Re: SRS Timestamp
"Stuart Gathman via spf-discuss" <[email protected]> Thu, 08 May 2014 15:24:00 -0400
| Newsgroups | gmane.mail.spam.spf.discuss |
|---|---|
| Organization | BWI Corporation |
| Message-ID | <[email protected]> |
This is a multi-part message in MIME format. --------------080902080408050105040305 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: quoted-printable On 05/08/2014 02:42 PM, Dale Moore via spf-discuss wrote: > I would post this to the mailing list [email protected]=20 > <mailto:[email protected]>, but it appers that list is=20 > deprecated in spite of being listed at=20 > http://www.libsrs2.org/support.html . > > Please forgive me if this has been discussed before, but I did some=20 > searching > and didn't find anything appropriate. > > In the paper http://www.libsrs2.org/srs/srs.pdf > > The resolution is one day, and it is a 10 bit number stored as 2=20 > base32 characters. > The timestamp is computed as unix time / (60 * 60 * 24) mod 2 ^ 10 . > > This gives 1024 possible timestamps, with approximately a 3.5 year=20 > cycle before a > timestamp becomes valid again. > > And on the wikipedia > http://en.wikipedia.org/wiki/Sender_Rewriting_Scheme > > * *The timestamp* (TT) has a one-day resolution in order to make the > address invalid after a few days. Computed as ^unix time > <http://en.wikipedia.org/wiki/Unix_time> =E2=81=84_(60*60*24) mod = 2^10 , > it can be stored as a two base32 > <http://en.wikipedia.org/wiki/Base32> characters, with a recycling > period of about 3.5 years. > > > So, wouldn't the cycle time be > - 2^10 days or > - 1024 days or > - (1024 / 365) years or > - 2.8 years? > > I don't understand where 3.5 years come from. Originally, the timestamp was 2 base64 characters for 4096 values, for=20 an 11.2 year cycle. But braindead mailers keep smashing case, and no=20 amount of preaching, or handing out copies of the email RFCs on street=20 corners seemed to be able to change that. So the developers switched to=20 two base 36 chars for the timestamp - for a 3.5 year cycle. A lot of=20 things were better with a power of 2 modulus, so the final version used=20 2 base 32 chars for a 2.8 year cycle. But the whitepapers were never=20 updated. ------------------------------------------- Sender Policy Framework: http://www.openspf.net [http://www.openspf.net] Modify Your Subscription: http://www.listbox.com/member/ [http://www.list= box.com/member/] Archives: https://www.listbox.com/member/archive/735/=3Dnow RSS Feed: https://www.listbox.com/member/archive/rss/735/6959934-50ec8f89 Modify Your Subscription: https://www.listbox.com/member/?member_id=3D695= 9934&id_secret=3D6959934-b7c4528d Unsubscribe Now: https://www.listbox.com/unsubscribe/?member_id=3D6959934= &id_secret=3D6959934-edadf31a&post_id=3D20140508152437:624A4DD6-D6E6-11E3= -8E6B-A19AC288ED6A Powered by Listbox: http://www.listbox.com --------------080902080408050105040305 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable <html> <head> <meta content=3D"text/html; charset=3DUTF-8" http-equiv=3D"Content-Ty= pe"> </head> <body bgcolor=3D"#FFFFFF" text=3D"#000000"> <div class=3D"moz-cite-prefix">On 05/08/2014 02:42 PM, Dale Moore via spf-discuss wrote:<br> </div> <blockquote cite=3D"mid:[email protected]= ail.com" type=3D"cite"> <div dir=3D"ltr">I would post this to the mailing list <a moz-do-not-send=3D"true" href=3D"mailto:[email protected]= ">[email protected]</a>, but it appers that list is deprecated in spite of being listed at=C2=A0<a moz-do-not-send=3D"true" href=3D"http://www.libsrs2.org/support.html">http://www.libsrs2= .org/support.html</a> . <div> <br> </div> <div>Please forgive me if this has been discussed before, but I did some searching</div> <div>and didn't find anything appropriate.<br> <div><br> </div> <div>In the paper=C2=A0<a moz-do-not-send=3D"true" href=3D"http://www.libsrs2.org/srs/srs.pdf">http://www.libs= rs2.org/srs/srs.pdf</a></div> <div><br> </div> <div> <div>=C2=A0 The resolution is one day, and it is a 10 bit num= ber stored as 2 base32 characters.</div> <div>=C2=A0 The timestamp is computed as unix time / (60 * 60= * 24) =C2=A0mod 2 ^ 10 .</div> <div><br> </div> <div>=C2=A0 This gives 1024 possible timestamps, with approximately a 3.5 year cycle before a</div> <div>=C2=A0 timestamp becomes valid again.