MIME tests one and two results - Netscape and Eudora

MIME tests one and two results - Netscape and Eudora

[Matt Armstrong]

Netscape/Eudora together capture a whole lot of the Win32 e-mail
reading population.

Screen shots of MIME tests one and two for both MUA's are at
http://www.best.com/~mattdav/mime/

-- 
matta

Re: MIME tests one and two results - Netscape and Eudora

[Matt Armstrong]

Matt Armstrong <matta@geoworks.com> writes:

> Netscape/Eudora together capture a whole lot of the Win32 e-mail
> reading population.
> 
> Screen shots of MIME tests one and two for both MUA's are at
> http://www.best.com/~mattdav/mime/

Now I've added PC-Pine to the page, which is perhaps more relevant to
Gnus since it is has a text based UI.


-- 
matta

Re: MIME tests one and two results - Netscape and Eudora

[Lars Magne Ingebrigtsen]

Matt Armstrong <matta@geoworks.com> writes:

> Screen shots of MIME tests one and two for both MUA's are at
> http://www.best.com/~mattdav/mime/

Interesting.

For the multipart/mixed, Netscape drew lines between each part, and
even centered the image/jpeg part.  For the multipart/alternative, it
didn't even indicate that it was such a message, it just displayed the 
image/jpeg part (centered).  Trés weird, in my opinion, to go so
overboard on the multipart/mixed and so minimal on the
multipart/alternative.  Hm.  Could you edit the "MIME Test One"
message to have a "Content-Disposition: inline" header in the
image/jpeg part to see if that makes any difference?

Eudora didn't attempt to display the multipart/mixed image part, but
just drew an icon and said "MimeAttachment" (sic).  Could you try with
that Content-Disposition thing with Eudora as well?  It didn't
understand multipart/alternative at all.

PC-Pine listed a "menu" after the headers, which was an OK way for a
text-based reader to do it.  The menu said what all the parts were,
and whether they were shown or not.

So, does anyone here have access to some Microsoft mail reader?  It
might be fun to see what it does.  And what does VM do?  And surely
there must be some MIME-aware unixoid mail/newsreader out there...
knews?  xrn?  trn?  slrn?

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> So, does anyone here have access to some Microsoft mail reader?  It
> might be fun to see what it does.

Outlook Express under Win98:
http://www.cs.cmu.edu/~karl/pgnus/mime-test/

"Fun," he says.  "`Fun'."  Hm, "`«fun»'".  I wonder how many levels
of derision quotes I can produce.

Comedy is not pretty.

Re: MIME tests one and two results - Netscape and Eudora

[Matt Armstrong]

Lars Magne Ingebrigtsen wrote:
> 
> Matt Armstrong <matta@geoworks.com> writes:
> 
> > Screen shots of MIME tests one and two for both MUA's are at
> > http://www.best.com/~mattdav/mime/
> 
> For the multipart/mixed, Netscape drew lines between each part, and
> even centered the image/jpeg part.  <snip>  Hm.  Could you edit the 
> "MIME Test One" message to have a "Content-Disposition: inline" 
> header in the image/jpeg part to see if that makes any difference?

It made no difference.


> Eudora didn't attempt to display the multipart/mixed image part, but
> just drew an icon and said "MimeAttachment" (sic).  Could you try with
> that Content-Disposition thing with Eudora as well?  <snip>

That'll have to wait until Monday.  I don't think Eudora can display
graphics 
though, so I doubt it'll make any difference.

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@justresearch.com> writes:

> Outlook Express under Win98:
> http://www.cs.cmu.edu/~karl/pgnus/mime-test/

Wowsie.  That was pretty pathetic, even taking into account how badly
their products usually are.  To summarize, it shows the first part of
a multipart/mixed inline, and the rest are attachments.  Even the
other text/plain part.  Quite amazing.

The multipart/altenative only showed the text/enriched part, and didn't
indicate in any way that there were alternatives.  And it didn't even
render the text/enriched part correctly.  Very interesting.

> "Fun," he says.  "`Fun'."  Hm, "`«fun»'".  I wonder how many levels
> of derision quotes I can produce.
> 
> Comedy is not pretty.

:-)

I want to create a MIME page with links to these test pages.  It might
be a nice resource for MIME developers.  Is that OK with everyone?  I
can copy the pages over to me if there's any problem with keeping
these permanently at your sites...

More test results, please.  :-)

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Netscape and Eudora

[Lars Magne Ingebrigtsen]

Matt Armstrong <matta@geoworks.com> writes:

> That'll have to wait until Monday.  I don't think Eudora can display
> graphics though, so I doubt it'll make any difference.

If it can't do graphics, then it probably will display it exactly the
same way.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> To summarize, it shows the first part of a multipart/mixed inline,
> and the rest are attachments.  Even the other text/plain part.

It occurred to me that there's one notable deficiency in Gnus'
handling of multipart/mixed in test one: If it's being displayed on an
ordinary terminal, the image/jpeg gets lost, without any indication
that Gnus was unable to display it.  This is no better than OE's
failure to note 2 parts' elision in multipart/alternative.  Example
screenshot follows.  It would be good if a buttonized indicator was
substituted for the image in the absence of a window system, so it
could be saved or displayed elsehow.

