headers
Doug Royer | 1 Mar 2003 01:24

Push of CAP to ietf


I am sending the 17-FEB-2003 version + some changes to the IETF list
tonight. I know there are some issues still, but by sending it
to the IETF it can be discussed in S.F.

I'll post a URL to the diffs.

-
--

-- 

  Doug Royer                     |   http://INET-Consulting.com
  -------------------------------|-----------------------------
  Doug <at> Royer.com                 | Office: (208)612-INET
  http://Royer.com/People/Doug   |    Fax: (866)594-8574
                                 |   Cell: (208)520-4044

                 We Do Standards - You Need Standards
Attachment (smime.p7s): application/x-pkcs7-signature, 4411 bytes
Permalink | Reply | 0 comments |
headers
Doug Royer | 1 Mar 2003 01:49

draft-ietf-calsch-cap-10


I have submitted draft-ietf-calsch-cap-10 (the deadline is tomorrow).

I placed copies at:

	http://INET-Consulting.com/draft-ietf-calsch-cap-10.txt
	http://INET-Consulting.com/draft-ietf-calsch-cap-10.html
	http://INET-Consulting.com/draft-ietf-calsch-cap-10.xml

The difference between cap-17-FEB-2003.txt and what was
submitted today:

	http://INET-Consulting.com/draft-ietf-calsch-cap-10.17-FEB.diff

The difference between draft-ietf-calsch-cap-09.txt
and what was submitted today:

	http://INET-Consulting.com/draft-ietf-calsch-cap-10.cap-09.diff

--

-- 

  Doug Royer                     |   http://INET-Consulting.com
  -------------------------------|-----------------------------
  Doug <at> Royer.com                 | Office: (208)612-INET
  http://Royer.com/People/Doug   |    Fax: (866)594-8574
                                 |   Cell: (208)520-4044

                 We Do Standards - You Need Standards
Attachment (smime.p7s): application/x-pkcs7-signature, 4411 bytes
(Continue reading)

Permalink | Reply | 0 comments |
headers
Andrea Campi | 1 Mar 2003 13:08
Picon

Re: The charset issue


On Fri, Feb 28, 2003 at 08:50:26AM -0500, Mark Swanson wrote:
> Perhaps you missed the discussion about this.
> There is a big problem with it: if test1.ics is in utf-8 and test2.ics is in 
> latin1 the web server will serve both as latin1. This can not work.

Uhm... I see no reason why that should be the case. It's an
implementation detail really, but just for an example, CERN-style
meta files have been around for like 10 years; if using Apache you
can also use mod_asis, even though it's not optimal. Or even if
using static files you could pass all requests through a script
which determines the appropriate charset via your preferred
mechanism, and adds it to the MIME headers.

Bye,
	Andrea

--

-- 
Andrea Campi                              mailto:a.campi <at> inet.it
I.NET S.p.A. - BT Ignite                  http://www.inet.it
Technical Dept. - R&D			  phone: +39 02 32863 ext 1
v. Darwin, 85 - I-20019			  fax: +39 02 32863 ext 7705
Settimo Milanese (MI), Italy
Permalink | Reply | 1 comment |
headers
Andrea Campi | 1 Mar 2003 13:20
Picon

Re: The charset issue


On Fri, Feb 28, 2003 at 09:17:04AM -0500, Mark Swanson wrote:
> On February 28, 2003 06:23 am, Andrea Campi wrote:
> >
> > Since the discussion has reached (IMO) a dead end, I'll voice my opinion.
> > I am 1000% of Doug's opinion. Mark, you are presenting a non-compliant
> > implementation as reason for changing RFC2445. Like it or not, an http
> 
> No Andrea. This is not why I am doing this at all.
> Perhaps a summary of reasons will help:
> 
> 1. MIME is not the universe (to steal a phrase from John Stacke :-). 2445 
> states iCalendar objects are to work fine outside the MIME environment. To 
> quote from 2445, "the format in this memo is equally applicable for use 
> outside of a MIME message content type.".

