# Its parsing is too complex
Complex or not, it's working just fine with quite an array of software.
# No sane firewall is going to let TXT records through
A firewall would need to examine the contents of packets to differentiate a TXT record from a, say, A record or cname. Comparable wizardry is already being performed by mail servers world wide, on a vast scale:
[smegma@cartman smegma]$ host -t txt 84.137.116.38.sbl.spamhaus.org
84.137.116.38.sbl.spamhaus.org text "http://www.spamhaus.org/SBL/sbl.lasso?query=SBL17 101"
# No sane firewall is going to let TCP DNS packets through
SPF does not rely on anything that every other app using DNS doesn't. Also see above.
# The parsing can loop forever
In a horribly written parser perhaps. The same could be said about IRC clients, Web Browsers and just about any application out there.
# It will increase DNS scaning as spamers hunt for broken SPF records
DNS is quite efficient. Unlike RBLs, SPF will work just fine with traditional SOA settings, so cache hits will be plentiful.
# Its too complex to be implimented inside the MTA where it needs to be done
Just where do you think it's already being implemented?
# It can't be properly parsed in sendmail
It is already being used with all popular MTAs, including sendmail, postfix, qmail, exim, courier and ms exchange.
# ISO 8839 8859 59-15 utf-8 issues for domain names may kill some dns servers
Huh?
Parsing complexity might become a bit of a concern with the advent of XML, but as of now, it's dead-simple.
3, Interesting? And I feel like I'm feeding a troll here!
Use your own domain for outgoing mail AND either do not publish an SPF record for it, or publish one pointing to your Internet Serive Provider's SMTP server.
If the DNS provider for your domain doesn't enable you to publish an SPF (needs a 'txt' record), you'll need to ask them to, or change providers. But something tells me the laggard will quickly figure it out, since SPF is nearing critical mass.
True, to some extent, but even power plants burning fossil fuels are orders of magnitude more efficient than small internal combustion engines. You also get the added benefit of mitigating city smog.
Re:I tried
on
The FragBook
·
· Score: 5, Informative
Disabling vertical sync can help, since it unclamps FPS from your monitor refresh.
A few hints on the fps/refresh subject:
For Q3A/RtCW engine based games, there's a console variable called 'r_displayRefresh'. This allows you to have the game set its own display refresh. I personally use r_displayRefresh 120 @ 800x600x32. Make sure you don't exceed your monitor's specifications.
Disabling vertical sync can greatly improve game performance for those stuck with low refresh rates. For example, one can easily do steady 125+ FPS while the monitor runs as 60Hz.
Use a mouse with a decent sampling rate and/or check its settings. Choppy performance (during turns especially) can often be attributed to low mouse rates and not frame rates.
Agreed. But another key advantage of big, public BLs vs private ones is to the unlikeliest of parties; the victims whose resources have been abused by spamming parasites.
If you'd find yourself blocked by thousands of admins' private lists, imagine getting the fiasco cleared up. Next to impossible. On the other hand, public blocklists usually have a simple, straight-forward removal procedure that immediately (blocklists typically use a very short refresh interval) takes effect for all its users.
If you don't agree with a BL's listing criteria or policies, don't use it.
There's a variety of DNSBLs out there. Some attempt to list spam sources (IPs from which spam is injected) with surgical precision whilst others go for the 'spam support' services, typically listing increasing swaths of space as the responsible party refuses to act (SPEWS for one).
In many cases the surgical approach simply won't do. Playing whack-a-mole with a fake ISP/spam support service isn't everybody's game.
Possibility 3) Intel wants to enable customers to better realize the differences between their own models; take a retailer's flashy ad for a "2.4GHz P4 Celeron" next to a Northwood at twice the price, yet showing the same clock rate. This could obviously help their (Intel's) customers.
Considering the present offerings from AMD - and more specifically their recent 64 bit models - I don't think Intel would want a head-to-head scale at this time. This would definitely be helpful to all consumers, but probably not what Intel wants right now.
Things will get very interesting in the next 2-3 years.
Re:Unfortunately much spam originates from the US.
on
UK Spam Law Goes Live
·
· Score: 2, Insightful
A lot of spam does indeed 'come from' Korea, China, Brazil etc, but a vast majority of it is still being sent by US spammers, using open relays and proxies in the aforementioned locations. Same with bullet-proof hosting of spamvertized webpages.
Sweden, for some very strage reason, has three providers offering symmetrical, 10 Mbps or bigger connections for less than 65 USD a month. I'm not a swede, and don't have information in English, but you should be able to decipher what matters here. The yellow box below the headline has the speeds and prices. One Swedish krona is roughly 13.5 US cents, prices are per month and anslutningsavgift is the one-time hookup fee.
