This whole discussion reminds me of the often quoted phrase "Premature optimisation is the root of all evil" but you bring up an interesting point that I disagree with. There is a place for gperf in command line processing, its just not for production programs. It is fine for experimental programs as a training exercise.
The fairy tale connection is why I tell some congresscritters that if they think a 20 year extension is good, then a 50 or even a 100 year extension should be better. Some of them have been paid beyond the ability to listen to the people who can vote them out of office, so the only other choice is to get them to extend things way past the line of common sense and have it get voted down on the floor.
I would guess 160 terrabytes per year. I figure one album per 1000 people per year for 1st world countries. The 1 album per 1000 people comes from a contest a few years ago where a radio station had a contest where bands had to send in a CD they made in the last year and they got about 3000 entries out of listening population of about 3 million people. 160 terrabytes is based on world wide production of CD at the same rate.
Did you know the current system is based on the concept that everything in the system can fail at once and planes won't run into each other? Thats one of the reasons that towers still move little bits of paper around using a very well defined procedure.
However, if you reject email delivered to postmaster@your_domain, then your mail system isn't configured right, and you should expect to be blacklisted.
I've been running email systems for over 2 decades and I can't remember the last time I got a message via smtp to a postmaster address that was legit. It was common in the early days of uucp/snmp/pmdf gateways to send something to the admins when things broke which is why the postmaster address showed up in the RFC but that was long ago and things moved on.
I suspect that the rfc-ignorant list encourages people to just set up a postmaster address that goes to/dev/null A SNMP bouce that says "Hi. if you need to contact a real person, please resend your message to where it will be properly addressed" This means that real problems can get fixed and the spamers have less people reading their junk.
Who gets legit mail to postmaster from external hosts?
I hate the rfc-ignorant list. My domain doesn't have any spam going out of it and it never will (due to a shoot first and ask questions later policy combined with terms and conditions involving using site abusers for medical experiments). I've annoyed a few spamers in the past so I get my domain name in from addresses from time to time so every once in a while I will get a real person with a legit complaint however the postmaster address is now getting several thousand messages a day and I have no choice but to remove it.
Maybe its time for RFC 3821 which says the human abuse and postmaster address should be encoded in the SMTP error message...
You point out the Swiss system as an option to the Canadian and UK systems that keep showing up in the news as if they are the only other choice. The Australian system is a public/private system where everyone is covered with a 1.5% income tax and if your income exceeds some threshold, there is a penalty if you don't get supplemental private insurance. The result is if you end up in the hospital and have private insurance, you get a private room and better TV but if you don't have the private insurance you will end up in the public hospital maybe in a shared ward but still get treated. The real advantage of this system is that it allows new techniques to be brought to the country and practised in the private hospitals while the treatments are still in their early adopter high price range. That results in a high level of care (like a good US research hospital) but with far less drain on everyone's pocket book.
I could get you 45mb for about AU$100,000/y in Sydney but it will cost you an extra $50k in Brisbane or Melbourne. If your more than a few km outside of the core areas then the price goes up very rapidly because of tail charges. In Sydney its about $200/mb to talk to Telstra, $200/mb to talk to the rest of the world and $20/mb to talk to the peering points. Then its about $100/mb to talk to the nearest major cities.
I've wondered if that point is invalid since no one knows what the later versions state. Paragraphs in contracts agreeing to unspecified and arbitrary terms later tends to not go over so well with judges. In the case of the GPL, the intent of the GPL can be applied to that claim maybe and I think its a big maybe.
While most people here think if RIAA as an evil anti-sharing group, back before they turned to the dark side, they used to set decent audio standards. Too bad that was in the era when hi-fi records were new.
The volume compression crud is one of their more recent "technology advancements". Volume compression is isn't data compression but reefers to horizontally compressing the waveform or boosting the quiet bits and cutting the loud bits.
This is why modern music has no emotion. The soft bits get boosted and the high energy bits get clipped. It is why most remastered CDs suck so bad.
Its also why rap is so popular. Rap's verbal beat messes up the auto-compressors and break them and since rap is about the only modern music that has an energy, its got a huge younger following.
I don't think the Google spell checker is purely statistical since it seems to cope with my inability to spell Old French words that find their way into modern English. The best spell checker that I've seen was part of the AT&T Documenters Workbench. It takes into account odd things like how a specific user mistypes as well as word associations. The scary thing is I used it with MS Word for Unix on a 3b2 over a serial terminal.
Its easier to replace the libraries and they are less likely to be properly checked.
The more complex systems get, the easier it is to hide stuff and there just aren't the tools to easily check these things. A diff of a text file isn't that much simpler than a diff of two xml files but it reduced the number of people who will ever check it. Binary config files are even worse. Many major Unix flavors now use binary files (often under the gules of optimizing startup) that can be hand edited with a binary editor to do very interesting and nearly undetectable things on shutdown. Its real handy to be able to do thing after the processing accounting system disk has been unmounted if you want to do things undetected.