</div> </div> <div><br> </div> <div>And on the wikipedia</div> <div>=C2=A0 <a moz-do-not-send=3D"true" href=3D"http://en.wikipedia.org/wiki/Sender_Rewriting_Schem= e">http://en.wikipedia.org/wiki/Sender_Rewriting_Scheme</a></div> <div><br> </div> <div> <ul style=3D"line-height:22.399999618530273px;margin:0.3em 0p= x 0px 1.6em;padding:0px;color:rgb(37,37,37);font-family:sans-serif;font-size:14= px"> <li style=3D"margin-bottom:0.1em"><b>The timestamp</b>=C2=A0= (TT) has a one-day resolution in order to make the address invalid after a few days. Computed as=C2=A0<span class=3D= "" style=3D"font-family:'Nimbus Roman No9 L','Times New Roman',Times,serif;font-size:17px;white-space:nowrap"><= span class=3D""><sup style=3D"line-height:1"><var><a moz-do-not-send=3D"true" href=3D"http://en.wikipedia.org/wiki/Unix_time" title=3D"Unix time" style=3D"text-decoration:none;color:rgb(11,0,12= 8);background-image:none">unix time</a></var></sup>=E2=81=84<sub style=3D"line-height:1">(60*60*24)</sub></span>=C2=A0= mod 2<sup style=3D"line-height:1">10</sup></span>, it can b= e stored as a two=C2=A0<a moz-do-not-send=3D"true" href=3D"http://en.wikipedia.org/wiki/Base32" title=3D"Base32" style=3D"text-decoration:none;color:rgb(11,0,128);backg= round-image:none">base32</a>=C2=A0characters, with a recycling period of about 3.5 years. =C2=A0</li> </ul> <div><font color=3D"#252525" face=3D"sans-serif"><span style=3D"font-size:14px;line-height:22.399999618530273p= x"><br> </span></font></div> </div> <div><font color=3D"#252525" face=3D"sans-serif"><span style=3D"font-size:14px;line-height:22.399999618530273px"= >So, wouldn't the cycle time be=C2=A0</span></font></div> <div><font color=3D"#252525" face=3D"sans-serif"><span style=3D"font-size:14px;line-height:22.399999618530273px"= >=C2=A0 - 2^10 days or=C2=A0</span></font></div> <div><font color=3D"#252525" face=3D"sans-serif"><span style=3D"font-size:14px;line-height:22.399999618530273px"= >=C2=A0 - 1024 days or =C2=A0</span></font></div> <div><font color=3D"#252525" face=3D"sans-serif"><span style=3D"font-size:14px;line-height:22.399999618530273px"= >=C2=A0 - (1024 / 365) years or</span></font></div> <div><font color=3D"#252525" face=3D"sans-serif"><span style=3D"font-size:14px;line-height:22.399999618530273px"= >=C2=A0 - 2.8 years? =C2=A0</span></font></div> <div><font color=3D"#252525" face=3D"sans-serif"><span style=3D"font-size:14px;line-height:22.399999618530273px"= ><br> </span></font></div> <div><font color=3D"#252525" face=3D"sans-serif"><span style=3D"font-size:14px;line-height:22.399999618530273px"= >I don't understand where 3.5 years come from.</span></font>= </div> </div> </div> </blockquote> <font color=3D"#252525">Originally, the timestamp was 2 base64 characters for 4096 values, for an 11.2 year cycle.=C2=A0 But brain= dead mailers keep smashing case, and no amount of preaching, or handing out copies of the email RFCs on street corners seemed to be able to change that.=C2=A0 So the developers switched to two base 36 cha= rs for the timestamp - for a 3.5 year cycle.=C2=A0 A lot of things wer= e better with a power of 2 modulus, so the final version used 2 base 32 chars for a 2.8 year cycle.=C2=A0 But the whitepapers were never updated.<br> </font><br> <div style=3D"width:auto;margin:0;padding:5px;background-color:#fff;cle= ar:both;border-top: 1px solid #ccc;" bgcolor=3D"#ffffff"> <p style=3D'font-family:Arial,sans-serif;margin:0.5em auto'>Sender Policy= Framework: <a href=3D"http://www.openspf.net">http://www.openspf.net</a>= <br> Modify Your Subscription: <a href=3D"http://www.listbox.com/member/">http= ://www.listbox.com/member/</a></p> <table border=3D"0" cellspacing=3D"0" cellpadding=3D"0" width=3D"100%" st= yle=3D"background-color:#fff" bgcolor=3D"#ffffff"> <tr> <td padding=3D"4px"> <font color=3D"#333333" size=3D"1" style=3D"font-family:helvetica, = sans-serif;"> <a style=3D"text-decoration:none;color:#669933;border-bottom: 1px s= olid #444444" href=3D"https://www.listbox.com/member/archive/735/=3Dnow" title=3D"Go to= archives for spf-discuss">Archives</a> <a border=3D"0" style=3D"text-decoration:none;color:#669933" href=3D"http= s://www.listbox.com/member/archive/rss/735/6959934-50ec8f89" title=3D"RSS= feed for spf-discuss"><img border=3D0 src=3D"https://www.listbox.com/ima= ges/feed-icon-10x10.jpg" /></a> | <a style=3D"text-decoration:none;color:#669933;border-bottom: 1px soli= d #444444" href=3D"https://www.listbox.com/member/?member_id=3D6959934&id_secret=3D6= 959934-b7c4528d" title=3D"">Modify</a> Your Subscription | <a style=3D"text-decoration:none;color:#669933;borde= r-bottom: 1px solid #444444" href=3D"https://www.listbox.com/unsubscribe/?member_id=3D6959934&id_secre= t=3D6959934-edadf31a&post_id=3D20140508152437:624A4DD6-D6E6-11E3-8E6B-A19= AC288ED6A" title=3D"">Unsubscribe Now</a> <td valign=3D"top" align=3D"right"><a style=3D"border-bottom:none;" href=3D= "http://www.listbox.com"> <img src=3D"https://www.listbox.com/images/listbox-logo-small.png" title=3D"Powered by Listbox" border=3D"0" /></a></td> </font> </td> </tr> </table> </div> </body> </html> --------------080902080408050105040305--