Sorry, but I don't know which hysteria you mean, and where you read that I say that "they" are interchangeable.
You wrote "the ij is originally a y with an umlaut" which is not true. Not because the dots on the U+00FF aren't an umlaut (which they aren't, functionally they aren't a diaeresis either), but because the origin of the Dutch letter 'ij' is otherwise.
The U+00FF Y DIAERESIS (dotted-y) is used in French names and in some other languages, but is erroneous for Dutch. The dots on the dotted-y are not an umlaut (like in Mann/Männer, compare also mouse/mice) or a functional diaeresis (like in coöperate, naïve, preëmptive, reënact, Chloë, Zoë), they are diacritical marks to show that it is a different letter. More on Umlaut
In old Dutch handwriting (about 16th century) you will find the mix I mentioned, such as a shape that looks a bit like the U+00FF dotted-y, but it shouldn't be represented by that.
In contemporary handwriting you often see a shape like a lowly connected i-j, with a dot on top of each of the two verticals. Just like a handwritten 'g' can look like an 'a' with a descender, a handwritten 'ij' can look like a 'u' with a descender (with or without dots).
The Russian character that when printed looks like 'bI' is often written with a shape like 'y'. That also doesn't mean that they are interchangeable.
Aaargh, stop calling the Dutch character ij (U+0133, capital U+0132) a dotted y (U+00FF).
One of the main origins of the Dutch letter ij is the double-i, of which the last i was written with extra length. You can find examples of ij and iij in old handwritings for numbering pages (Roman numbers 2 and 3). In a word like "bijzonder" the "ij" still has the "long-i" sound.
In old Dutch handwriting, the ij-ligature and the y and the dotted y were freely mixed. This resulted in many spelling variations of the same name, for example Bruin, Bruyn and Bruijn. In Dutch phonebooks, Bruyn and Bruijn are sorted together. An opinion: The Dutch "letter" IJ
Fighting spam should first be done at the gate, by the receiving mail server. To facilitate that, SMTP should be extended, to create a non-bulk variant of the standard.
This requires a few extensions to SMTP, so that the receiving mail server is allowed to
limit the number of recipients per message
limit, per IP-nr and subnet, the total number of messages and recipients per time window (like 24 hours)
limit the number of concurrent connections, per IP-nr and subnet
communicate these limits, and the current quota for the specific client, at the initiation phase
Of course, current SMTP implementations can already impose such limits, by rejecting to process the message, like with a 554-reply.
The good thing of hashes here is that they are not unique.
Create hashes with enough bits on file-parts, for example SHA-1.
Each part is just random anonymous data, findable by its hash.
When a reader has retrieved m-out-of-n parts, it can create a result file. With (m-1) parts it can not create a file that comes close to that.
The result file is the original file, or a file with a clue about how to change it into the original (like XOR it with the same number of bytes from some popular CD-R).
The indexes (like filenames, and a list of belonging hashes, and ueber-hashes) MUST be totally separated from the parts. The indexes are published 'somewhere else', like on specialized and actively mirroring websites/usenet/irc.
There MUST be no central notion at all of which node can supply the data from any specific hash.
Clients tell the neighbouring clients which hashes they are looking for, and that gets spread.
A client MUST not know anything of relations between the hashes that is has. A client can serve any collection of hashes that it finds fit.
PRINCE (PRojects IN Controlled Environments) is a structured method for effective project management. It is a de facto standard used extensively by the UK Government and is widely recognised and used in the private sector, both in the UK and internationally. PRINCE, the method, is in the public domain, offering non-proprietorial best-practice guidance on project management.
It also does a 20 second delay before sending the reject code, to slow down the spammer from moving on to their next target.
Delays are useful (see also sendmail's greet_pause) but it won't slow spammers down much, because they don't follow rfc2821. They don't wait for responses but just start pumping bytes. Consider blocking at the firewall.
The allowed time-out of the Initial 220 Message: 5 minutes (see rfc2821, 4.5.3.2). Plenty of time to do DNSBL first, no need to wake-up the SMTP-server.:)
I have been creating spam-detection-software for many years now. My rules act on structure and metadata, not on content, never on content. All you need is procmail and sed.
