Just when I thought I had seen it all...

Just when I thought I had seen it all...

[Lars Magne Ingebrigtsen]

[-- Attachment #1: Type: text/plain, Size: 147 bytes --]

MimeOLE (sic) creates vcards by inserting them as uuencoded, and not
marking anything as, er, anything.  Look, ma, no MIME.

The head, she reels.


[-- Attachment #2: Type: message/rfc822, Size: 862 bytes --]

From: "Trygve Mongstad" <tmongstad@bigfoot.com>
Subject: Re: Ahhhh, At spam kan fremkalle s=E5mye primitiv glede :-)
Date: Mon, 8 Feb 1999 19:14:31 +0100
Message-ID: <79n9no$d5b$1@romeo.dax.net>

--
TrM



begin 666 Trygve Mongstad.vcf
M0D5'24XZ5D-!4D0-"E9%4E-)3TXZ,BXQ#0I..DUO;F=3DS=3D&%D.U1R>6=3DV90T*
M1DXZ5')Y9W9E($UO;F=3DS=3D&%D#0I%34%)3#M04D5&.TE.5$523D54.G1M;VYG
M<W1A9$!B:6=3DF;V]T+F-O;0T*4D56.C$Y.3DP,C X5#$X,30S,5H-"D5.1#I6
&0T%21 T*
`
end

[-- Attachment #3: Type: text/plain, Size: 105 bytes --]


-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@ifi.uio.no * Lars Magne Ingebrigtsen

Re: Just when I thought I had seen it all...

[Jack Vinson]

>>>>> "LMI" == Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

LMI> MimeOLE (sic) creates vcards by inserting them as uuencoded, and not
LMI> marking anything as, er, anything.  Look, ma, no MIME.

LMI> The head, she reels.

Yep.  I saw this from some version of MS IE or Netscape that was sending
vcards.  The problem, of course, is that any text is buried under the
assumption that it is the introductory "you shouldn't be reading this" text
that many MIME mailers insert.

How does one simply view the entire message, gory MIME attachments and all?
gnus-summary-show-article doesn't do it.  

-- 
Jack Vinson <jvinson@chevax.ecs.umass.edu>    http://www.cis.upenn.edu/~vinson/
Zippy: I'm not available for comment..

Re: Just when I thought I had seen it all...

[Lars Magne Ingebrigtsen]

Jack Vinson <jvinson@chevax.ecs.umass.edu> writes:

> How does one simply view the entire message, gory MIME attachments and all?
> gnus-summary-show-article doesn't do it.  

`C-u g' will show you the completely raw article.

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

Re: Just when I thought I had seen it all...

[Kai.Grossjohann]

Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

  > MimeOLE (sic) creates vcards by inserting them as uuencoded, and not
  > marking anything as, er, anything.  Look, ma, no MIME.

How wonderful.  Btw, did you see that there was a button in the
article buffer but the digest group thing didn't show it?  Shouldn't
these two mechanisms see the same `attachments'?

kai
-- 
I like _\bb_\bo_\bt_\bh kinds of music.

Re: Just when I thought I had seen it all...

[Hrvoje Niksic]

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

> How wonderful.  Btw, did you see that there was a button in the
> article buffer but the digest group thing didn't show it?  Shouldn't
> these two mechanisms see the same `attachments'?

I lobbied for the same thing, but to no avail.  The reason that this
happens is that nndoc doesn't use the mm-* functions to parse MIME --
it uses its own /ad hoc/ mechanisms of parsing MIME structure.  In
nndoc's defense, it should be noted that it was written before mm
existed.

Anyway, I think it would be worth it to convert nndoc to use mm
because the latter is more complete and actively maintained.  I
haven't yet managed to convince Francois of this, nor did I contribute
any code of my own.

Re: Just when I thought I had seen it all...

[Lars Magne Ingebrigtsen]

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

> How wonderful.  Btw, did you see that there was a button in the
> article buffer but the digest group thing didn't show it?  Shouldn't
> these two mechanisms see the same `attachments'?

Well, nndoc parses MIME, while the article buffer display thing also
uses mm-uu to pretend that those are MIME articles as well.

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

Re: Just when I thought I had seen it all...

[François Pinard]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 1951 bytes --]

Hrvoje Niksic <hniksic@srce.hr> writes:

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

> > How wonderful.  Btw, did you see that there was a button in the
> > article buffer but the digest group thing didn't show it?  Shouldn't
> > these two mechanisms see the same `attachments'?

> I lobbied for the same thing, but to no avail.  The reason that this
> happens is that nndoc doesn't use the mm-* functions to parse MIME -- it
> uses its own /ad hoc/ mechanisms of parsing MIME structure.  In nndoc's
> defense, it should be noted that it was written before mm existed.