There was a month before they figured out what they had done. In the chain of trust situation, the spamer would have made use of that trust for a whole month which means billions of messages. If a full chain of trust thing happens, there will be people all over the social networking sties offing popular people $100 to be in their chain of trust. You plan fails to take into account that the spammers have money to throw around.
Lets just take 1st degree. I only know one person who murdered another person I knew and due to mental instability, no one would have predicted that. Take two steps to people I know who are in the corrections field who work with lots of murders. If I'm two steps on your chain of trust of people to who deserve to be in jail forever, where does that leave your chin of trust? Chain of trust isn't going to work either.
Oh, I also know people how bought marketing services from people who ended up being spammers as well.
What universe are you living in? That only applies if your toaster gets an address from your fridge that gets its from the home computer. In the real world we have several upstream providers and IPv6 sucks at that.
I get a few messages a month from people that your system would say are spamers. There is no way to tell a legit 1st contact email message from a spammer on todays net.
They don't allocate IP addresses, they allocate routes entry and with route entries, you get way more addresses than most need. The solution for this is to start allocating non-contigious/24... Force everyone to fix their routing and treat the wold as a 2^24/24 ranges and get over it. To do this right requires less than 8mb of cache tag ram in most routers that want full feeds and enough ram to process the bgp routing updates.
Going to IPv6 doesn't fix the fact that routers are running out of routes. This problem will get plenty of attention in about 2 months when the big Cisco routers start to dump routes because they are too big and adding IPv6 only makes the problem much worse.
Yep, limit it to 16,777,216/24 routes. That will fix it. If your router has 16 interfaces, you can do this with 8mb of cache ram to make the quick decisions and whatever else you need to processes the routes.
leads me to think this was a problem which was probably reported numerous times to middle management and perpetually postponed Who's middle management? Cisco's?
Their routers have been perpetually running out of memory for reasonable routing tables since at least 1992.
Slashdot Mirror
Active servers don't need any content.
I would like to see the stats of domains with at least 10 pages.
This whole discussion reminds me of the often quoted phrase "Premature optimisation is the root of all evil" but you bring up an interesting point that I disagree with.
There is a place for gperf in command line processing, its just not for production programs. It is fine for experimental programs as a training exercise.
The fairy tale connection is why I tell some congresscritters that if they think a 20 year extension is good, then a 50 or even a 100 year extension should be better. Some of them have been paid beyond the ability to listen to the people who can vote them out of office, so the only other choice is to get them to extend things way past the line of common sense and have it get voted down on the floor.
I would guess 160 terrabytes per year.
I figure one album per 1000 people per year for 1st world countries. The 1 album per 1000 people comes from a contest a few years ago where a radio station had a contest where bands had to send in a CD they made in the last year and they got about 3000 entries out of listening population of about 3 million people. 160 terrabytes is based on world wide production of CD at the same rate.
Did you know the current system is based on the concept that everything in the system can fail at once and planes won't run into each other? Thats one of the reasons that towers still move little bits of paper around using a very well defined procedure.
However, if you reject email delivered to postmaster@your_domain, then your mail system isn't configured right, and you should expect to be blacklisted.
/dev/null
I've been running email systems for over 2 decades and I can't remember the last time I got a message via smtp to a postmaster address that was legit. It was common in the early days of uucp/snmp/pmdf gateways to send something to the admins when things broke which is why the postmaster address showed up in the RFC but that was long ago and things moved on.
I suspect that the rfc-ignorant list encourages people to just set up a postmaster address that goes to
A SNMP bouce that says "Hi. if you need to contact a real person, please resend your message to where
it will be properly addressed"
This means that real problems can get fixed and the spamers have less people reading their junk.
Who gets legit mail to postmaster from external hosts?
I hate the rfc-ignorant list.
My domain doesn't have any spam going out of it and it never will (due to a shoot first and ask questions later policy combined with terms and conditions involving using site abusers for medical experiments). I've annoyed a few spamers in the past so I get my domain name in from addresses from time to time so every once in a while I will get a real person with a legit complaint however the postmaster address is now getting several thousand messages a day and I have no choice but to remove it.
Maybe its time for RFC 3821 which says the human abuse and postmaster address should be encoded in the SMTP error message...
You point out the Swiss system as an option to the Canadian and UK systems that keep showing up in the news as if they are the only other choice. The Australian system is a public/private system where everyone is covered with a 1.5% income tax and if your income exceeds some threshold, there is a penalty if you don't get supplemental private insurance. The result is if you end up in the hospital and have private insurance, you get a private room and better TV but if you don't have the private insurance you will end up in the public hospital maybe in a shared ward but still get treated. The real advantage of this system is that it allows new techniques to be brought to the country and practised in the private hospitals while the treatments are still in their early adopter high price range. That results in a high level of care (like a good US research hospital) but with far less drain on everyone's pocket book.
Its what we in the programming field would call the Data Segment.
I could get you 45mb for about AU$100,000/y in Sydney but it will cost you an extra $50k in Brisbane or Melbourne. If your more than a few km outside of the core areas then the price goes up very rapidly because of tail charges. In Sydney its about $200/mb to talk to Telstra, $200/mb to talk to the rest of the world and $20/mb to talk to the peering points. Then its about $100/mb to talk to the nearest major cities.
I've wondered if that point is invalid since no one knows what the later versions state. Paragraphs in contracts agreeing to unspecified and arbitrary terms later tends to not go over so well with judges. In the case of the GPL, the intent of the GPL can be applied to that claim maybe and I think its a big maybe.
While most people here think if RIAA as an evil anti-sharing group, back before they turned to the dark side, they used to set decent audio standards. Too bad that was in the era when hi-fi records were new.
The volume compression crud is one of their more recent "technology advancements". Volume compression is isn't data compression but reefers to horizontally compressing the waveform or boosting the quiet bits and cutting the loud bits.
This is why modern music has no emotion. The soft bits get boosted and the high energy bits get clipped. It is why most remastered CDs suck so bad.
Its also why rap is so popular. Rap's verbal beat messes up the auto-compressors and break them and since rap is about the only modern music that has an energy, its got a huge younger following.
I'm researching ways around the at&t contract since their towers don't reach this far over the Pacific and I want a shiny new toy to crack open.
Anyone know how to open one yet? I'm hoping its like an ipod or I'll break a shiny new toy.
Who will be 1st to get it working on an non-at&t network?
I don't think the Google spell checker is purely statistical since it seems to cope with my inability to spell Old French words that find their way into modern English.
The best spell checker that I've seen was part of the AT&T Documenters Workbench. It takes into account odd things like how a specific user mistypes as well as word associations. The scary thing is I used it with MS Word for Unix on a 3b2 over a serial terminal.
Thats ok because someone rewrote the compiler to put in back doors too.
Its easier to replace the libraries and they are less likely to be properly checked.
The more complex systems get, the easier it is to hide stuff and there just aren't the tools to easily check these things. A diff of a text file isn't that much simpler than a diff of two xml files but it reduced the number of people who will ever check it. Binary config files are even worse. Many major Unix flavors now use binary files (often under the gules of optimizing startup) that can be hand edited with a binary editor to do very interesting and nearly undetectable things on shutdown. Its real handy to be able to do thing after the processing accounting system disk has been unmounted if you want to do things undetected.
There was a month before they figured out what they had done. In the chain of trust situation, the spamer would have made use of that trust for a whole month which means billions of messages. If a full chain of trust thing happens, there will be people all over the social networking sties offing popular people $100 to be in their chain of trust. You plan fails to take into account that the spammers have money to throw around.
Lets just take 1st degree. I only know one person who murdered another person I knew and due to mental instability, no one would have predicted that. Take two steps to people I know who are in the corrections field who work with lots of murders. If I'm two steps on your chain of trust of people to who deserve to be in jail forever, where does that leave your chin of trust? Chain of trust isn't going to work either.
Oh, I also know people how bought marketing services from people who ended up being spammers as well.
What universe are you living in? That only applies if your toaster gets an address from your fridge that gets its from the home computer. In the real world we have several upstream providers and IPv6 sucks at that.
I get a few messages a month from people that your system would say are spamers. There is no way to tell a legit 1st contact email message from a spammer on todays net.
If IETF has been working on it for years , its broken.
They don't allocate IP addresses, they allocate routes entry and with route entries, you get way more addresses than most need. The solution for this is to start allocating non-contigious /24... Force everyone to fix their routing and treat the wold as a 2^24 /24 ranges and get over it. To do this right requires less than 8mb of cache tag ram in most routers that want full feeds and enough ram to process the bgp routing updates.
Going to IPv6 doesn't fix the fact that routers are running out of routes. This problem will get plenty of attention in about 2 months when the big Cisco routers start to dump routes because they are too big and adding IPv6 only makes the problem much worse.
Yep, limit it to 16,777,216 /24 routes. That will fix it. If your router has 16 interfaces, you can do this with 8mb of cache ram to make the quick decisions and whatever else you need to processes the routes.
leads me to think this was a problem which was probably reported numerous times to middle management and perpetually postponed
Who's middle management? Cisco's?
Their routers have been perpetually running out of memory for reasonable routing tables since at least 1992.
Thats why more and more things are using 3 batteries... but the charges are designed to charge in pairs.