begin 644 mime-no-image.gif
M1TE&.#=A]P$$`O<```````!D`%A<6+B\N/C\^'@O`0(4`$`(`/A0$R_T`!3_
M``B_`#`%`3YU`!0"``A``.2TV#8X-A04%`@("+O0_$#S\A3__T"_OWP$W.X`
M\AD`_T``OWR$)V!<>!`*`@A`0$3X^/4O+_\4%+\("/#X,/0O/K1$W$P^-D`(
M"$0;#_6V-_\`%+\`"'P$,&``/A``%`@`"'L$-F``@1```@P`0`#X?0`O```4
M```(`'P$#>X``!D``$```'SX>F`O`!`4``@(`(2,)%ST-PK_%$"_"&$%.'-U
M\P4"_T!`OP,$0@``:@``!P``""CX-C8O@104`@@(0%P$?0(``(3X#5PO``H4
M`!BD>O7T`/__`+^_`.D%+'QU-P4"%$!`"+CP.#4V\Q04_P@(OR@D=#;T:A3_
M!PB_"%P$^`(`+P``%(2$@UQ<``H*0$!`!D3X,/4O/MCX?>PO``,4`+A$$#4^
M`!04`"@;>S:V`!0```@``(`$$O4``/\``+\``"'X#4`O``@4`)S41$?T\PS_
M_PB_OP`%-@!U<0`"!P!`"/0D'/4W7_\4$PA4`:#T`"S_`$"_```$B@``L0``
M$%B$5/5<\_\*_[]`OQ;X/>PO90,4!_#XT/4O7/1$H/4^^/\4_[\(OP@`5*``
M]BP`_P`P8P#\?``9!1<!#8(```<```@#`*```"P```$$H```^```_P``OP`$
MH`#XQ`$O]0`4_P`(OP`$`&[X```<``#U``#_``"_```%``!U`/\"`+]``"/`
MH`$V^'RD,`#T_`#_&0"_0``G`0!X```"``!``(2L;#GT-P#_%`"_"`PG,`AX
M/@`"%(3X`0`P`P`^``"\!`#T`&$G^'-X+P``````````````````````````
M````````````````````````````````````````````````````````````
M````````````````````````````````````````````````````````````
M`````````````````````"P`````]P$$`@`(_@`)"!Q(L"!!`083*ES(L*'#
MAQ`C2IQ(L:+%BQ@S:MS(L:/'CR`?(JPHH*3)DRA3JES)LJ7+ES!CRIQ)LZ;-
MFSASZMS)LZ?/GT"#"D4I<&1$HR&3*EW*M*G3IU"C2O6(5.3$J@Y'8F58<JK7
MKV##BAU+MFS6K0N['D6K4.W(`7#CRI6K=J[=NWCSZMW+MZ_?OX`#"QY,N+#A
MPX@3*U[,F''=OH_Y1M[K5J#?R8TS:][,^3```'`_$Q;=.2[ITJA3JU:,.6]K
MO*_O5B9P^>WJV[AS>P8]^'1IW[J#"U<=VV[QN8_3PIU=V_+PY]`'@/X\_31U
MWM*IFQ9=/73WZ]Z!_H?GS1T[^.SHSV>WKAW]7/7GJ\N/3O_Y<;JV)=NF+7NY
M5N?Z`5C?@*F19F!WZ9'W77G>N7=@@W(]J!V#"5ZGH(,(4KA=@Q)>Z!Z!(*)V
M7UPC^N<<?\:92,!;#0DP0(DAQMB;AQ26AQV&.#Z(8X0W[KAA>\#I^&./05ZH
MHW@R)HD8C)@-Y")^)\*F(HMZN0BCDECJ=:1\[:T'88T9*J@>A!MB.*%Y1-((
M7I'O&:EFEG`:UMJ*QN4'%W]/DKB?E"_^1QN*_5T9YZ`[:MAFF#F^B1>;"7*(
M9IN.1LHCI(WZ2.BEE-EY9YYZ"C@`GLCMV5^?15FV8DND`HKIJI,F*BF#_F=Z
MV>&;';Z*H*N5@HFKKA^RZFNGJKJFZ:F<IGHGG\Q!INFOF&XYWG;LN1KM>H]2
M^Z68UD*+)GM<7LNMHD@R>RF3R\(F:HJIEBN;NN*V&UBX[L8K;W/!FNNIL%&.
MFFR`]<[K[Z(]_BOPP"K2:_"G+$UYK[W]$BPPO`Y'S"RY"Z];<9VE-ASJQ1)W
M[/''A%*L,90C`XL03<0.I?+*++?L\LLPQRSSS"C7/%/*,.%,\\X\]^SSST`'
M+?30.JU(``!')XWTT@(A;=;34$<M]=145RV6GTQGK;357'?M]==@AUW6?UIK
MG;38:*>M]MILMXWUUF5[1!U#GT%4=T%W/^6T_D)SM^WWWX"G33;<6Y_-T=Y\
M]]U0WE\Q37?@D$<N.=1OQ^WXX70[;>'=UQW=N>>:APXZXA?M33K>DZ>N^NI.
M#6YYTW)GWG3GGUOH^>RXCPZ[0;;;SOM`IQ,4/.O$%V\\1)43?OE&P\^^N?"(
M<RZZ\TM_GI'ICQ^O_?;<'U3JZYK'OI#TC),_?=ZT-V_1\@E%C[KPW<<O/]K)
MO]Z1XGB'SKCNTNN>N__,RYSIT$=`]<WO@`B\VO>4=S8#)E`BI(O@`R=(P;#4
MCX$5S*`&-Z@VUV'0@1P,H0A'.)4+QBTB^,O?_L8'PJBH+X4DC*$,0^+!$[9P
M=XG[G^S,PK[?S?"'_D#DB`D)A\(=`F]NO:M=W]*W1!A6!'OC"Z(4ITB1&A+Q
MAH9KW_"2>+O<*1&)3CPB&)T(1;Y1\8QH3$O&P)=%A[PP@L^#WA&I1\?;60\C
M96Q?&O?(1RN:#8MOK.,=[=A$\^D0BX\S("+YR,@,#M%L=ENA&/4W2-`)$HS4
M"^,3!=C(3@;1CX5;I"='24K)/;)PI4RE*HL'2L>)<I6PC"773ME#3K)P<]4K
MY"OU>$M9^G**K6P@!"-I2/_M$G6*_*4R?TC+-BZ.F)2L(^9PR,ME6E.$P0Q?
M$1^"R]%53X!U\]W[G`F_:YIS@\T$)#2AU[_[`2][YXSG!+-)33?N\([A_OPF
M.3>93'GZ$X'IA*`DZ5@^T56R=+;\IT*[1\]]+O2A$+5(0"-*T8I6<8'V.Z9%
M-RK/B1+SEOJLG3M[R=&2FA*C'QPF"R_).<SUTZ0P_9M'N6E$;RXQ@/4<9TQW
MVD&4VE"E.9RD2'G'1$GFT8<\3:K89FK/E<[1F#AUJ%252M6H-52=3OVB-C52
MRW)6]:M58^KB!LH_@VJ2(F_4*%C7JI2KLO6MLA0K7.?:2;?2]:Z>E"M)M9A/
M08XT<6K%JV`E8E>[W5.7.5V?`P,[V,:V:(T,[*H9G>H\J')UJIAUK&8EZM,K
M`E6+[/1=\)*HR<5N]K14@2P;MYE#TK9TFO!$_JUL+U)8FF85L9DU[#-GRUN2
MJ#:R`FU>^G!W4+06]ZR]3:Y!:JO<YL9/K\Z-+O&8*]WJL@ZZS%-<`;>HW4`^
M%8`@(2M1HR=<$+;3NA6DKMS(V]<NRI&09-WN>\-;T)J.E9N,16_JL,O5`1ZR
MO%T<:#L)&!+K%=>]W<UG&$F;26GJEY6=_>-2Q&O)H)XW?U[T;U+J2]D`PW&E
M3:SL<!]L//Y>S[NWU>%[STMA/`*2O)-4\8K/9];\DKAKZKT?@(/;0GS:&,/?
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MG=MM6Y/:Y"YFAFM-9)!>>K35)K6Y%=KMPY8UCKC49U1?.&]NNUF8T`9R03D\
MR'`&.K;]_F>WXUMK<!.Y@'_5<\(7BNZ)6[PMR2;BQ3=^E7]OE>,@STK&_B&)
M7^Y:NW&Z#GE'/4YN5)^\+))MN<I_6>]#Z[+(KUUTNF>N<):76ZHK)!^B<7WK
M)S.:WSQ?N=&`&W.70_FIJI[>N(&>=*4C9+58_3/4P7WI@T>QZN=<.'>MW+]N
