 |
[Design] A Sort by subject annoyanceMark Nielsen Fri, 8 Nov 2002 16:43:00 -0400
--____PTBXKLZLXMGURRIRPUMD____ Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable That's exactly my point Arnulv and thanks for defining to the others what = Ang: and Betr: stood for, the question is what are all the multilingual = 'turds' out there so we could ignore them in the sorting of subjects? As = mentioned earlier by Kaitlin (who has written some good messaging articles = I must add), time has proven that "because the RFCs are publicly available = and usable, any idiot can write a MUA -- and many idiots have." Because of = this, you can't depend on Message-Id or Referenece headers because they = never became a requirement/standard for a Mail User Agent. I've yet to see any client that can thread SMTP mail perfectly because of = this exact issue.=20 Regards, Mark Nielsen >>> Arnulv Rudlandrudland@hastenrath.de> 11/07/02 09:10PM >> Fascinating thoughts! but the given examples are, AFAIK, examples of multilingual "turds". Every language has its own: e.g. in German the Re: equivalent is mostly = AW:, Fwd: mostly WG: but also FW: and so on. I _guess_ that both examples above are "Subject:" turds, "Ang:" is Danish/Norwegian, "Betr:" German. But this only illustrates the problem: every language has its own turds, = and I had to use my own cultural background and my personal knowledge about common Danish names (hello Mark!) to guess. A software would have to know all generally used turds in all languages, with all ambiguities. That's the point. In good old times -- as the usenet and not the web was = the internet -- this worked with even the most primitve command-line reader IIRC. Since then we've had technological progress. Best Regards, Arnulv Rudland _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Open Source Applications Foundation "Design" mailing list http://lists.osafoundation.org/mailman/listinfo/design --____PTBXKLZLXMGURRIRPUMD____ Content-Type: multipart/related; boundary="____KRWBIFSFOVNOWDRPFGCM____" --____KRWBIFSFOVNOWDRPFGCM____ Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META http-equiv=3DContent-Type content=3D"text/html; charset=3Diso-8859-1"= > <META content=3D"MSHTML 5.50.4919.2200" name=3DGENERATOR></HEAD> <BODY style=3D"MARGIN-TOP: 2px; FONT: 10pt Comic Sans MS; MARGIN-LEFT: = 2px"> <DIV>That's exactly my point Arnulv and thanks for defining to the others = what=20 Ang: and Betr: stood for, the question is what are all the multilingual = 'turds'=20 out there so we could ignore them in the sorting of subjects? As = mentioned=20 earlier by Kaitlin (who has written some good messaging articles I must = add),=20 time has proven that "because the RFCs are publicly available and usable, = any=20 idiot can write a MUA -- and many idiots have." Because of this, you = can't=20 depend on Message-Id or Referenece headers because they never became a=20 requirement/standard for a Mail User Agent.</DIV> <DIV> </DIV> <DIV>I've yet to see any client that can thread SMTP mail perfectly = because of=20 this exact issue. </DIV> <DIV> </DIV> <DIV>Regards,</DIV> <DIV>Mark Nielsen<BR><BR>>>> Arnulv Rudland<A=20 href=3D"mailto:rudland@hastenrath.de> 11/07/02 09:10PM >>">rudland@hastenra= th.de>=20 11/07/02 09:10PM >></A><BR><BR>Fascinating thoughts! but the = given=20 examples are, AFAIK, examples of<BR>multilingual "turds".<BR><BR>Every = language=20 has its own: e.g. in German the Re: equivalent is mostly AW:,<BR>Fwd: = mostly WG:=20 but also FW: and so on.<BR><BR>I _guess_ that both examples above are = "Subject:"=20 turds, "Ang:" is<BR>Danish/Norwegian, "Betr:" German.<BR><BR>But this = only=20 illustrates the problem: every language has its own turds, and<BR>I had to = use=20 my own cultural background and my personal knowledge about<BR>common = Danish=20 names (hello Mark!) to guess.<BR><BR>A software would have to know all = generally=20 used turds in all languages,<BR>with all ambiguities.<BR><BR><BR>That's = the=20 point. In good old times -- as the usenet and not the web was the<BR>intern= et --=20 this worked with even the most primitve command-line reader<BR>IIRC. Since = then=20 we've had technological progress.<BR><BR>Best Regards,<BR><BR>Arnulv=20 Rudland<BR><BR>_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _<BR><BR>Open Source=20 Applications Foundation "Design" mailing list<BR><A=20 href=3D"http://lists.osafoundation.org/mailman/listinfo/design">http://list= s.osafoundation.org/mailman/listinfo/design</A><BR></DIV></BODY></HTML> --____KRWBIFSFOVNOWDRPFGCM____-- --____PTBXKLZLXMGURRIRPUMD____--
|