With SpamAssassin you can achieve about the same result as I do. But disable all content-based rules, because they don't scale and they get worked around. Even checks on the geographical origin of the urls in a message, won't survive.
Detecting spam on content is (and always have been) a dead end street. When Bayesian filters came around, I just thought: o no, another weak spot.
IPv6 can be the next anti-spam-problem: just too many IP-addresses to blacklist.
For some people, small doses of cafeine, nicotine, etc. saves their lives. Without it they are grumpy suicidal unsocial people.
These substances act almost directly on the brain and if you are relatively lacking behaviour that gets equalled by a little coffee or a few cigarettes, like 10% of humanity does, then just keep using them.
If two cigarettes a day keep your shrink away, smoke'm.
"The message digest algorithm used by older versions of PGP is the MD5 Message Digest Algorithm, placed in the public domain by RSA Data Security, Inc.
MD5 is a 128-bit hash algorithm. In 1996, MD5 was all but broken by a German cryptographer, Hans Dobbertin. Although MD5 was not completely broken at that time, it was discovered to have such serious weaknesses that no one should keep using it to generate signatures. Further work in this area might completely break it, allowing signatures to be forged.
If you don't want to someday find your PGP digital signature on a forged confession, you might be well advised to migrate to the new PGP DSS keys as your preferred method for making digital signatures, because DSS uses SHA as its secure hash algorithm." [page 234 of the pdf which comes with PGP 6.5.1. int]
The Euro-sign is visible here [] in the preview. Of course, it doesn't exist in ISO-8859-1 (the HTML default). It is in ISO-8859-15. And in Unicode of course: [] should show it.
Nikita recently invented and implemented a clever bit of code that keeps track of the highest node in the tree that spans a directory, and then performs [...]
Slashdot Mirror
Sorry, but I don't know which hysteria you mean, and where you read that I say that "they" are interchangeable.
You wrote "the ij is originally a y with an umlaut" which is not true. Not because the dots on the U+00FF aren't an umlaut (which they aren't, functionally they aren't a diaeresis either), but because the origin of the Dutch letter 'ij' is otherwise.
The U+00FF Y DIAERESIS (dotted-y) is used in French names and in some other languages, but is erroneous for Dutch. The dots on the dotted-y are not an umlaut (like in Mann/Männer, compare also mouse/mice) or a functional diaeresis (like in coöperate, naïve, preëmptive, reënact, Chloë, Zoë), they are diacritical marks to show that it is a different letter. More on Umlaut
In old Dutch handwriting (about 16th century) you will find the mix I mentioned, such as a shape that looks a bit like the U+00FF dotted-y, but it shouldn't be represented by that.
In contemporary handwriting you often see a shape like a lowly connected i-j, with a dot on top of each of the two verticals. Just like a handwritten 'g' can look like an 'a' with a descender, a handwritten 'ij' can look like a 'u' with a descender (with or without dots).
The Russian character that when printed looks like 'bI' is often written with a shape like 'y'. That also doesn't mean that they are interchangeable.
Aaargh, stop calling the Dutch character ij (U+0133, capital U+0132) a dotted y (U+00FF).
One of the main origins of the Dutch letter ij is the double-i, of which the last i was written with extra length. You can find examples of ij and iij in old handwritings for numbering pages (Roman numbers 2 and 3). In a word like "bijzonder" the "ij" still has the "long-i" sound.
In old Dutch handwriting, the ij-ligature and the y and the dotted y were freely mixed. This resulted in many spelling variations of the same name, for example Bruin, Bruyn and Bruijn. In Dutch phonebooks, Bruyn and Bruijn are sorted together. An opinion: The Dutch "letter" IJ
Fighting spam should first be done at the gate, by the receiving mail server. To facilitate that, SMTP should be extended, to create a non-bulk variant of the standard.