M_ES0$@=[K'W^:K7O.=AN/W;%XR[WD:.RY%=NL=YO2.7/1IONGV2[WR5MW^RZ
M4]N`EV'-C1ANU^HOQ@!\MJ"1GO@95GRQ-#;?N[G,^2/S]=OPWGGE+6_WILO;
MVOF>\8"EWN!K(U3THX_AY=W(^JBO?MTT/CS:8X_-TO_\K/(MZR41K7FBMWJO
MO.^]T;">?(O#O?F#GCOTH^_[%Y=7Z$]LN[-3_C[]]`H^X#-N]GW#TG3M=[_-
M2V=CU@D_Y?8&'8Z@1_+QJ7Y^#<X>G%L?_M,M;5D5QA^IE%=_%+1XX[=Z<21E
M-D5)?==4]">``_A]NF5AN9>`IT-PB>9UC>:`"21V(,52MN9_H%=V"VA;@*6!
M#[A\P'5V)GACS[>"^F57^1:#,CB#-%B#-GB#.)B#.KB#/-B#/OB#0!B$0CB$
M1%B$1GB$-)@Q(Z%^F82$-2A43GAS,4B!2-B$03AK56AI,DB%[_:#"3AK5GB`
M_X-)45B&9GB&J;=T5Y>""HB&EI1+7'B#;YAZ.!B'7>B%13B'=RB$=IB#?6B#
M-F6%?XA$;P@W:'B(_HB8A1.5B(08?X_W<*%D<#-H1\3U:8](B5@(AQ"7AT7E
M2@36B&'X;7/TB`;6B3U8=('HA\\CB8S8BJYXBFK(=+LC05Y53_"WA?`3:DK#
MBI[81FD88F8U=&-4B8W(BUJ%APU'C+S80'NHB08'7WHH?+7HB\BTBJLVC5HV
MC-A(:K18;=UHB]P8CMLXCMXHCN6X7-47BF6XBP%6=,L8C,W85\7X3:O63828
MB^PH8D0HC9LWC\Q(A=%H9`TWD#SXA?98AXXXB*^XD*^HA++(C(E8B/.H8/G8
MBP1)D?XHCQ#IC%E#.[LXC)R(D>[8D;#3ADPDD02%D2%XB@?)D5LXD67#_I`R
M.9.^LX@S.8I9UE+QB(N:&(WPU5US*$;YB'M#F(F!6%F&.(FE>)29N)-^B(T#
M](0J1H\T694W&8O,IY!6Z7AF:)!;Z81::8U?.99D698%&8NR:))F28>(&)9K
MJ8-N:8IO.9=T.98.F951$P!ZN9<N:%$>M68JM'WQ)A![R9<_-FJ'(WG3F('<
M*&"&=E,E>5!&]W7(-VR%A9B:AE]#YC2%J9<J.'E*L4MJ%7H`>'3O9%Z#YU7F
MUUM_.9B9^4SAPUX$T)D!X$PW!7&@A7V4J&_`=YK^18KM2&"CV&4FASXXF5FR
MB93'=9KMYVB7.8)8Y3[E1)L.%5*N=G=A%IRN_FDX.CF+OAF9.`1%B+E%J(0]
M'2E>9=1#E&>=F,F"Z5B2J>ENKDF=%7B/[.1GMF:<D5E:IME^SXB=U%0^K\F=
M_0F@U11J[4F@Y@F8+PB!'[=[$4B-L]F9$FJ@B66>S*F:R&F:WZB=I?D^TOFA
M!`J?#"A<WEF:&%IJK2ES2+E-"#JAA2EN'AE:^,E2X=FB?R>>N&F=@8FCN/FA
MD/F1\76?X,5P6W6)<29]3,&9%#H5"?IZGRFB?0D5+=B5M%:79CFER(:"&:6E
ML-:"7NI<2AJF8OJ>9/IV$'BF<`:F:LI;8]JF;FJF<.IE;SJGJ,6F=KI9=9JG
MFH6G?-I8>_JG@^6G_H**5X%:J'=%J(@Z5X>ZJ'"EJ([*5HT:J8\FIY2J7)-Z
MJ:9FJ9H:IUR:4IV*J9P:JG>:IJ0J6Y!ZJCN5J:H*4ZG:JB;%JK#*4:\ZJ]-F
MJK8ZJ*.:JXF*J[S:J^GWD+^JJY_Z4\-*K$LHK,<*K&O8I<MJJ+OZK&`EJ]+J
M3[5:K5;WD*N)K><6K=R:5-3ZK6L7K*LEKM/JJ^;*4]>:KLH4KNSJ2^OZKG&%
MKO):4O%:KZODKOBJ2O>ZKZ6DK_Y*2OT:L'E%KP0+40-[L(P$L`J[L-[:L#U7
MK)X%L7[YL!0;3PQ[L6>4L!H+3`;;L>;$L2`+1!D[LB1KL29+<Q^;LBI+_JXI
MR++9BI<P&W8H.[/Y^GU8FK,ZN[,\V[,^.XEHR81QR8CJJ(X@"8M-"8=0V75&
M"X1)VY:8U+0769`M"8CA-K0_^Y5WR8922Y,CN9)BB8PJR8Y3:(Q.Z8,!2;09
M:;5\B&]RV'5<F;4\ZY!IV;6<6(G!R6XI^8N3!I'[&;?G&8ED*(I@V8F%6)%&
M*9#X]K?VV&!MZY%OJY)3*[=8NK7.VJ&8"X*_:)%)^;7-V+?`^)-<"8E&!HUR
M^;A`"8J6,Z/ON(I)N3\UIF_@N(TM&8Y'JXWGF+NSJ[L5NKN^V[O`F[GDZ#TN
MFU%J^XZ=JY1LR9%!R85?JX<)R6Z%V[BJJSSN_L-%7GF>RPNX<%FU0"N4W$NY
M=6FYH-J*$BF)'^FY(PFZ(IF$IDBVK2ND9\N2[1N;3&>VL2F6Z>N]8+N#V>N^
MDMN_XDN7="NTYIN1S=N40JF<4Y;`4SB5HMBB5F:W5+NC^YO`\BN7#LR_6"N&
MO@C`O+E=`YRSY&NL/1N'0PN](PRU`KS"+NS"!:RM+^R%%#S#?%C#-IS#)(R5
M7&NS-%N\=#:9405;VC>"I^=TBSE-BHEJ^/-AV\F84%JF$BMA+,J`2)QK5;Q)
M\Q?%$9=?F/F@8+&MKFJF$K2CE[-@.=J<N7F)Z/NC!]K`P*F3&J:,/WJ;`9A%
M=KS`".9YMFBBLYB<_K_CQF)LKPX:8N6I<7ZL1]V92-2H38M<G5`'<`L*GB,*
M<'@\3CSJ=/#VR!IZQI#G:>E9H"M6R5'ZJ&1LCLJ&</SS=T!JR`IZ8*'LGYY,
MFHH&?XZ[G=\XHT1ZHP(:;XD<R^24HJ2<7)>)RHA,F;R,S!R:H<M<HLS<QR2*
MR3Y4QM*,S*'\RX,9HKDK=<!\S8\I72OJRSF'I/RWRC3ZQL1WHV#8HSH:E91<
MG^Z\=1;<<FQLG]*48'ILRQ+LB?%<8>V<Q8[%JH,<@6WWI`0=/QHUT(Q:LVEC
MQ,ZVQ)5YL27KPR$T4V@\H!B-SD&Z>5<:H6E'T9PUQ=DYS`#-F-BLSA9Z_L10
M+,H@W7%`K&S=2)+?A<^1)V"\>9RHF<_J)LTAK-#F6EL;W8[?',U@;**>#)],
M6Z/1_,23?,@MG1!R-<N9IH"V"53=O*&MIUWP-,L,:K/%3*+N5TXQ?:*0C-+"
MO-5?!\8*^LQ/#=5RFKJ(RYV"W,_5'-1"'*`[;:3J[-`I^Z9J#7YMK60OK7'9
M%]@G*])O9MCS.MB0=->L97V*K3K,A7@K#9J1?5*,C9V`7%J]2,F[>60R_<\E
M.)3_>=FIA=B6;-:X/(:1[,@3^\CE]YTC;=JGG:SEJMK.C-O&*9QK_<?DN=3B
M1]NT9:I7/5GZV=N\FZ*P6Z#LX]-MS52;S7!^_CNB-.W/CZF].EW4$"W<7+&R
M*HU6V?S$$$J9SNW#(OO1/>H4?,W=W8W:?\W>:13.8P?8$7W%_?7;+B1YZP=;
MZZS>U.?>N%W-5@S*XEW0X>U"XLW%_<7235'>G/;6^P3/[US=RCR<UNW;X!W(
M\8RA;NS1>+O'XUS3=,W(*)W>U9W'>,O9OXFCT`K@%?JBKTS6!?[;RWW6*&1R
MA]S4IL=^2'UT]BOC-AZA%]W,.?[CI0V;9.W@+6O;+YO2OFF?[UW9U'V/NCG@
M/,W,ZOEL_';4R/UQ,'[09G9S+][E+'[.QZWD\.J@!Z[90]V>_?2?!H7)^%VC
MRBU9<_S=QYS,0FWG_D,FX,DTUL+<V6O>W`G^X)G-Y?/-P.$YXD+56J!6I/+9
MRBM^BTG<>G4=XIBHQO@4Z0MVXO/\CTE-:_N,YFGNXLYM?E\,-:2NQ9-CT`M]