This is exactly the reason why I think no change is necessary. If
you are in a MIME environment, use MIME. If you are not, you MUST
use UTF-8. I think this is quite reasonable in fact.

> There are the obvious examples of using an iCalendar object outside the MIME 
> environment: http/ftp/custom protocols - even copying the iCalendar object as 
> presented by Outlook (in text format, which does not show the MIME headers as 
> part of the iCalendar object) to the clipboard.  John Strake has stated that 
> no MIME implementations (MUA or HTTP) include the MIME headers when it saves 
> a MIME body-part.

You see - implementation errors. ;-) Even for clipboards, you do
have a choice - you can trancode to UTF-8 before sending the data
to the clipboard, i.e. before throwing away the charset information.
(Continue reading)

Permalink | Reply | 10 comments |
headers
Andrea Campi | 1 Mar 2003 13:23
Picon

Re: The charset issue


On Fri, Feb 28, 2003 at 01:43:07PM -0500, Lawrence Greenfield wrote:
>    There are the obvious examples of using an iCalendar object outside
>    the MIME environment: http/ftp/custom protocols - even copying the
>    iCalendar object as presented by Outlook (in text format, which
>    does not show the MIME headers as part of the iCalendar object) to
>    the clipboard.  John Strake has stated that no MIME implementations
>    (MUA or HTTP) include the MIME headers when it saves a MIME
>    body-part.
> 
> Of course not. They should save objects in the local charset, doing
> conversions if necessary.

Absolutely not, they should convert everything to UTF-8. I honestly
can't see the reason for NOT doing that - since any conforming
implementation MUST be able to handle at least UTF-8, why go to
the effort of using anything else as a native charset? I mean I
can see the need to import or even export data in different
charsets, but UTF-8 should be the main one.

Bye,
	Andrea

--

-- 
Andrea Campi                              mailto:a.campi <at> inet.it
I.NET S.p.A. - BT Ignite                  http://www.inet.it
Technical Dept. - R&D			  phone: +39 02 32863 ext 1
v. Darwin, 85 - I-20019			  fax: +39 02 32863 ext 7705
Settimo Milanese (MI), Italy
(Continue reading)

Permalink | Reply | 0 comments |
headers
pregen | 3 Mar 2003 00:15

Re: Push of CAP to ietf


Thanks.  We're getting ready to update the calsch website as well.  I have 
a person who has volunteered to keep it up to date - I'm getting too busy 
to be very current.
___________________
Patricia Egen Consulting
www.egenconsulting.com
423-875-2652

Doug Royer <Doug <at> royer.com>
Sent by: owner-ietf-calendar <at> mail.imc.org
02/28/2003 19:24
Please respond to "ietf-calendar <at> imc.org"

 
        To:     "ietf-calendar <at> imc.org" <ietf-calendar <at> imc.org>
        cc: 
        Subject:        Push of CAP to ietf

I am sending the 17-FEB-2003 version + some changes to the IETF list
tonight. I know there are some issues still, but by sending it
to the IETF it can be discussed in S.F.

I'll post a URL to the diffs.

-
--

-- 

  Doug Royer                     |   http://INET-Consulting.com
  -------------------------------|-----------------------------
(Continue reading)

Permalink | Reply | 0 comments |
headers
Mark Swanson | 3 Mar 2003 04:14

Re: The charset issue


On March 1, 2003 07:08 am, Andrea Campi wrote:
> On Fri, Feb 28, 2003 at 08:50:26AM -0500, Mark Swanson wrote:
> > Perhaps you missed the discussion about this.
> > There is a big problem with it: if test1.ics is in utf-8 and test2.ics is
> > in latin1 the web server will serve both as latin1. This can not work.
>
> Uhm... I see no reason why that should be the case. It's an
> implementation detail really, but just for an example, CERN-style
> meta files have been around for like 10 years; if using Apache you

My web server doesn't support CERN-style meta files.