> Anyway, I think it would be worth it to convert nndoc to use mm because
> the latter is more complete and actively maintained.  I haven't yet managed
> to convince Francois of this, nor did I contribute any code of my own.

In my opinion, `nndoc' is much more a debugging device than a MIME viewer.
It is meant to dissect a message to see its internal organs.  I guess that
`mm-*' is much more user oriented, and should be used for the real things.
About modifying `nndoc' to use `mm-*', it might depend if this would make
`nndoc' a more useful debugging device, or not.  It probably does the job
conveniently as it stands.

Trying to turn `nndoc' into a MIME viewer is doomed to fail, in my opinion,
or at least, make a very poor one.  It is too much of a bad start for this.

I remember this publicity, made by some computer company which based its
fast machine on his own chip design, instead of reusing an existing chip.
The poster was showing a big pig, hopelessly and desperately flapping a pair
of small wings at full speed, with no chance of flying ever, of course.
It was saying: "The design has to be good from the start.", or something
along those lines.  Just think at a big pig when you consider `nndoc'! :-)

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

Re: Just when I thought I had seen it all...

[Hrvoje Niksic]

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

> In my opinion, `nndoc' is much more a debugging device than a MIME
> viewer.

An interesting view.  Anyway, since you are the author of the code,
you have the final say over this.  It's just that I consider it sad to 
have *two* separate MIME parsers in Gnus, requiring separate updates
and bugfixes.  Given that Lars abstracted mm into a library, it would
really be nice if it were used as such.

> Trying to turn `nndoc' into a MIME viewer is doomed to fail, in my
> opinion, or at least, make a very poor one.  It is too much of a bad
> start for this.

Heh, and when I asked for a Summary-based MIME viewer, I was always
referred to nndoc.  :-(

Re: Just when I thought I had seen it all...

[François Pinard]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 2411 bytes --]

Hrvoje Niksic <hniksic@srce.hr> écrit:

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

> > In my opinion, `nndoc' is much more a debugging device than a MIME
> > viewer.

> An interesting view.  Anyway, since you are the author of the code,
> you have the final say over this.

No, no!  The code exists, it is not mine anymore, and could be amended
and modified by anyone.  Lars, of course, try to be wise among diverging
trends, and cannot always make absolutely everyone happy on everything.

> It's just that I consider it sad to have *two* separate MIME parsers in
> Gnus, requiring separate updates and bugfixes.  Given that Lars abstracted
> mm into a library, it would really be nice if it were used as such.

It could be indeed.  I wonder if it is worth the effort of featuring `nndoc'
more than it is.  If it was becoming too clever, it might be loosing its
property of clearly dumping MIME internal structure.  On the other hand,
it might become a tool for debugging the `mm' library.  Most probably that
Lars has plenty of other ways to debug that library, and does not need
`nndoc' for this.

> > Trying to turn `nndoc' into a MIME viewer is doomed to fail, in my
> > opinion, or at least, make a very poor one.  It is too much of a bad
> > start for this.

> Heh, and when I asked for a Summary-based MIME viewer, I was always
> referred to nndoc.  :-(

MIME summary is kind of antithetic.  MIME is meant by design to be as
transparent as possible, and so, a "summary view" of MIME structure is
just going against the very idea of MIME.  By asking a Summary-based MIME
viewer, you were calling for something unsuited, and this, naturally,
yielded people to suggest `nndoc' :-).

Hmph!  I fear my humour is a bit hermetic.  My friends often tell me so.

Seriously, yes, I sometimes use `nndoc' myself.  But each time I do,
I have the feeling that something is missing in Gnus, and I go through
`nndoc' around the limitation.  The need of `nndoc' should slowly fade
away as MIME gets better and better implemented in Gnus.  The feeling
that we need a summary should also progressively disappear.  Nevertheless,
I presume Lars will retain `nndoc' for MIME for a good while, the same as
for the capability of seeing a raw message (`C-u g').

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

Re: Just when I thought I had seen it all...

[Steinar Bang]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 527 bytes --]

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

> Seriously, yes, I sometimes use `nndoc' myself.  But each time I do,
> I have the feeling that something is missing in Gnus, and I go
> through `nndoc' around the limitation.

Is nndoc the thing fired off when you do C-d in the summary buffer?
If so, I rather like it, and have used it to step my way through
forwarded articles and digests and even to reply to and forward
individual messages in a digest.

Damn!  I thought it was a _feature_, and a rather good one...! :-)

Re: Just when I thought I had seen it all...

[Hrvoje Niksic]

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

> MIME summary is kind of antithetic.  MIME is meant by design to be as
> transparent as possible, and so, a "summary view" of MIME structure is
> just going against the very idea of MIME.

We've been through this before, and I don't particularly feel like
debating your views here.  Suffice it to say that for me it would at
least be useful to have a Summary-based view of attachments, like
gnus-uu and nndoc allowed since time immemorial.