MZ*4L->O=7+.^J:8.WZ-TWKA^/!.]Z[S.T+XN>]X=[(H'[,2N?,T*JL>.1KJ^
M[%]VZ\Z^L<8>[8XT[-2.3F2LRY4^8?1=TL=D6A5.%MHL2B^6WXOY[95-[B0T
MV86NZN.MRO&Y[;&N-_(><?[=[H!]F.MNJ4U=RWI-VNAYI,K9Z76L;D,NT_Z,
MQA8LQPO<TTH=?+)KXBL^E'C\[\PF\(O.VPY?D2D4YZR-X1]>T<0=R27^_L:P
MG3U!KFTI7^\Z#J+W5(WO'.-.7LGCKM:D&9,!3ME!/-4`^*`\6H',F4>)'8G[
M#NMC7>('_\J].>5/REZZJ<^![-1#]?!1RYZZ3,Y!S_,,[J-9C^GB)NI1C^CG
M3O+:[KN-+>=07O3);JS,O>:=S-9I/94R+M9E[?.*G.-M;O)!G[^[VZ$]'J!$
M;=F1%:(OM<E$#_@T+]0`^N-9#_AVO\?8;O1^/O#_+MK3/)P^QGAY_8%1S_1W
MKJ'F+,X"#\MR#6,D+>)ZOLLJ?O,ISF$73J.C]?C3;?G/6.O79>U.6M+&X_=L
ML^H9;C6^'U;3;NX:Q/MK$_SQ3JNX?^W>)_G,_A]XT/[\S#3\TG]`O5[]Z,?D
M3(;]:J^MW%_LSO_](W3]XM]3X5_^'$3^Z+]4U+_^);;\[J\]S1[_7Z/^](]C
M[7__D@W_^G_[Y]__`$%`X$""!0T>1)A0X4*&#0T*$`@1`(&)%2E>Q.A0XT:.
M'3U^!!E2Y$B2)4V>1)E2Y4J6#"$2>&E1YL6)+6W>Q)E3YTZ>/7W^!)KPI42:
M16O6#)I4Z5*F39T^A;IRJ,"95:->Q9I5ZU:N73=.M6HQHU>R9<V>19O6XU2C
M;9&JA1M7[ERZ0<&ZQ?BV[EZ^??W^%1J1*EZQ@`T?1IRXZ]VP@PGJ51Q9\F3*
M)]DV+DP1`.3*G3U__@8-4S!FQP,YAT:=6O7>RX3'KH8=6[9:QGA?S\:=6_?3
MUJ1W_P8>G&=MJ[>%'T>>_&-OU\J=/X<>6+3OT]&M7__-O#%V[MUU$V]KW/MX
M\IVUVRZ?7KUD\#/%KX<??^[YXO+MWX?;ONA[_/W]1Z4OO/\&))"WT9JKKD`%
M%\0I0/<8A#!"F_03*T$)+\2P(P?WR[!##SFB,*\/1R11H0TS*S%%$D,\2D47
M5Q3,-11?I#%"%DNK,4<&3QQK,PLU^C'!'UDZ:C,=C\SIQHR0&I*A(!5J4J7"
MHD2R2@UC)(U)@WQT2,N*F#12LS"WU,RT,+\<;$PN:0(3S!ZMA#,E)=]:_E-+
M)TTK\[46$SJS(#393!///KVDRLXX#R6)1\ZF;*B^(OOD$T\Q_Z1T4D9]--)0
M*A'E]*`Y)>70S8,<7=*Q105%==`]046QR$Y?%4G1Q]0<D[\*)<743(3LS%4O
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M5!PM]')53:G6U7+%(9U<<%X=7Q38-*]FM=3,`RV34-(;]3QU_F;?4>7F+D<U
M4)G@ECAT*/LV7>^J8^>UR9Y/E]A/+D='7'5<9TW><><5-W7P3)MG%&U+8<^>
MU:%'79RH:K$7/54.R10Q<1P%13U]4%_?_.[.P;<<\]S3C__ZZL\??OJMZ5X?
MM/2G)^-\3W?O&];Q_A#8H9K)3WO%&]NLA%<X]$%/@I.:F_JB-ZRU:="`>=,4
MX5I'P3QICWX3#%*MCM8^_=6I=RK\'P6#A[?U>8AC4'&:2)A'HAPNI(?]`1D.
ME;+!F<7GAD5$HD:"N*6BK5!Z):1=\WKR0_7Y[(-/\R'UDOB0EX5G4S^DW/Z>
M$D8J<G!J'W%:&0NVQ`H^L(0O#%;E`KBW.$Z03&*+E-F8V#<[*B]/<-Q='RMX
MMA8&CI!'8Q,>"W<VBAVQC>QC'?FZ![<:RHZ#I8MDZ8A(QT%EL57X`V`EW19*
MS?W1E!I\'[3BID9LL5&*O=HC_!1X&SI6D8SWTZ,>E3=`0<J.=YN#6`P?_J>W
M`<ZH1[-\("Q%N<8NNN>+J:NE&5/YR.)9,GZQ>QTU[_?(Q+D.FK#KIA1-R31:
M!I"`RR28*^VH3&*N4Y%Z>F<A13A#$EKME[S$X`DCZ,;'^:V4)@36W]R)R'SR
M,XK\<J16FDB7,(ZDH49;X!8)HLX1$?%F5Q0:1B6:4(EV=**VVXY'14JS9NZ'
ME2-M)$C1@U*6?I1QXVMI3&,4DWW)-*8<M:G,*)I3EN&4IRC;Z4]%YE.AABRH
M1>T849&JL:,N-:4O+9E3-ZK2^DBUB$JU*L.:FE6$896KS+18EKXZ,Z^.%6!;
M-6N_RII6?*&5K?=:ZUOGY5:YPBNN=6477?&Z_JZ[[O5<>O4KN?H:6'`!EK#?
M&NQAMV58Q;:RI+=J;,(8&]EH)9:RS)KL98ME6<T2*[.=?15G0=NISXX64:(U
M[:%*FUHXH9:U5EKM:Y'D6MD>*;:US1%M<5NCV^[V1;KUK8MZ&]P4`9>X2J.J
M@(Z+V,>Z;[G<&NYS;=C<]DFWLLE]D'6O95SM8BBZW940=\%K(^SR;[R>I2XZ
MSZO:\AISO9P2[WL+]%WY$BB^]?T/??'KG_ON%S_Z]>]]^AM@^0"8P$9,[T$/
M7-SV*GC!X:OI21_,WP9+>,+_3;"%+VP?`V]X/`/VL'<Z'&+N@)C$V!GQB:UC
M8A5')\4M?@Z+8>R<_A?/.#DRMC%R:IQCX>"8Q\'9\8^SDV$A?RC(1<Z-CY&<
MY`HO.4-*=K)LCAQEV$"9RN%J\I4A9&4MHV;*70[-L2(V9C*7V<QG1G.:U;QF
M-K?9S6^&<YSE/&<ZU]G.=\9SGO6LYYD>$&9[YK.E@BGH,:,MSH9VT_)V2<FT
MF9G0E`1TF3'HZ(@]NFZ%QO/VGKCF?-XSTI\&-:C[3)U&AUK.VPMFFBW-9D1_
MR=6)I#2G8^WJ4$_:TJLF,ZYQS6I)IUK5M$:UJ84][#H+1@`TI0ZQ"PDXY`$;
MA))>I3R;;;KE@6[7S@;H"GWM3+'1>=+7<Q4*7YBI<9/;V[G>]IE;'6QE_K?;
MW6B."$3$%]5S(E"5I2[T"$>X.UKW^]9U2R:@T)U(0Y]OU@6O],%=]6H[?_MK
MK=OVNJL-[*'=FX#Y]K4X&6YN<VO<XYO^N+U#;DZ0EUSD)G<ILA'4;H2#F^#J
M[G7')_Y'=,M<YM<VH1PGCFEPX[/AC(8UQVT><>CU^]`Q-SJTA9?N=S?=Z341
M,\L5/?67]\KJ,P<XUG6-]9O_6NA6AWC$&7YM-TM<T&&7^-"O?NZKA]WK&]_Y
MT^6N[#Y?#-^UGGK/#SE/11X2T<1LN[`47?1TSZ_3N18=N0'/]I8/>MP#7?J=
M_2@W7BM>AG/'_+!'W3BR9W[M/']SJST_>M*7_M[TIV>YQ51>+=0W??&MAWWL
M93][V6\>IF">;Y8=3+ZL(#-GN+\25&$64<Z9Y:0:AK&V?A7/AY_.@4PTRC\M
MFNUVUM.>@W.^]:G\,-[G3DV=@RSZP@TYPE]2>@^E7C''2?PE*Y_W[+QFN'?V
MS7[V;YJ`]X@W+P7T[1.9Y/73.?K;FG""H@*R)!$AI_];.%S:/1Y3OGNJ/FE+
MH"NBIWV;OL>;'PS,MD1;I&!!OOJ*KUM2KYVPN`\4,OVZ0(T"OK[@LA5,%]US
M02"*MV.CP1JTP1O$P1S4P1WDP1[TP1\$PB`4PB$DPB(TPB-$PB14PB5DPB9T
2PB>$PAI,F1B$CI?@`XB````[
`
end