As you can see in the Aftonbladet article, Telia has just entered the fray. They were literally forced to do so, by competition from Bredbandsbolaget and Bostream.
Bredbandsbolaget, apparently not content with losing their edge (their connections are generally considered better than Bostream's, dispite the bandwidth advertised), are preparing to roll out a 100 Mbps service next year, with a 300GB/mo traffic limit, rumored to cost in the neighbourhood of 120 USD/mo; I wonder how people will survive such terribly restrictive limits, heh.
In most urban scenarios, there isn't any divide and conquer going on. Having many providers competing in the same areas has its advantages.
I've never even been to Sweden, but happen to know a lot of Swedish netizens. Most of those are hard-core gamers, the most demanding users you'll find; anything short of 1MB/sec downloads and 10ms latency domestically, and they'll be screaming.
Take a look at the network traffic of any university. Can you really blame electronics companies for not being trusting of their target market?
Looks like you're confusing electronics companies with content providers. An elecronics company doesn't care much as long as people are buying its products. The computing world in particular sees 'users' where content providers see 'customers'. The demise of the all-purpose Personal Computer to satisfy the content providers would be the ultimate loss to all, in my opinion.
1. Block all email that contains HTML.. I mean how exciting can a text email be:)... Kills the marketing BS.
Agreed, this is an immensely useful measure; HTML e-mail simply isn't too useful. This'll also kill all the tracking bugs.
2. Institute a block all email except where you have whitelisted the sender...
Powerful, but a huge sacrifice. Feels like throwing in the towel to me.
3. Allow the sender to get prioritized by requiring them the first time to respond to an email and identify who they are and why they are contacting you.
Challenge-Response causes backscatter to innocent bystanders. Think of worms and spam with falsified from: headers. Using C-R makes you a part of the problem, not the solution.
An acceptable price to pay for a pretty effective measure, in my opinion. Just buy a domain, or use your ISP's mail servers. DUL lists are already employed by various companies, to kill spam from end-user ranges and viruses using their own smtp engine. It may by annoying, but it sure is effective.
In Iceland, around two thirds of the 290.000 inhabitants live in the capital area, but every town with 1000 people or more has access to ADSL (up to 2048/512). By the end of this year, the threshold will be 500. The second largest town has around 15000 people, the third closer to 5000.
And yes, the two telcos (it's essentially a duopoly) compete in the same geographical areas, on price, service and innovation; a far cry from the divide and empera strategies seemingly used in the US.
The real issue is that of sharing the files; that's when you violate the rights granted to the author by copyright law, and I don't think shelves full of CDs will change that. Of course, IANAL.
Slashdot Mirror
# Its parsing is too complex
7 101"
Complex or not, it's working just fine with quite an array of software.
# No sane firewall is going to let TXT records through
A firewall would need to examine the contents of packets to differentiate a TXT record from a, say, A record or cname. Comparable wizardry is already being performed by mail servers world wide, on a vast scale:
[smegma@cartman smegma]$ host -t txt 84.137.116.38.sbl.spamhaus.org
84.137.116.38.sbl.spamhaus.org text "http://www.spamhaus.org/SBL/sbl.lasso?query=SBL1
# No sane firewall is going to let TCP DNS packets through
SPF does not rely on anything that every other app using DNS doesn't. Also see above.
# The parsing can loop forever
In a horribly written parser perhaps. The same could be said about IRC clients, Web Browsers and just about any application out there.
# It will increase DNS scaning as spamers hunt for broken SPF records
DNS is quite efficient. Unlike RBLs, SPF will work just fine with traditional SOA settings, so cache hits will be plentiful.
# Its too complex to be implimented inside the MTA where it needs to be done
Just where do you think it's already being implemented?
# It can't be properly parsed in sendmail
It is already being used with all popular MTAs, including sendmail, postfix, qmail, exim, courier and ms exchange.
# ISO 8839 8859 59-15 utf-8 issues for domain names may kill some dns servers
Huh?
Parsing complexity might become a bit of a concern with the advent of XML, but as of now, it's dead-simple.
3, Interesting? And I feel like I'm feeding a troll here!
Use your own domain for outgoing mail AND either do not publish an SPF record for it, or publish one pointing to your Internet Serive Provider's SMTP server.
If the DNS provider for your domain doesn't enable you to publish an SPF (needs a 'txt' record), you'll need to ask them to, or change providers. But something tells me the laggard will quickly figure it out, since SPF is nearing critical mass.
True, to some extent, but even power plants burning fossil fuels are orders of magnitude more efficient than small internal combustion engines. You also get the added benefit of mitigating city smog.