> Seriously, yes, I sometimes use `nndoc' myself.  But each time I do,
> I have the feeling that something is missing in Gnus, and I go
> through `nndoc' around the limitation.  The need of `nndoc' should
> slowly fade away as MIME gets better and better implemented in Gnus.

You mean the need of nndoc/MIME?  Even so, the statement may be true
for *you*, as you don't want Summary-based MIME representation.

Re: Just when I thought I had seen it all...

[François Pinard]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 1675 bytes --]

Steinar Bang <sb@metis.no> écrit:

> Is nndoc the thing fired off when you do C-d in the summary buffer?

Yes.

> If so, I rather like it, and have used it to step my way through forwarded
> articles and digests and even to reply to and forward individual messages
> in a digest.  Damn!  I thought it was a _feature_, and a rather good
> one...! :-)

Yes, `nndoc' has many niceties, and is undoubtedly useful.  Among its
capabilities, it can display the internal MIME structure tree and handle
physical MIME parts as if each was a genuine, complete message in itself,
and this was only this capability that was under discussion.

They were two motivations behind making `nndoc' able to split MIME parts.
The first was to be a debugging aid, to ease studying how MIME messages are
really made inside, at a time Lars was starting to ponder how to implement
MIME for real in Gnus.  The second was an attempt to alleviate the trend
of overloading MIME presentation with a lot of gadgets that would merely
drawn the fish.  Dumping internal structure, saving physical parts, finding
suboptimal alternate parts, and many such other things, are very auxiliary
to presentation.  Some MIME viewers just make an horrible mix of everything,
and miss the essentials.  I wanted `nndoc' to be a way to un-pollute the
MIME arena in advance, by providing right away an alternate device to soften
various compulsions induced by other MIME readers.  In so, it increased
the chance that Lars would address the real thing with more freedom.

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

Re: Just when I thought I had seen it all...

[François Pinard]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 520 bytes --]

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

> The second [reason for `nndoc'] was an attempt to alleviate the trend
> of overloading MIME presentation with a lot of gadgets that would merely
> drawn the fish.

Yes, indeed, I'm mixing "to draw" and "to drown".  Those are quite different
words.  My knowledge of English is approximative, and it shows.  Sorry!

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

Re: Just when I thought I had seen it all...

[François Pinard]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 780 bytes --]

Hrvoje Niksic <hniksic@srce.hr> écrit:

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

> > MIME is meant by design to be as transparent as possible [...]

> We've been through this before, and I don't particularly feel like
> debating your views here.

My own views are really unimportant, here.  We should have much more
confidence in Nathaniel Boreinstein's views about what MIME really is.
I'm merely conveying the bits I understood from his prose and examples.

P.S. - It seems now that there is a probability that we meet next month, and
hopefully, we will find _plenty_ of (other) things to speak or laugh about!

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

Re: Just when I thought I had seen it all...

[Robert Bihlmeyer]

Hi,

>>>>> On 23 Feb 1999 09:26:44 -0500
>>>>> François Pinard <pinard@iro.umontreal.ca> said:

 FP> MIME summary is kind of antithetic. MIME is meant by design to be
 FP> as transparent as possible, and so, a "summary view" of MIME
 FP> structure is just going against the very idea of MIME.

Most books/papers/reports/etc. are written to be read as a whole.
Still, a summary or table of contents is normally very useful in
practice. For example to skip to an important part quickly.

I think it just natural, that sometimes you want to see the thing
itself, and sometimes you want to see the structure, the bones, the
innards. It would be nice if Gnus supported both alternatives.

	Robbe

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

Re: Just when I thought I had seen it all...

[François Pinard]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 1569 bytes --]

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

>  FP> MIME summary is kind of antithetic.  MIME is meant by design to
>  FP> be as transparent as possible, and so, a "summary view" of MIME
>  FP> structure is just going against the very idea of MIME.

> Most books/papers/reports/etc. are written to be read as a whole.
> Still, a summary or table of contents is normally very useful in
> practice.  For example to skip to an important part quickly.

Agreed, of course.  But in case of a book, the table of contents is designed
to well represent the intended logical structure of the book.  MIME parts
are not necessarily related to the intended logical structure of a message.
They are more a way to assemble various bits, like texts from many scripts,
images and various other things, into a continuum.  Nobody would sensibly
insert, in the table of contents of a book, all transitions between French
and Chinese.  This example should help to see that a summary view of MIME
is nothing much meaningful, it cannot be compared to a table of contents.

> I think it just natural, that sometimes you want to see the thing
> itself, and sometimes you want to see the structure, the bones, the
> innards.  It would be nice if Gnus supported both alternatives.

When you want to see the bones and the innards, you perform a dissection.
An autopsy is to medico-legal people what debugging is to programmers. :-)

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