This requires a few extensions to SMTP, so that the receiving mail server is allowed to
- limit the number of recipients per message
- limit, per IP-nr and subnet, the total number of messages and recipients per time window (like 24 hours)
- limit the number of concurrent connections, per IP-nr and subnet
- communicate these limits, and the current quota for the specific client, at the initiation phase
Of course, current SMTP implementations can already impose such limits, by rejecting to process the message, like with a 554-reply.Some cams break when you alternatively use Sony and TDK tapes. article with links
Simplex is also used for wavelet decomposition.
The good thing of hashes here is that they are not unique.
The indexes are published 'somewhere else', like on specialized and actively mirroring websites/usenet/irc.
a{n}b{n}c{n}
What is PRINCE? http://www.prince2.com/whatisp2.html
PRINCE (PRojects IN Controlled Environments) is a structured method for effective project management. It is a de facto standard used extensively by the UK Government and is widely recognised and used in the private sector, both in the UK and internationally. PRINCE, the method, is in the public domain, offering non-proprietorial best-practice guidance on project management.
I am looking forward to a 'HowNotTo'-operator.
Not really started in Russia:
Historical Context of Phage Therapy
More: BBC
It also does a 20 second delay before sending the reject code, to slow down the spammer from moving on to their next target.
:)
Delays are useful (see also sendmail's greet_pause) but it won't slow spammers down much, because they don't follow rfc2821.
They don't wait for responses but just start pumping bytes.
Consider blocking at the firewall.
The allowed time-out of the Initial 220 Message: 5 minutes (see rfc2821, 4.5.3.2). Plenty of time
to do DNSBL first, no need to wake-up the SMTP-server.
But that would remove the Prozac from the water of the UK, I am not sure of that to be a good idea.
google: prozac uk drinking
That bandwidth is not spilled if you make your MTA do SMTP-REJECTs, based on the high-quality-blacklists around.
base64- and qp-decoding in 'just procmail' (see a.o. bq_demo.rc)
the temporary restraining order (TRO) against SpamCop.net [...] has been dissolved
.
There is no difference necessary. Your consciousness evens out anything, to keep you happy with what you believe. It's a jungle out there, you know!
Another morse encoder and decoder, with JavaScript-source: http://www.xs4all.nl/~rvtol/morse_js.html
I have been creating spam-detection-software for many years now. My rules act on structure and metadata, not on content, never on content. All you need is procmail and sed.
With SpamAssassin you can achieve about the same result as I do. But disable all content-based rules, because they don't scale and they get worked around. Even checks on the geographical origin of the urls in a message, won't survive.
Detecting spam on content is (and always have been) a dead end street. When Bayesian filters came around, I just thought: o no, another weak spot.
IPv6 can be the next anti-spam-problem: just too many IP-addresses to blacklist.
For some people, small doses of cafeine, nicotine, etc. saves their lives. Without it they are grumpy suicidal unsocial people.
These substances act almost directly on the brain and if you are relatively lacking behaviour that gets equalled by a little coffee or a few cigarettes, like 10% of humanity does, then just keep using them.
If two cigarettes a day keep your shrink away, smoke'm.
PZ's text is also on page 45 of IntroToCrypto.pdf
Phil Zimmermann says this about MD5:
"The message digest algorithm used by older versions of PGP is the MD5 Message Digest Algorithm, placed in the public domain by RSA Data Security, Inc.
MD5 is a 128-bit hash algorithm. In 1996, MD5 was all but broken by a German cryptographer, Hans Dobbertin. Although MD5 was not completely broken at that time, it was discovered to have such serious weaknesses that no one should keep using it to generate signatures. Further work in this area might completely break it, allowing signatures to be forged.
If you don't want to someday find your PGP digital signature on a forged confession, you might be well advised to migrate to the new PGP DSS keys as your preferred method for making digital signatures, because DSS uses SHA as its secure hash algorithm." [page 234 of the pdf which comes with PGP 6.5.1. int]
because spammers react on them. An other alternative (SMTP-with-a-delay)
greylisting
The Euro-sign is visible here [] in the preview. Of course, it doesn't exist in ISO-8859-1 (the HTML default). It is in ISO-8859-15. And in Unicode of course: [] should show it.
Euro:
Do n ' t f o r g e t t o p a t e n t i t.