> can also use mod_asis, even though it's not optimal. Or even if

mod_asis is not supported by my web browser, and even though I agree it would 
serve the purpose of providing the MIME headers I still do not like the fact 
that requiring MIME breaks the fundamental encapsulation of iCalendar.

> using static files you could pass all requests through a script
> which determines the appropriate charset via your preferred
> mechanism, and adds it to the MIME headers.

You could, but imposing that as a restriction to using iCalendar is not 
reasonable.

Again, the fundamental objection is: 2445 should work equally well outside of 
a MIME environment.

--

--
(Continue reading)

Permalink | Reply | 0 comments |
headers
Mark Swanson | 3 Mar 2003 04:25

Re: The charset issue


On March 1, 2003 07:20 am, Andrea Campi wrote:
> On Fri, Feb 28, 2003 at 09:17:04AM -0500, Mark Swanson wrote:
> > On February 28, 2003 06:23 am, Andrea Campi wrote:
> > > Since the discussion has reached (IMO) a dead end, I'll voice my
> > > opinion. I am 1000% of Doug's opinion. Mark, you are presenting a
> > > non-compliant implementation as reason for changing RFC2445. Like it or
> > > not, an http
> >
> > No Andrea. This is not why I am doing this at all.
> > Perhaps a summary of reasons will help:
> >
> > 1. MIME is not the universe (to steal a phrase from John Stacke :-). 2445
> > states iCalendar objects are to work fine outside the MIME environment.
> > To quote from 2445, "the format in this memo is equally applicable for
> > use outside of a MIME message content type.".
>
> This is exactly the reason why I think no change is necessary. If
> you are in a MIME environment, use MIME. If you are not, you MUST
> use UTF-8. I think this is quite reasonable in fact.

I mostly quite like that.
Unfortunately 2445 does not agree with us. 
Instead of a CHARSET property would you agree to rewording 2445 so UTF-8 MUST 
be used if not in a MIME environment?

> > There are the obvious examples of using an iCalendar object outside the
> > MIME environment: http/ftp/custom protocols - even copying the iCalendar
> > object as presented by Outlook (in text format, which does not show the
> > MIME headers as part of the iCalendar object) to the clipboard.  John
(Continue reading)

Permalink | Reply | 9 comments |
headers
pregen | 3 Mar 2003 04:58

IETF San Francisco - who's going?


This is a quick note to get a count of who will be at the IETF meeting in 
San Francisco.  If you are going and plan on being at the CalSch session, 
let us know.  Also, we will do our traditional Monday night dinner and 
meet in the lobby of the hotel around 6:30 pm.
___________________
Patricia Egen Consulting
www.egenconsulting.com
423-875-2652
Permalink | Reply | 0 comments |
headers
Doug Royer | 3 Mar 2003 18:56

Re: The charset issue


Mark Swanson wrote:

> 
> I mostly quite like that.
> Unfortunately 2445 does not agree with us. 
> Instead of a CHARSET property would you agree to rewording 2445 so UTF-8 MUST 
> be used if not in a MIME environment?

That is not going to happen ether as several asian countries charsets
are more efficient then UTF-8 for them. We did not pick UTF-8 and the
wording in 2445 at random. It was discussed over a few years and agreed
upon. Changing 2445 in an incompatible way is just not going to happen.
If we changed it to a MUST it would make currently compliant applications
non compliant. Do you think they are just going to ignore that issue?

 >>
 >> ...

> Oh? You are going to expect all past and future software to do this? 
> Notepad.exe for example?

It is *not* the job of an interoperability protocol to be view-ready.
It is the job of an interoperability protocol to be interoperable
with multiple venders across multiple countries. Your OS tool specific
example is not relevant. The point made was that you can write
code to do copy/paste, do it or not but do not expect every
vendor to agree that they have to change.

>>I agree with you that it would be nice to be universally compatible;
(Continue reading)

Permalink | Reply | 8 comments |

Gmane