Thats fun. A friend of mine did that with a mainboard (w/486/DX66+3 PCI!) of ME! He is still a friend, though.:)
He stuck the chip in the socket and I reckoned he did it correctly, as that is his job. I reckoned incorrectly. EPROMS have a tiny little window where you can direct ultraviolet radiation inside to erase the content. A bright flash of radiation came from INSIDE the chip when I turned the system on. I knew instantly that this was not a good sign.
The chip was fried and I had no BIOS Backup. I wrote emails all over the world but could not find the manufacturer. The funny thing of the story is that if I had remembered the BIOS-ID string which is on the screen when the system boots, Award could have told me the manufacturer.
No BIOS, no bootup, no BIOS-ID, no Manufacturer, no image. Damn.:)
The idea is to place the public-key in DNS and the private key(s) on the mailserver(s). Not the dial-up users sign the mails, but the relay-servers. The dial-up users log in with their SMTP-AUTH password as they are used to.
So if a virus is hijacking a dial-up machine, it only must catch the SMTP-AUTH password, which boils down to the same situation as today.
SRS/SPF still have a large number of problems to solve. SPF alone is very good idea, but the special-case of mail-forwarding is not compatible with the current design. As SPF breaks forwarding, SRS is an ugly hack which tries to repair the damage done.
I am not convinced that SRS does not introduce ugly bugs, which enables spammers to circumvent SPF alltogether.
Specifically I think SRS can as easily be forged as mails can be forged today, as noone hinders you to fake a forward which hasnt taken place in the first time. If this forward looks ok, you are in.
I did not see any drop of traffic today. I checked the logs and saw a 50% reduction in spam coming from comcast and attbi. I also saw a 50% increase of spam coming from t-ipconnect and others.
In my opinion Comcast should use a silimar system like cbl.abuseat.org. However, they should ensure that spam-zombies can not automatically remove themselves from the list. Perhaps removal should require the dialup-password or something similar.
As for spam-blocking dynamic ips:
For all IPs in dialup-rbls, IPs without a reverse-ptr and IPs with more then one digit in the hostname I use greylisting with a delay of 300 seconds.
This has served me equally well with a maximum efficiency.
Thats exact the same arguments people had against RBLs (Realtime-Blackhole-Lists): "Wholesale blocking of a complete machine is a lazy, destructive answer to the problem. It may stop the flow of spam in the short term, but it also seriously harms legitimate users of the machine."
I use relays.ordb.org, opm.blitzed.org, block.dnsbl.sorbs.net, zombie.dnsbl.sorbs.net, socks.dnsbl.sorbs.net,misc.dnsbl.sorbs.net,smtp.dn sbl.sorbs.net,web.dnsbl.sorbs.net and sbl.spamhaus.org on a VERY large mailserver and had 2 complaints in 2 years! For comparison: Every day 200.000 mails are beeing relayed through that system.
So I think if the automation isn't totally braindamaged, you can use them (RBLs) without harming innocent bystanders. Blocking ports in a network is technically EXACTLY the same thing as blocking IPs on a mailserver. Given a list of targets, you block a resource. See?
Heck, they could (and probably do) even use the same RBLs for the port-blocking.
The concern here is not that they DO block, but on what data-source the block and how long.
...disable your catchall-address on your 4 domains, and only set up the addresses you need. You will see that the rate of spam you get will drop.
Furthermore the overall traffic you cause on the net will drop also, because the spam will be blocked directly at the mailserver with a 550. The mail will not be transmitted at all.
There are at least a dozen of other methods to block mail from entering a mailserver (given you really have admin-rights on the MX of your domains). There is no need to forward 3000 mails a day to some unlucky bureaucrat.
I want to remind you that once one elevator is ready a new one can be built for far less money then the first one. So we can have...say 2 of them for the price of the first one, because transportation is now cheaper as its our business now and not the rocket stinkers. Thank you, universe!
Its a symbol for mankind to take pride in their achievement. Its a symbol that the universe is out there for us to expand to, and we can! Its a symbol of progress! Its something that can trigger something an any person on this planet.