Re: MIME tests one and two results - Outlook Express

[Wes Hardaker]

>>>>> On 22 Nov 1998 08:26:00 +0100, Lars Magne Ingebrigtsen <larsi@gnus.org> said:

Lars> The multipart/altenative only showed the text/enriched part, and
Lars> didn't indicate in any way that there were alternatives.  And it
Lars> didn't even render the text/enriched part correctly.  Very
Lars> interesting.

I think many mailers treat the alternative as "their choice" type of
actions, not "users choice".

IE, since the contents are all alternatives of each other, the mail
reader can pick the one it can display the prettiest and use it.

I didn't say its right, this is just what I suspect...

-- 
"Ninjas aren't dangerous.  They're more afraid of you than you are of them."

Re: MIME tests one and two results - Outlook Express

[Maciej Matysiak]

Karl Kleinpaste <karl@justresearch.com> writes:

> It  occurred to  me  that  there's one  notable  deficiency in  Gnus'
> handling of multipart/mixed  in test one: If it's  being displayed on
> an  ordinary   terminal,  the  image/jpeg  gets   lost,  without  any
> indication that  Gnus was  unable to display  it.  This is  no better
> than OE's failure to  note 2 parts' elision in multipart/alternative.
> Example  screenshot  follows.   It  would  be good  if  a  buttonized
> indicator was  substituted for the image  in the absence  of a window
> system, so it could be saved or displayed elsehow.

would it be possible to display image on console with zgv? right now, i
have to recover them from /tmp/emm*.

m.nia.m.
-- 
 use gnus not guns!

Re: MIME tests one and two results - Outlook Express

[Karl Kleinpaste]

Maciej Matysiak <phoner@blah.pl> writes:
> would it be possible to display image on console with zgv?

Only if there is a $DISPLAY available.  But there are those of us who
expressly don't want things like xv invoked when we're wire-wise a
long (slow) way from where Gnus executes, and that's why we use xterm
and ssh rather displaying under X directly in the first place.

Some indicator of the non-displayed images (or, generally, unrendered
parts) must be visible.

Re: MIME tests one and two results - Outlook Express

[Aaron M. Ucko]

Wes Hardaker <wjhardaker@ucdavis.edu> writes:

> >>>>> On 22 Nov 1998 08:26:00 +0100, Lars Magne Ingebrigtsen <larsi@gnus.org> said:
> 
> Lars> The multipart/altenative only showed the text/enriched part, and
> Lars> didn't indicate in any way that there were alternatives.  And it
> Lars> didn't even render the text/enriched part correctly.  Very
> Lars> interesting.
> 
> I think many mailers treat the alternative as "their choice" type of
> actions, not "users choice".

That's what emime does, and what section 5.1.4 of RFC 2046 suggests:

   Systems should recognize that the content of the various parts are
   interchangeable.  Systems should choose the "best" type based on the
   local environment and references, in some cases even through user
   interaction.  As with "multipart/mixed", the order of body parts is
   significant.  In this case, the alternatives appear in an order of
   increasing faithfulness to the original content.  In general, the
   best choice is the LAST part of a type supported by the recipient
   system's local environment.

(Emime is Marc Horowitz's MIME-parsing code for Gnus; the version I
use is available at <http://web.mit.edu/amu/elisp/emime.el>.)

-- 
Aaron M. Ucko, KB1CJC <amu@mit.edu> (finger amu@monk.mit.edu)

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@justresearch.com> writes:

> It would be good if a buttonized indicator was substituted for the
> image in the absence of a window system, so it could be saved or
> displayed elsehow.

Yup.  Fix in Pterodactyl Gnus v0.54.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

amu@mit.edu (Aaron M. Ucko) writes:

> That's what emime does, and what section 5.1.4 of RFC 2046 suggests:
> 
>    Systems should recognize that the content of the various parts are
>    interchangeable.  Systems should choose the "best" type based on the
>    local environment and references, in some cases even through user
>    interaction.  As with "multipart/mixed", the order of body parts is
>    significant.  In this case, the alternatives appear in an order of
>    increasing faithfulness to the original content.  In general, the
>    best choice is the LAST part of a type supported by the recipient
>    system's local environment.

Right.  Then I think that Gnus should get rid of the buttons for the
multipart/alternative as well.  I'll default to that in 0.54.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Robert Bihlmeyer]

Hi,

>>>>> On 23 Nov 1998 09:36:59 -0800
>>>>> Wes Hardaker <wjhardaker@ucdavis.edu> said:

 Wes> IE, since the contents are all alternatives of each other, the
 Wes> mail reader can pick the one it can display the prettiest and
 Wes> use it.

Only that OE doesn't seem to be able to render text/enriched corectly.

	Robbe

-- 
Robert Bihlmeyer	reads: Deutsch, English, MIME, Latin-1, NO SPAM!
<robbe@orcus.priv.at>	<http://stud2.tuwien.ac.at/~e9426626/sig.html>

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Right.  Then I think that Gnus should get rid of the buttons for the
> multipart/alternative as well.  I'll default to that in 0.54.

FWIW, I think the user's choice that Gnus does is much better.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Union break!

Re: MIME tests one and two results - Outlook Express

[Robert Bihlmeyer]

Hi,

>>>>> On 24 Nov 1998 12:32:36 +0100
>>>>> Hrvoje Niksic <hniksic@srce.hr> said:

 Hrvoje> Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
 >> Right. Then I think that Gnus should get rid of the buttons for
 >> the multipart/alternative as well. I'll default to that in 0.54.

