mail applications to use PGP
Hooks are no problem, just change your SMTP and POP3 to localhost, and make a PGP encrypted
email forwarder/reader. You can even scan the TO: line and see if it's on the list of people who can recieve encrypted email, and scan incoming email for X-ENCRYPT-ENABLED: fields
and add them to your lists automatically.
Ok, you've convinced me - Unicode can't handle the large number of asian "letters." I never liked Unicode anyway - let's just go back to 8 bit codes, admit that letters only work for European languages, and force everyone else to use graphics.
"shared source" is selling people the right to look at, but not copy, source code. It isn't a software developement philosophy, it isn't even sharing, that's just sleezy marketing. If I have to pay to look at something I can't touch it, then it wasn't shared with me.
The panel may decide SS is a better business model than GPL, but I doubt that will surprise anybody who understand what the two are.
GPL is designed for people who use software, not for people who make it. It's not going to make anybody rich, though it might make us all wealthy.
Dial up is on the rise. Maybe broadband is rising faster, but the fact remains there are more dialup users out there than ever. J.D. claims it's become a standard.
But it's not the speed that the market is standardizing on, it's the price. Dial up costs around $20 a month. It's been that price for a long time. Until DSL or Cable can offer something that's price competitive with that, dial up will continue to be the major player.
What I would invest in is DSL coupled with wireless modems and a neighbor who splits the bandwidth/cost with you.
A working group of the country-code top level domains (ccTLDs),
voted unanimously to pull out of ICANN's
Domain Name Supporting Organization (DNSO)
They've been talking for months and aren't happy.
The ccTLD supply 1/3 of ICANNs funding.
ICANN has two other supporting organizations.
Each supporting organization gets three seats on ICANN's 19-member board.
(No, the article does not mention how many members there are in the working group,
or if by pulling out they remove a full 1/3 of ICANN's funding or only a proportionate share.)
Corrections:
It would be more correct to say "They've been talking for years", not months.
There are 18 seats on ICANN, not 19.
In theory, 9 are elected by the three supporting agencies, and 9 elected by the internet "at large" but in reality, 6 are elected "support" seats (none of which are the ccTLD seats) and 4 elected "at large" seats. The rest are held by legacy appointees.
Some history on this:
ICANN was formed in 1998, essentially by the US, who appointed 9 people to the board at that time.
'Round about September 1999, 3 of the seats were supposed to have been filled by elections of the ccTLDs, but they didn't elect them. ICANN claims that they "decided to defer" the election. The ccTLD's claim they were not allowed to participate in this "decision." IOW, (if you believe them) they were defrauded of their three promised seats in 1999. It's hard to know who to believe in this, but I note that the seats are still aren't held by ccTLD's elected members.
I think it's extremely unlikely that the amendments to the
ICANN bylaws would have passed had those seats been filled by the ccTLDs.
Among other changes, the amendments have allowed 5 of the legacy appointees to remain without contest. Without the changes, they would have had to win an election to remain seated, and I for one, believe they wouldn't win.
If tax dollars are funding a project, then the results of that development should be available to everyone and not just people who use one particular license. This is the rationale behind the laws which prevent the government from enjoying a copyright on the data it produces.
Government code should be public domain, not placed under a restrictive license like the GPL.
<DEVILS ADVOCATE>
If US tax dollars fund something, why should the Ausies get it for free?
Besides, governments have funded a lot of research that ended up being classified, patened, copywritten, or released with far more restrictive licences than GPL. Allowing the government to GPL what it pays for is a much better alternative to forcing them to licence in a particular way, even if that way is public-domain. </DEVILS ADVOCATE>
If you don't like the news, make some of your own.
on
Legitimacy Of ICANN?
·
· Score: 2
So you hate ICANN, propose something else.
If you can get a reasonable number of people to agree with your proposal, then propose it to the DoC. Even if they don't agree, it's likely that the proposal will start moving things in the right direction.
For example,
I propose that ICANN and any entity they delegate authority to, be open to public inspection, including a detailed accounting of all money taken in or spent.
Slashdot Mirror
If I have a trademark on Foogiston, I should not have to pay Network Solutions to defend it.
It depends on what the alternatives are. New.net looks a lot better than Network Solutions.
These would be hard for a human to remember, (20-30 random characters) but they have the other advantages of DNS, and a few extra;
They are guaranteed to be unique,
A single computer (IP address) can support multiple names
- and -
Hi-jacking a domain is nearly impossible,
Anyone who wanted to could become a listing service (competition)
They can be cached essentially forever
If this were a software project, we wouldn't listen to anyone who didn't contribute code.
Show me the patches ICANN.
What TLDs have you created?
What has been done to promote competition?
What have you done for the public?
Instead of wasting time tearing down the others, make them all moot.
If ICANN got off it's ass and approved a few hundred new TLDs,
then no one would care a whit about new.net, alternic, or any of the others.
Seize all property, cash and other holdings, Microsoft has in Europe.
Legalize the copying and distribution of Microsoft software.
Deport, detain, or even kill Microsoft employees in Europe.
Demand the extradition of Microsoft executives for trail in Europe.
Sure, none of these things are likely, but they do point out that you shouldn't ever treat an organization as large as a government lightly.
Governments have used their right of eminent domain to seize real property.