Mankind can create something that is so big that interplanetary distances seem not so big anymore!
And to add: Not the symbol of freedom caused your 9/11. I sympathize with your condition though. Would be a shock to me too.
You must calculate the angle and the length of the ribbon (the position and mass of the counterweight) to even out the centripetal force of the earths gravitation. I can not calculate these for myself but I am sure there are people out there who can.:)
This could be THE chance for bootstrapping an industry on extraterrestrial objects like the moon or the asteroids.
First isolate bacteria for extraction of other elements. If you have enough diversity take a small craft filled with bacteria and their life-support system which certainly isnt as big as one for humans. Then land the craft on the object and start mining and sorting out the elements.
Later send crafts to pick up that elements and produce something out of it. Perhaps first a larger scale mining operation, or a larger scale production facility.
Use greylisting. I recently implemented it on a large mailserver with modifications I found on the postfix-users mailing-list. Sorry but I do not remember who posted it. Here is how it works.
My current (modified) strategy is: Only greylist IPs which are
listed in a DNSBL(***) of your choice or
contain several digits in their resolving hostname which would indicate a dial-up host.
(***) i use l2.spews.dnsbl.sorbs.net and cbl.abuseat.org. I would never reject any mail with these dnsbls as the false-positives are too high, but for greylisting they work perfect.
This keeps the number of false-positives low and is really effective, as only suspicious hosts (dialup, dnsbled) are checked.
I am very satisfied by the results. The number of mails in the deferred queue dropped from ~15k to ~600, the system-load dropped from 2 to 0.5 despite the additional checking and database-lookups done. My system sends ~ 3-5 mails/second and rejects/deferrs 10-15 mails/second.
Greylisting implementations for your favourite MTA are allready available. You only have to use them.
lets see... if the station-wagon has a usable volumen of 1 m^3 ( 1000 litres ) and the harddrives have a size of (15*10*2.5) cm^3 and store 120GB, and the wagon does 80mph then the speed of the data is:
Lets wait until ITER is completed, and then talk again. The estimated power release of ITER is said to be higher then the power you need to keep the fusion running. Even JET got to 65% ratio 1997.
H3 is a better fuel for ANY fusion-reactor because with nearly the same procedure you would not have to have as much shielding as built into ITER. This is because the currently used fuel produces neutrons as byproducts (waste) and neutron radiation can only be shielded by large thick plates of metal. H3 produces only electrical charged particles as byproducts which can be magnetically deflected, thus enabling you to build smaller reactors e.g. for use in spaceships or sea-vessels. The problem is, H3 is too expensive.
See? There is even a marketing possibility here.:)
interesting. how would you define good/evil in a computational environment?
for me evil would be if a segment of code wants to consume resources for itself without offering a benefit to the user or all other code segments. could be interpreted as minor altruism or the opposite of greed. the code should be able to make selections of the type of code it wants to give a benefit.
this could produce interesting structures in the ongoing process.
ps. i think a good way to generate interesting settings is to run them through evolutionary algorithms. why try to find them themselfs if we can "evolve" them.:)
Slashdot Mirror
...its funny. :)
Thats fun. A friend of mine did that with a mainboard (w/486/DX66+3 PCI!) of ME! He is still a friend, though. :)
:)
He stuck the chip in the socket and I reckoned he did it correctly, as that is his job. I reckoned incorrectly. EPROMS have a tiny little window where you can direct ultraviolet radiation inside to erase the content. A bright flash of radiation came from INSIDE the chip when I turned the system on. I knew instantly that this was not a good sign.
The chip was fried and I had no BIOS Backup. I wrote emails all over the world but could not find the manufacturer. The funny thing of the story is that if I had remembered the BIOS-ID string which is on the screen when the system boots, Award could have told me the manufacturer.
No BIOS, no bootup, no BIOS-ID, no Manufacturer, no image. Damn.
Thanks. Nice idea. Will try that.
Nobody forces you via abuse of market power gained by a monopoly to use Linux.
I cant say that about Windows.
A 1:1 comparison of this distinct situations is definitely a fallacy.
I searched for "wheel mouse" and got a microsoft.com advertisment page for their hardware. On google I get Colas Nahaboo X mouse wheel scroll page. :)
I searched for "postfix policy" to get a description on the postfix policy framework and I got:
Search Error
MSN Search is temporarily unable to process your request.
Please try again in a few minutes.
EID: f:618926422 - 1041:1041:10004:1059
HC: 71d61b13
That has not to be the case.
The idea is to place the public-key in DNS and the private key(s) on the mailserver(s). Not the dial-up users sign the mails, but the relay-servers. The dial-up users log in with their SMTP-AUTH password as they are used to.
So if a virus is hijacking a dial-up machine, it only must catch the SMTP-AUTH password, which boils down to the same situation as today.
SRS/SPF still have a large number of problems to solve. SPF alone is very good idea, but the special-case of mail-forwarding is not compatible with the current design. As SPF breaks forwarding, SRS is an ugly hack which tries to repair the damage done.
I am not convinced that SRS does not introduce ugly bugs, which enables spammers to circumvent SPF alltogether.
Specifically I think SRS can as easily be forged as mails can be forged today, as noone hinders you to fake a forward which hasnt taken place in the first time. If this forward looks ok, you are in.
Now: Forging Spammer -> Recipient
SRS/SPF: Forging Spammer -> Recipient
^
|
Imaginary Origin
Why should this work at all? Please enlighten me.
I did not see any drop of traffic today. I checked the logs and saw a 50% reduction in spam coming from comcast and attbi. I also saw a 50% increase of spam coming from t-ipconnect and others.
For me this didn't work.
In my opinion Comcast should use a silimar system like cbl.abuseat.org. However, they should ensure that spam-zombies can not automatically remove themselves from the list. Perhaps removal should require the dialup-password or something similar.
As for spam-blocking dynamic ips:
For all IPs in dialup-rbls, IPs without a reverse-ptr and IPs with more then one digit in the hostname I use greylisting with a delay of 300 seconds.
This has served me equally well with a maximum efficiency.
Thats exact the same arguments people had against RBLs (Realtime-Blackhole-Lists): "Wholesale blocking of a complete machine is a lazy, destructive answer to the problem. It may stop the flow of spam in the short term, but it also seriously harms legitimate users of the machine."
n sbl.sorbs.net,web.dnsbl.sorbs.net and sbl.spamhaus.org on a VERY large mailserver and had 2 complaints in 2 years! For comparison: Every day 200.000 mails are beeing relayed through that system.
I use relays.ordb.org, opm.blitzed.org, block.dnsbl.sorbs.net, zombie.dnsbl.sorbs.net, socks.dnsbl.sorbs.net,misc.dnsbl.sorbs.net,smtp.d
So I think if the automation isn't totally braindamaged, you can use them (RBLs) without harming innocent bystanders. Blocking ports in a network is technically EXACTLY the same thing as blocking IPs on a mailserver. Given a list of targets, you block a resource. See?
Heck, they could (and probably do) even use the same RBLs for the port-blocking.
The concern here is not that they DO block, but on what data-source the block and how long.
...disable your catchall-address on your 4 domains, and only set up the addresses you need. You will see that the rate of spam you get will drop.
Furthermore the overall traffic you cause on the net will drop also, because the spam will be blocked directly at the mailserver with a 550. The mail will not be transmitted at all.
There are at least a dozen of other methods to block mail from entering a mailserver (given you really have admin-rights on the MX of your domains). There is no need to forward 3000 mails a day to some unlucky bureaucrat.
Please see:
Greylisting - the next step in the spam-control war (generic)
Anti-UCE Cheat-Sheet (Postfix)
Security-Sage Anti-Spam Guide (Postfix)
Nice post. Would be total crazy!
I want to remind you that once one elevator is ready a new one can be built for far less money then the first one. So we can have...say 2 of them for the price of the first one, because transportation is now cheaper as its our business now and not the rocket stinkers. Thank you, universe!