 Hrvoje> FWIW, I think the user's choice that Gnus does is much
 Hrvoje> better.

Yes. Please don't let us fall too far into the users-are-dumb-pit.

	Robbe

-- 
Robert Bihlmeyer	reads: Deutsch, English, MIME, Latin-1, NO SPAM!
<robbe@orcus.priv.at>	<http://stud2.tuwien.ac.at/~e9426626/sig.html>

Re: MIME tests one and two results - Outlook Express

[Per Abrahamsen]

Karl Kleinpaste <karl@justresearch.com> writes:

> Maciej Matysiak <phoner@blah.pl> writes:
> > would it be possible to display image on console with zgv?
> 
> Only if there is a $DISPLAY available.  

Sure?  zgv is a Linux image viewer using VGAlib directly on the
console.  No X11 involved.

Re: MIME tests one and two results - Outlook Express

[Karl Kleinpaste]

Per Abrahamsen <abraham@dina.kvl.dk> writes:
> Sure?  zgv is a Linux image viewer using VGAlib directly on the
> console.  No X11 involved.

Not the point.  The point is that, when I'm using XEmacs in anything
like a terminal, it's because I'm on a slow PPP link and I don't want
piggish 800x600 images displayed over that link, no matter what.  I'm
using a dumb terminal for a reason, expressly to limit bandwidth use.

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Robert Bihlmeyer <e9426626@stud2.tuwien.ac.at> writes:

>  Hrvoje> FWIW, I think the user's choice that Gnus does is much
>  Hrvoje> better.
> 
> Yes. Please don't let us fall too far into the users-are-dumb-pit.

I don't think we're in any danger of doing that.  I do think falling
into the "let's make reading pleasurable" pit is worth stumbling into, 
though.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[William M. Perry]

Karl Kleinpaste <karl@justresearch.com> writes:

> Per Abrahamsen <abraham@dina.kvl.dk> writes:
> > Sure?  zgv is a Linux image viewer using VGAlib directly on the
> > console.  No X11 involved.
> 
> Not the point.  The point is that, when I'm using XEmacs in anything
> like a terminal, it's because I'm on a slow PPP link and I don't want
> piggish 800x600 images displayed over that link, no matter what.  I'm
> using a dumb terminal for a reason, expressly to limit bandwidth use.

  This only works on the console of the linux box, it wouldn't work over
the network like you seem to be running it.

  So the viewer for MM would need to be careful and not start up zgv unless
TERM=linux and you were actually on one of the real virtual consoles for
that box.  Would using `tty' and checking if it was in /dev/tty[0-9] be
sufficient?

-Bill P.

Re: MIME tests one and two results - Outlook Express

[Karl Kleinpaste]

wmperry@aventail.com (William M. Perry) writes:
> So the viewer for MM would need to be careful and not start up zgv unless
> TERM=linux and you were actually on one of the real virtual consoles for
> that box.  Would using `tty' and checking if it was in /dev/tty[0-9] be
> sufficient?

I suppose, but gosh, that seems like an awfully Linux-specific (one
might even say -peculiar) sort of solution.  Does this generalize to
any other platforms?

Considering that I never use ttyN for anything of substance -- I
certainly don't run Gnus on ttyN -- it won't affect me one iota, so I
suppose I don't care if it's done.

Re: MIME tests one and two results - Outlook Express

[Robert Bihlmeyer]

Hi,

>>>>> On 24 Nov 1998 09:31:53 -0500
>>>>> wmperry@aventail.com (William M. Perry) said:

 Bill> So the viewer for MM would need to be careful and not start up
 Bill> zgv unless TERM=linux and you were actually on one of the real
 Bill> virtual consoles for that box. Would using `tty' and checking
 Bill> if it was in /dev/tty[0-9] be sufficient?

Hmm, IMHO zgv should just terminate if controlling_tty != some_console

	Robbe

-- 
Robert Bihlmeyer	reads: Deutsch, English, MIME, Latin-1, NO SPAM!
<robbe@orcus.priv.at>	<http://stud2.tuwien.ac.at/~e9426626/sig.html>

Re: MIME tests one and two results - Outlook Express

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> Right.  Then I think that Gnus should get rid of the buttons for the
> multipart/alternative as well.  I'll default to that in 0.54.

I've looked at the ChangeLog and seen the reference to new default
alternative choice in mm-decode.el.  I'd really, really, really prefer
to retain the buttonized alternatives' visibility, as they had been up
to this point, but I don't see, or perhaps just don't understand, how
to use mm-alternative-precedence and mm-preferred-alternative to
resurrect that previous behavior.  At the moment, last week's "MIME
Test Two" displays in 0.54 as badly (IMO) as in Outlook Express,
showing only the image/jpeg without any indication that alternatives
are even available.  This is wrong, I am convinced.

Default to a particular preference, sure; even default to not showing
a selector for other alternatives; but please don't deny the ability
even to learn that alternatives are available, or to choose those
alternatives on the fly as we examine a given message, for those of us
who want that ability.

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

Karl Kleinpaste <karl@justresearch.com> writes:

> I've looked at the ChangeLog and seen the reference to new default
> alternative choice in mm-decode.el.  I'd really, really, really
> prefer to retain the buttonized alternatives' visibility, as they
> had been up to this point, but I don't see, or perhaps just don't
> understand, how to use mm-alternative-precedence and
> mm-preferred-alternative to resurrect that previous behavior.  At
> the moment, last week's "MIME Test Two" displays in 0.54 as badly
> (IMO) as in Outlook Express, showing only the image/jpeg without any
> indication that alternatives are even available.  This is wrong, I
> am convinced.

Well said.  A friend of mine, who has some experience with using and
implementing MIME, has been pleasantly surprised when I showed him the
buttonized approach to multipart/alternative; he liked it more than
anything else he's seen so far.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Bumper sticker on the stealth bomber:
"IF YOU CAN READ THIS, THEN WE WASTED 50 BILLION BUCKS."

Re: MIME tests one and two results - Outlook Express

[François Pinard]

Karl Kleinpaste <karl@justresearch.com> writes:

> Default to a particular preference, sure; even default to not showing
> a selector for other alternatives; but please don't deny the ability
> even to learn that alternatives are available [...]

This is close to debugging, I think.  If the alternative leads to a proper
choice, there is not that much need to debug MIME.  Of course, the user
wants some way to adequately describe his/her environment (slow PPP lines,
dumb terminals, and such).

> or to choose those alternatives on the fly as we examine a given message,
> for those of us who want that ability.

Yes, Gnus should indeed offer some MIME debugging features.  Yet, by default,
this should not become any encumbrance for the real thing.

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

François Pinard <pinard@iro.umontreal.ca> writes:

> Karl Kleinpaste <karl@justresearch.com> writes:
> 
> > Default to a particular preference, sure; even default to not showing
> > a selector for other alternatives; but please don't deny the ability
> > even to learn that alternatives are available [...]
> 
> This is close to debugging, I think.

The trouble is that you can almost always say that, with the sentence
seemingly making sense.  But I believe that in this case we are
talking about more than mere difference in taste.  Not only do I want
to be able to choose between the alternatives, but I want to *know*
that what I'm getting is an alternative.

> > or to choose those alternatives on the fly as we examine a given
> > message, for those of us who want that ability.
> 
> Yes, Gnus should indeed offer some MIME debugging features.  Yet, by
> default, this should not become any encumbrance for the real thing.