Most wouldn't even think twice about seizing the intellectual property of a convicted criminal.
Welcome to the internet.
The idiots who complain of probes on 192.168.1.10 are annoying (and yes, I have gotten such a complaint) but the ones I hate are those who think attempting to connect to a service on their machine is an attack.
I wonder if these people call the phone company to complain when someone dials their phone number by mistake.When it can get over 30% on the S.A.T., I'll be impressed.
Getting spam through an open relay doesn't consume any more of your resources, than getting it direct. If the operator of an open relay wants to spend the money to carry the traffic, that should be his decision, not ORBS.
Further, the costs are minuscule. If one assumes the incremental cost of a gigabyte of data at about $4, then you can relay over 25,000 messages for under a buck. Even if you recieve 100 spams a day, your part of the delivery cost is less than a dime a month. It costs the average user more to throw away the junk snail mail he receives than to download spam.
This would also allow the box to be considerably smaller.
You can also use a networked drive, and probably other components can be pushed into another room as well.
A year from now, we should be seeing 2 gigahertz stuff at the same price.
The only important benchmarks are the ones we can run today.
If you pay $20 a month for a 56K dial up, and use it only 10 minutes a day, then it's still less than a 1/10 of a cent to download most spam. And the true cost of delievery is a lot less, and falling every day. It is the time wasted reading enough of the spam to know it should be deleted that really costs.
Charging for UBE is an interesting idea, but until there's a practical suggestion of how to do so, it's just a fantasy.
Then there are the "gray" ports, like the non-standard Gnutella ports. Attempting to connect to 6347 after failing to connect to 6346 might be an attempt to ask for permission to speak to a Gnutella client, though it's unlikely.
That still leaves a lot of ports, and attempting to connect to any of them is suspect. If it is only once though, then we're still in the "gray" area. However, attempting to connect to hundreds of ports sequentially, even as slow as once a minute, puts us firmly in the "black" area.
Ok, now let me go back to your questions:
Does a program have to used? If I send 50 http GET requests to a computer within one second, is that port scanning? What about 50 TCP requests to a computer to 50 different ports in one second?How many TCP ports need to be probed in how much time to be defined as port scanning?
Corporate entities are not freed from moral or legal responsibility, just because they are corporate entities.
Businesses can't kill people for money anymore than private citizens can.
The goal of business is to make money, and in the US, corporations have an obligation to do so. But that doesn't mean they can break the law just to make a buck. Microsoft has already been convicted of abusing monopoly power. In other words, Microsoft has already proven more than once that they can not be trusted with monopoly power, and their recent "innovations" are little more than a ruthless attempt to extend that abusive power.
It may be that /.ers have a knee jerk reaction to everything Microsoft does, but that's a learned reaction, based on long exposure to numerous abuses.
Paraphased:
"The users pay Napster. They also pay their ISP's for the bandwith used to share music with other users. And they are sharing songs they've already paid for."
Hooks are no problem, just change your SMTP and POP3 to localhost, and make a PGP encrypted email forwarder/reader. You can even scan the TO: line and see if it's on the list of people who can recieve encrypted email, and scan incoming email for X-ENCRYPT-ENABLED: fields and add them to your lists automatically.
The panel may decide SS is a better business model than GPL, but I doubt that will surprise anybody who understand what the two are. GPL is designed for people who use software, not for people who make it. It's not going to make anybody rich, though it might make us all wealthy.
But it's not the speed that the market is standardizing on, it's the price. Dial up costs around $20 a month. It's been that price for a long time. Until DSL or Cable can offer something that's price competitive with that, dial up will continue to be the major player.
What I would invest in is DSL coupled with wireless modems and a neighbor who splits the bandwidth/cost with you.
Corrections:
It would be more correct to say "They've been talking for years", not months.
There are 18 seats on ICANN, not 19. In theory, 9 are elected by the three supporting agencies, and 9 elected by the internet "at large" but in reality, 6 are elected "support" seats (none of which are the ccTLD seats) and 4 elected "at large" seats. The rest are held by legacy appointees.
Some history on this:
ICANN was formed in 1998, essentially by the US, who appointed 9 people to the board at that time.
'Round about September 1999, 3 of the seats were supposed to have been filled by elections of the ccTLDs, but they didn't elect them. ICANN claims that they "decided to defer" the election. The ccTLD's claim they were not allowed to participate in this "decision." IOW, (if you believe them) they were defrauded of their three promised seats in 1999. It's hard to know who to believe in this, but I note that the seats are still aren't held by ccTLD's elected members. I think it's extremely unlikely that the amendments to the ICANN bylaws would have passed had those seats been filled by the ccTLDs.
Among other changes, the amendments have allowed 5 of the legacy appointees to remain without contest. Without the changes, they would have had to win an election to remain seated, and I for one, believe they wouldn't win.
If US tax dollars fund something, why should the Ausies get it for free?
Besides, governments have funded a lot of research that ended up being classified, patened, copywritten, or released with far more restrictive licences than GPL. Allowing the government to GPL what it pays for is a much better alternative to forcing them to licence in a particular way, even if that way is public-domain.
</DEVILS ADVOCATE>
If you can get a reasonable number of people to agree with your proposal, then propose it to the DoC. Even if they don't agree, it's likely that the proposal will start moving things in the right direction.
For example,