A few hints on the fps/refresh subject:
Agreed. But another key advantage of big, public BLs vs private ones is to the unlikeliest of parties; the victims whose resources have been abused by spamming parasites.
If you'd find yourself blocked by thousands of admins' private lists, imagine getting the fiasco cleared up. Next to impossible. On the other hand, public blocklists usually have a simple, straight-forward removal procedure that immediately (blocklists typically use a very short refresh interval) takes effect for all its users.
If you don't agree with a BL's listing criteria or policies, don't use it.
There's a variety of DNSBLs out there. Some attempt to list spam sources (IPs from which spam is injected) with surgical precision whilst others go for the 'spam support' services, typically listing increasing swaths of space as the responsible party refuses to act (SPEWS for one).
In many cases the surgical approach simply won't do. Playing whack-a-mole with a fake ISP/spam support service isn't everybody's game.
Possibility 3) Intel wants to enable customers to better realize the differences between their own models; take a retailer's flashy ad for a "2.4GHz P4 Celeron" next to a Northwood at twice the price, yet showing the same clock rate. This could obviously help their (Intel's) customers. Considering the present offerings from AMD - and more specifically their recent 64 bit models - I don't think Intel would want a head-to-head scale at this time. This would definitely be helpful to all consumers, but probably not what Intel wants right now. Things will get very interesting in the next 2-3 years.
A lot of spam does indeed 'come from' Korea, China, Brazil etc, but a vast majority of it is still being sent by US spammers, using open relays and proxies in the aforementioned locations. Same with bullet-proof hosting of spamvertized webpages.
This map is a bit more readable.
Sweden, for some very strage reason, has three providers offering symmetrical, 10 Mbps or bigger connections for less than 65 USD a month. I'm not a swede, and don't have information in English, but you should be able to decipher what matters here. The yellow box below the headline has the speeds and prices. One Swedish krona is roughly 13.5 US cents, prices are per month and anslutningsavgift is the one-time hookup fee.
As you can see in the Aftonbladet article, Telia has just entered the fray. They were literally forced to do so, by competition from Bredbandsbolaget and Bostream.
Bredbandsbolaget, apparently not content with losing their edge (their connections are generally considered better than Bostream's, dispite the bandwidth advertised), are preparing to roll out a 100 Mbps service next year, with a 300GB/mo traffic limit, rumored to cost in the neighbourhood of 120 USD/mo; I wonder how people will survive such terribly restrictive limits, heh.
In most urban scenarios, there isn't any divide and conquer going on. Having many providers competing in the same areas has its advantages.
I've never even been to Sweden, but happen to know a lot of Swedish netizens. Most of those are hard-core gamers, the most demanding users you'll find; anything short of 1MB/sec downloads and 10ms latency domestically, and they'll be screaming.
Take a look at the network traffic of any university. Can you really blame electronics companies for not being trusting of their target market?
Looks like you're confusing electronics companies with content providers. An elecronics company doesn't care much as long as people are buying its products. The computing world in particular sees 'users' where content providers see 'customers'. The demise of the all-purpose Personal Computer to satisfy the content providers would be the ultimate loss to all, in my opinion.
I like this NANAE post by Steve Linford much better. Especially the last paragraph.
1. Block all email that contains HTML.. I mean how exciting can a text email be :)... Kills the marketing BS.
Agreed, this is an immensely useful measure; HTML e-mail simply isn't too useful. This'll also kill all the tracking bugs.
2. Institute a block all email except where you have whitelisted the sender...
Powerful, but a huge sacrifice. Feels like throwing in the towel to me.
3. Allow the sender to get prioritized by requiring them the first time to respond to an email and identify who they are and why they are contacting you.
Challenge-Response causes backscatter to innocent bystanders. Think of worms and spam with falsified from: headers. Using C-R makes you a part of the problem, not the solution.
Voila. Detailed information to be found there.
Gigli++! *Shivers*.
An acceptable price to pay for a pretty effective measure, in my opinion. Just buy a domain, or use your ISP's mail servers. DUL lists are already employed by various companies, to kill spam from end-user ranges and viruses using their own smtp engine. It may by annoying, but it sure is effective.
Enter The Darknet.
In Iceland, around two thirds of the 290.000 inhabitants live in the capital area, but every town with 1000 people or more has access to ADSL (up to 2048/512). By the end of this year, the threshold will be 500. The second largest town has around 15000 people, the third closer to 5000. And yes, the two telcos (it's essentially a duopoly) compete in the same geographical areas, on price, service and innovation; a far cry from the divide and empera strategies seemingly used in the US.
The real issue is that of sharing the files; that's when you violate the rights granted to the author by copyright law, and I don't think shelves full of CDs will change that. Of course, IANAL.