Its a symbol for mankind to take pride in their achievement.
Its a symbol that the universe is out there for us to expand to, and we can!
Its a symbol of progress! Its something that can trigger something an any person on this planet.
Mankind can create something that is so big that interplanetary distances seem not so big anymore!
And to add: Not the symbol of freedom caused your 9/11. I sympathize with your condition though. Would be a shock to me too.
It means what it does. To elevate things very high. So...
The proposed location is also very remote (somewhere in the pacific ocean) and can easy be made off limits for flight-lines.
I also can not see what a terrorist could gain from destroying a fanal of humanitys transition to become an interplanetary race.
Processes to make fibres of nanotubes have allready been developed:
2 0 4.asp
http://www.nature.com/nsu/040308/040308-10.html
http://www.technologyreview.com/articles/rnb_041
You must calculate the angle and the length of the ribbon (the position and mass of the counterweight) to even out the centripetal force of the earths gravitation. I can not calculate these for myself but I am sure there are people out there who can. :)
This could be THE chance for bootstrapping an industry on extraterrestrial objects like the moon or the asteroids.
:)
First isolate bacteria for extraction of other elements. If you have enough diversity take a small craft filled with bacteria and their life-support system which certainly isnt as big as one for humans. Then land the craft on the object and start mining and sorting out the elements.
Later send crafts to pick up that elements and produce something out of it. Perhaps first a larger scale mining operation, or a larger scale production facility.
Then construct all the other stuff. Dreaming...
Short answer: GPS
My current (modified) strategy is: Only greylist IPs which are
- listed in a DNSBL(***) of your choice or
- contain several digits in their resolving hostname which would indicate a dial-up host.
(***) i use l2.spews.dnsbl.sorbs.net and cbl.abuseat.org. I would never reject any mail with these dnsbls as the false-positives are too high, but for greylisting they work perfect.This keeps the number of false-positives low and is really effective, as only suspicious hosts (dialup, dnsbled) are checked.
I am very satisfied by the results. The number of mails in the deferred queue dropped from ~15k to ~600, the system-load dropped from 2 to 0.5 despite the additional checking and database-lookups done. My system sends ~ 3-5 mails/second and rejects/deferrs 10-15 mails/second.
Greylisting implementations for your favourite MTA are allready available. You only have to use them.
lets see... if the station-wagon has a usable volumen of 1 m^3 ( 1000 litres ) and the harddrives have a size of (15*10*2.5) cm^3 and store 120GB, and the wagon does 80mph then the speed of the data is:
87.3125 Petabits*m/s (1.2288142 * 10^16 byte-meters/s)
proof, see here
given your link to google I found out that you only told half of the story.
:)
look here for the rest.
you also proved the first paragraph of this page
thank you.
Hey. Lets render this. If it works out, this guy will have my kudos, which of course isn't worth anything. ;-)
Lets wait until ITER is completed, and then talk again. The estimated power release of ITER is said to be higher then the power you need to keep the fusion running. Even JET got to 65% ratio 1997.
:)
H3 is a better fuel for ANY fusion-reactor because with nearly the same procedure you would not have to have as much shielding as built into ITER. This is because the currently used fuel produces neutrons as byproducts (waste) and neutron radiation can only be shielded by large thick plates of metal. H3 produces only electrical charged particles as byproducts which can be magnetically deflected, thus enabling you to build smaller reactors e.g. for use in spaceships or sea-vessels. The problem is, H3 is too expensive.
See? There is even a marketing possibility here.
interesting. how would you define good/evil in a computational environment?
:)
for me evil would be if a segment of code wants to consume resources for itself without offering a benefit to the user or all other code segments. could be interpreted as minor altruism or the opposite of greed. the code should be able to make selections of the type of code it wants to give a benefit.
this could produce interesting structures in the ongoing process.
ps. i think a good way to generate interesting settings is to run them through evolutionary algorithms. why try to find them themselfs if we can "evolve" them.