Labeling ability to choose between the alternative parts as
``debugging'' to make it less appealing to our current implementor is
a dirty word trick.  Even rfc2046 allows the option of
user-interaction when selecting the alternative:

5.1.4.  Alternative Subtype

   The "multipart/alternative" type is syntactically identical to
   "multipart/mixed", but the semantics are different.  In particular,
   each of the body parts is an "alternative" version of the same
   information.

   Systems should recognize that the content of the various parts are
   interchangeable.  Systems should choose the "best" type based on the
   local environment and references, in some cases even through user
                                                                ^^^^
   interaction.  As with "multipart/mixed", the order of body parts is
   ^^^^^^^^^^^
   significant.  In this case, the alternatives appear in an order of
   increasing faithfulness to the original content.  In general, the

The "user-interaction" possibility can, of course, be understood
differently, but I think that what Gnus did up to recently was close
to the spirit of text -- choose what it thinks is best, and present
the rest of the options to the user.

If I receive HTML junk, and must press `K b' in order to check whether
there are perhaps some alternatives to the junk (because HTML looks
ugly in W3[1]), then something must be wrong.


[1]
With apologies to Bill Perry, who really does his best to make it not
suck.  IMHO current Emacs redisplay is simply not up to that task,
especially with tables in play.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
You have an unusual magnetic personality.  Don't walk too close to
metal objects which are not fastened down.

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@justresearch.com> writes:

> At the moment, last week's "MIME Test Two" displays in 0.54 as badly
> (IMO) as in Outlook Express, showing only the image/jpeg without any
> indication that alternatives are even available.  This is wrong, I
> am convinced.

I'm a bit uncomfortable with the, er, lack of control given by the new 
buttonless MIME interface, but I think it's healthy to just let go and 
let things happen.  MIME messages now look like ordinary messages,
which is kinda boring -- which is how it should be, so that I can
concentrate on reading messages instead of studying their structure.

> Default to a particular preference, sure; even default to not showing
> a selector for other alternatives; but please don't deny the ability
> even to learn that alternatives are available, or to choose those
> alternatives on the fly as we examine a given message, for those of us
> who want that ability.

This is Gnus; of course we can have both jam and hunny.  (Hold the
bread.)  `gnus-unbuttonized-mime-types' is the variable to fiddle
with. 

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@srce.hr> writes:

> Not only do I want to be able to choose between the alternatives,
> but I want to *know* that what I'm getting is an alternative.

We could put something in the mode line...

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> > Default to a particular preference, sure; even default to not
> > showing a selector for other alternatives; but please don't deny
> > the ability even to learn that alternatives are available, or to
> > choose those alternatives on the fly as we examine a given
> > message, for those of us who want that ability.
> 
> This is Gnus; of course we can have both jam and hunny.  (Hold the
> bread.)  `gnus-unbuttonized-mime-types' is the variable to fiddle
> with.

Yes, but it would be nicer if Gnus had sane defaults.  When
introducing a new victim to Gnus, I hate twiddling with their
variables to avoid them coming to me and asking questions.

An example of bad defaults that was recently corrected was the
infamous default value of `gnus-read-active-file'.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Idle RAM is the Devil's playground.

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Hrvoje Niksic <hniksic@srce.hr> writes:
> 
> > Not only do I want to be able to choose between the alternatives,
> > but I want to *know* that what I'm getting is an alternative.
> 
> We could put something in the mode line...

Maybe, but it is quite possible for the alternative multipart to be
only one part in a multipart structure.  The button approach solved
this quite well.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
- Now what did we learn from this?
- I learned what my liver looks like!

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@srce.hr> writes:

> An example of bad defaults that was recently corrected was the
> infamous default value of `gnus-read-active-file'.

Well, it was bad, but it worked all the time.  With some obscure
newsservers, the `some' default now doesn't work at all.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@srce.hr> writes:

> Maybe, but it is quite possible for the alternative multipart to be
> only one part in a multipart structure. 

That's true.  Hm.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Robert Bihlmeyer]

Hi,

>>>>> On 24 Nov 1998 20:53:03 +0100
>>>>> Hrvoje Niksic <hniksic@srce.hr> said:

 Hrvoje> The "user-interaction" possibility can, of course, be
 Hrvoje> understood differently, but I think that what Gnus did up to
 Hrvoje> recently was close to the spirit of text -- choose what it
 Hrvoje> thinks is best, and present the rest of the options to the
 Hrvoje> user.

An example: I want to send a scan of a Kafka manuscript (as image/gif) to
someone. I'm also nice and include a text/plain transcription. The
recipient's MUA should IMHO display the gif, if it is able to, but
inform her that a text/plain alternative is availiable, permitting her 
to view it, too. Cut'n'paste is infinitely easier from text then from
images.

	Robbe

-- 
Robert Bihlmeyer	reads: Deutsch, English, MIME, Latin-1, NO SPAM!
<robbe@orcus.priv.at>	<http://stud2.tuwien.ac.at/~e9426626/sig.html>

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Robert Bihlmeyer <e9426626@stud2.tuwien.ac.at> writes:

> An example: I want to send a scan of a Kafka manuscript (as image/gif) to
> someone. I'm also nice and include a text/plain transcription. The
> recipient's MUA should IMHO display the gif, if it is able to, but
> inform her that a text/plain alternative is availiable, permitting her 
> to view it, too. Cut'n'paste is infinitely easier from text then from
> images.

This is actually a rather classic problem -- on the one hand, one
wants to be given as much information as possible.  On the other hand, 
too much irrelevant information leads to problems with grasping what
you're presented with.

Take this message, for instance -- what information has been hidden,
and what is shown?  I'd guess that most people would not want to see
all the Received headers, and the Message-ID header is probably not
very interesting.  But how about the From header?  It's probably
displayed in the summary lines.  And the Subject?  The same.  How
about the Date?  Is that interesting?  The X-Newsreader line?  The
Organization?  When was the last time you actually found that header
useful?

Gnus defaults to showing things only that we think is pertinent to
a fulfilling reading experience.  This is not a science, and we find
these things out by using them for a while and seeing how comfortable
we are with them.

(The other important thing is that it must be easy to have Gnus
present you with the full information when necessary.  Sometimes I
want to read Received lines, so I use `C-u g' or `t'.  Perhaps we
should bind `M-t' to the same as `K b'?)

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> Perhaps we should bind `M-t' to the same as `K b'?)

It seems to me that, once upon a time, `M-t' was a way of toggling
automatic MIME viewing with metamail or TM.  If `M-t' could implement
a toggle of buttonizedness, it would actually be a good and useful
keybinding to have available again.

Re: MIME tests one and two results - Outlook Express

[Dale Hagglund]

> (The other important thing is that it must be easy to have Gnus
> present you with the full information when necessary.  Sometimes I
> want to read Received lines, so I use `C-u g' or `t'.  Perhaps we
> should bind `M-t' to the same as `K b'?)

The user has *got* to have a clue the information is there to begin
with, before he can know to use `M-t' or `K b' or whatever.  If
there's no visible clue, Robert's hypothetical friend has no reason to
guess that Robert has been as nice as to send two forms of the Kafka
manuscript.

Dale.

Re: MIME tests one and two results - Outlook Express

[Kai.Grossjohann]

Dale Hagglund <rdh@best.com> writes:

  > The user has *got* to have a clue the information is there to begin
  > with, before he can know to use `M-t' or `K b' or whatever. 

Put `Mime' in the mode line of the summary of article.

kai
-- 
Life is hard and then you die.

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

Kai.Grossjohann@CS.Uni-Dortmund.DE writes:

> Dale Hagglund <rdh@best.com> writes:
> 
>   > The user has *got* to have a clue the information is there to begin
>   > with, before he can know to use `M-t' or `K b' or whatever. 
> 
> Put `Mime' in the mode line of the summary of article.

Forcing me to press `M-t' on each message with `Mime' in the mode line 
is not particularly nice.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Depression is merely anger without enthusiasm.

Re: MIME tests one and two results - Outlook Express

[Kai.Grossjohann]

Hrvoje Niksic <hniksic@srce.hr> writes:

  > Forcing me to press `M-t' on each message with `Mime' in the mode line 
  > is not particularly nice.

You can also set gnus-unbuttonized-mime-types, no?

kai
-- 
Life is hard and then you die.

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

Kai.Grossjohann@CS.Uni-Dortmund.DE writes:

> Hrvoje Niksic <hniksic@srce.hr> writes:
> 
>   > Forcing me to press `M-t' on each message with `Mime' in the
>   > mode line is not particularly nice.
> 
> You can also set gnus-unbuttonized-mime-types, no?

Of course I can do that, but I want the defaults to work right.
That's what it's all about; I know that *I* can make Gnus do (almost)
anything I want.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Speak softly and carry a +6 two-handed sword.

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@justresearch.com> writes:

> If `M-t' could implement a toggle of buttonizedness, it would
> actually be a good and useful keybinding to have available again.

Ok; I've added this to 0.55.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Dale Hagglund <rdh@best.com> writes:

> The user has *got* to have a clue the information is there to begin
> with, before he can know to use `M-t' or `K b' or whatever.  If
> there's no visible clue, Robert's hypothetical friend has no reason to
> guess that Robert has been as nice as to send two forms of the Kafka
> manuscript.

Yes...  Well, Gnus now displays "(2 parts)" if there are two MIME
parts in the buffer, so that would be a clue.  It the user wants to
see what that really is, then `M-t' is available.

But normally (as in 99% of the cases), the user doesn't want to see
what it really is.  A multipart/alternative, for instance, is
virtually excusively used for sending a text/plain and a
text/not-very-plain part, and Gnus will display the part that is
"nicest".  In the text/plain and image/jpeg case (which is not the
normal case), the user may very well want to see the textual version. 

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Matt Armstrong]

Hrvoje Niksic <hniksic@srce.hr> writes:

> Kai.Grossjohann@CS.Uni-Dortmund.DE writes:
> > Hrvoje Niksic <hniksic@srce.hr> writes:
> > 
> >   > Forcing me to press `M-t' on each message with `Mime' in the
> >   > mode line is not particularly nice.
> > 
> > You can also set gnus-unbuttonized-mime-types, no?
> 
> Of course I can do that, but I want the defaults to work right.
> That's what it's all about; I know that *I* can make Gnus do
> (almost) anything I want.

This is going nowhere.  I think the defaults work correctly right now.
If we change it to the way you want, _then_ they will be wrong.

-- 
matta

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

Matt Armstrong <matta@geoworks.com> writes:

> > Of course I can do that, but I want the defaults to work right.
> > That's what it's all about; I know that *I* can make Gnus do
> > (almost) anything I want.
> 
> This is going nowhere.  I think the defaults work correctly right
> now.  If we change it to the way you want, _then_ they will be
> wrong.

Simply disagreeing with me isn't "going nowhere".  The "right" or
"wrong" defaults are obviously subjective.  Having said that, I really
do wish there were an indication of the existence of choice for
multipart/alternative.  Right now the default is to hide it
completely.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
HOW YOU CAN TELL THAT IT'S GOING TO BE A ROTTEN DAY:
        #15 Your pet rock snaps at you.

Re: MIME tests one and two results - Outlook Express

[Dale Hagglund]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Dale Hagglund <rdh@best.com> writes:
> > The user has *got* to have a clue the information is there to begin
> > with, before he can know to use `M-t' or `K b' or whatever.

> Yes...  Well, Gnus now displays "(2 parts)" if there are two MIME
> parts in the buffer, so that would be a clue.

Fair enough.  Still being at 0.51, I hadn't noticed that change yet.
Is this displayed in the message area, or added to the summary buffer
mode line?

Dale.

Re: MIME tests one and two results - Outlook Express

[François Pinard]

Hrvoje Niksic <hniksic@srce.hr> écrit:

> Labeling ability to choose between the alternative parts as ``debugging''
> to make it less appealing to our current implementor is a dirty word trick.

You read me that way?  We may disagree here or there, Hrvoje, but when I
write, I try to honestly express what I really think.  No dirty tricks...

> If I receive HTML junk, and must press `K b' in order to check whether
> there are perhaps some alternatives to the junk (because HTML looks ugly
> in W3[1]), then something must be wrong.

If HTML looks ugly, you perceive it as some kind of a bug, and `K b' is
a way around that bug.  We are still in the realm of debugging.  If you
could just inform Gnus that W3 rendering does not interest you, then Gnus
could automatically use the alternative to select something you like more.
Best would be to improve the HTML renderer, of course.  (I would like it
speedier, for one, yet this is much more easily said than done! :-)

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

François Pinard <pinard@IRO.UMontreal.CA> writes:

> If HTML looks ugly, you perceive it as some kind of a bug, and `K b'
> is a way around that bug.  We are still in the realm of debugging.
> If you could just inform Gnus that W3 rendering does not interest
> you, then Gnus could automatically use the alternative to select
> something you like more.

Yes, but in other circumstances I may want to see the HTML, or
whatever other choices the sender provided.  So I don't perceive
defaulting to HTML as a bug.  If anything is a bug, it's choosing one
alternative and not providing a way to know of existence of others.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Unspeakable horrors from outer space paralyze the living and
resurrect the dead!

Re: MIME tests one and two results - Outlook Express

[Maciej Matysiak]

wmperry@aventail.com (William M. Perry) writes:

> So the viewer  for MM would need  to be careful and not  start up zgv
> unless TERM=linux  and you were actually  on one of  the real virtual
> consoles for that  box.

that's ok. i always check my mail  on one of the server's vc. with TERM
linux.

so, is it  possible to make zgv[1] display  attached pictures? it would
be nice just to open the picture with it in C-d mode.


m.nia.m.

[1] yes, zgv is linux-only console picture viewer. it uses [curse] svgalib.
-- 
 use gnus not guns!

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Dale Hagglund <rdh@best.com> writes:

> Fair enough.  Still being at 0.51, I hadn't noticed that change yet.
> Is this displayed in the message area, or added to the summary buffer
> mode line?

The article buffer mode line.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: MIME tests one and two results - Outlook Express

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Karl Kleinpaste <karl@justresearch.com> writes:
> 
> > If `M-t' could implement a toggle of buttonizedness, it would
> > actually be a good and useful keybinding to have available again.
> 
> Ok; I've added this to 0.55.

Ungh!  I hate the "permanent" toggles in Gnus (yes, I hated the old
`M-t' behaviour too.)  It's perhaps a matter of taste, but I feel
cheated when pressing M-t to see the buttons results in all the
subsequent message being silently buttonized.

Could we please make M-t analogous to `t', i.e. make it work on just
the current article?

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Speak softly and carry a +6 two-handed sword.

Re: MIME tests one and two results - Outlook Express

[Lars Magne Ingebrigtsen]

Hrvoje Niksic <hniksic@srce.hr> writes:

> Ungh!  I hate the "permanent" toggles in Gnus (yes, I hated the old
> `M-t' behaviour too.)  It's perhaps a matter of taste, but I feel
> cheated when pressing M-t to see the buttons results in all the
> subsequent message being silently buttonized.

I'm no big fan of those types of commands, but that was how the old
`M-t' used, and many people liked how it worked.  So I think it makes
sense to have it be permanent.  

> Could we please make M-t analogous to `t', i.e. make it work on just
> the current article?

`K b' buttonizes just one message.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen