Generation of message id - Mozilla

This is a discussion on Generation of message id - Mozilla ; I have a little discussion about the generation of a valid (rfc compliant) message id. Some news reader permit non-compliant message id's some don't. What is the status foe SeaMonkey mail/news and thunderbird. Is there a possibility to force, f.e. ...

+ Reply to Thread
Results 1 to 14 of 14

Thread: Generation of message id

  1. Generation of message id

    I have a little discussion about the generation of a valid (rfc
    compliant) message id.
    Some news reader permit non-compliant message id's some don't.
    What is the status foe SeaMonkey mail/news and thunderbird.
    Is there a possibility to force, f.e. in the user.js or otherwise, a
    non compliant message id?

    --
    Belgische Ardennen http://users.kbc.skynet.be/fi001005 Ardennes Belge



  2. Re: Generation of message id

    Bernard Mercier wrote:
    > I have a little discussion about the generation of a valid (rfc
    > compliant) message id.
    > Some news reader permit non-compliant message id's some don't.
    > What is the status foe SeaMonkey mail/news and thunderbird.
    > Is there a possibility to force, f.e. in the user.js or otherwise, a
    > non compliant message id?
    >


    correct me if I'm wrong on this, but message IDs don't come
    from the program, but rather from the server. Its the
    server that gives messages their IDs. Am I not correct on
    this?

    --
    *IMPORTANT*: Sorry folks, but I cannot provide email
    help!!!! Emails to me may become public

    Notice: This posting is protected under the Free Speech
    Laws, which applies everywhere, except for some strange
    reason, not to the mozilla.org servers, where you may get
    banned.

    Peter Potamus & His Magic Flying Balloon:
    http://www.toonopedia.com/potamus.htm

  3. Re: Generation of message id

    Peter Potamus the Purple Hippo wrote:
    > Bernard Mercier wrote:
    >> I have a little discussion about the generation of a valid (rfc
    >> compliant) message id.
    >> Some news reader permit non-compliant message id's some don't.
    >> What is the status foe SeaMonkey mail/news and thunderbird.
    >> Is there a possibility to force, f.e. in the user.js or otherwise, a
    >> non compliant message id?
    >>

    >
    > correct me if I'm wrong on this, but message IDs don't come from the
    > program, but rather from the server. Its the server that gives messages
    > their IDs. Am I not correct on this?
    >

    You are semi not correct, the server *will* and usually *does* generate
    MsgID. At that point server tuning takes over, I've tuned servers to
    always generate MsgID, to accept the user given ID (or reject it if it
    was a duplicate) or to rename the old MsgID as "X-Original-Message-ID"
    and generate a new one.

    Same goes for non-compliant IDs, many servers reject them, because many
    *other* servers reject them, and why send out a message many sites won't
    use?

    --
    Bill Davidsen
    "We have more to fear from the bungling of the incompetent than from
    the machinations of the wicked." - from Slashdot

  4. Re: Generation of message id

    On Sat, 08 Mar 2008 10:43:07 -0800, Peter Potamus the Purple Hippo
    wrote:

    > correct me if I'm wrong on this, but message IDs don't come
    > from the program, but rather from the server. Its the
    > server that gives messages their IDs. Am I not correct on
    > this?


    Most news servers (every one that I've used - including
    news.mozilla.org) allow the news client to assign Message-ID's. If
    one is not present, the server will add it.

    Generating your own ID's at the client can be very handy for custom
    filtering. For example, all my posts, and all followups to my
    posts are highlighted based on my uniqe MID suffix (as shown in my
    headers here).

    -sw

  5. Re: Generation of message id

    On 3/8/08 1:43 PM, _Peter Potamus the Purple Hippo_ spoke thusly:
    > Bernard Mercier wrote:
    >> I have a little discussion about the generation of a valid (rfc
    >> compliant) message id.
    >> Some news reader permit non-compliant message id's some don't.
    >> What is the status foe SeaMonkey mail/news and thunderbird.
    >> Is there a possibility to force, f.e. in the user.js or otherwise, a
    >> non compliant message id?
    >>

    >
    > correct me if I'm wrong on this, but message IDs don't come from the
    > program, but rather from the server. Its the server that gives messages
    > their IDs. Am I not correct on this?


    Some clients allow the server to generate it; some don't. By default,
    SeaMonkey leaves it to the server, but you can make it generate a
    message ID.

    You have to use a fully qualified domain name.

    --
    Chris Ilias
    List-owner: support-firefox, support-thunderbird, test-multimedia

  6. Re: Generation of message id

    Chris Ilias wrote:
    > On 3/8/08 1:43 PM, _Peter Potamus the Purple Hippo_ spoke thusly:
    >> Bernard Mercier wrote:
    >>> I have a little discussion about the generation of a valid (rfc
    >>> compliant) message id.
    >>> Some news reader permit non-compliant message id's some don't.
    >>> What is the status foe SeaMonkey mail/news and thunderbird.
    >>> Is there a possibility to force, f.e. in the user.js or otherwise, a
    >>> non compliant message id?
    >>>

    >> correct me if I'm wrong on this, but message IDs don't come from the
    >> program, but rather from the server. Its the server that gives messages
    >> their IDs. Am I not correct on this?

    >
    > Some clients allow the server to generate it; some don't. By default,
    > SeaMonkey leaves it to the server, but you can make it generate a
    > message ID.
    >
    > You have to use a fully qualified domain name.
    >


    OK, thanks, then I stand corrected.

    --
    *IMPORTANT*: Sorry folks, but I cannot provide email
    help!!!! Emails to me may become public

    Notice: This posting is protected under the Free Speech
    Laws, which applies everywhere, except for some strange
    reason, not to the mozilla.org servers, where you may get
    banned.

    Peter Potamus & His Magic Flying Balloon:
    http://www.toonopedia.com/potamus.htm

  7. Re: Generation of message id

    Bill Davidsen a émis l'idée suivante :
    > Peter Potamus the Purple Hippo wrote:
    >> Bernard Mercier wrote:
    >>> I have a little discussion about the generation of a valid (rfc compliant)
    >>> message id.
    >>> Some news reader permit non-compliant message id's some don't.
    >>> What is the status foe SeaMonkey mail/news and thunderbird.
    >>> Is there a possibility to force, f.e. in the user.js or otherwise, a non
    >>> compliant message id?
    >>>

    >>
    >> correct me if I'm wrong on this, but message IDs don't come from the
    >> program, but rather from the server. Its the server that gives messages
    >> their IDs. Am I not correct on this?
    >>

    > You are semi not correct, the server *will* and usually *does* generate
    > MsgID. At that point server tuning takes over, I've tuned servers to always
    > generate MsgID, to accept the user given ID (or reject it if it was a
    > duplicate) or to rename the old MsgID as "X-Original-Message-ID" and generate
    > a new one.
    >
    > Same goes for non-compliant IDs, many servers reject them, because many
    > *other* servers reject them, and why send out a message many sites won't use?

    Hi, do you mean that when a news reader program permits own msg-ID not
    compliant with the rfc, that there are servers rejecting such a
    message?

    --
    Belgische Ardennen http://users.kbc.skynet.be/fi001005 Ardennes Belge



  8. Re: Generation of message id

    Chris Ilias a formulé ce zaterdag :
    > On 3/8/08 1:43 PM, _Peter Potamus the Purple Hippo_ spoke thusly:
    >> Bernard Mercier wrote:
    >>> I have a little discussion about the generation of a valid (rfc compliant)
    >>> message id.
    >>> Some news reader permit non-compliant message id's some don't.
    >>> What is the status foe SeaMonkey mail/news and thunderbird.
    >>> Is there a possibility to force, f.e. in the user.js or otherwise, a non
    >>> compliant message id?
    >>>

    >>
    >> correct me if I'm wrong on this, but message IDs don't come from the
    >> program, but rather from the server. Its the server that gives messages
    >> their IDs. Am I not correct on this?

    >
    > Some clients allow the server to generate it; some don't. By default,
    > SeaMonkey leaves it to the server, but you can make it generate a message ID.
    >
    > You have to use a fully qualified domain name.

    Thanks for the info.

    --
    Belgische Ardennen http://users.kbc.skynet.be/fi001005 Ardennes Belge



  9. Re: Generation of message id

    Bernard Mercier wrote:

    > Bill Davidsen a émis l'idée suivante :
    >
    > > Same goes for non-compliant IDs, many servers reject them, because
    > > many *other* servers reject them, and why send out a message many
    > > sites won't use?

    >
    > Hi, do you mean that when a news reader program permits own msg-ID
    > not compliant with the rfc, that there are servers rejecting such a
    > message?


    Yes. AFAIK, most servers can only check that the syntax conforms with
    the RFC and that there's no MID collision with an article they already
    have. I don't know of any that check that the FQDN is actually the
    domain from which the article is coming, which may or may not be
    required by the RFC. (It's a very old RFC, much more ambiguous than
    ones written today.)


  10. Re: Generation of message id

    »Q« wrote:
    > Bernard Mercier wrote:
    >
    >> Bill Davidsen a émis l'idée suivante :
    >>
    >> > Same goes for non-compliant IDs, many servers reject them, because
    >> > many *other* servers reject them, and why send out a message many
    >> > sites won't use?

    >>
    >> Hi, do you mean that when a news reader program permits own msg-ID
    >> not compliant with the rfc, that there are servers rejecting such a
    >> message?

    >
    > Yes. AFAIK, most servers can only check that the syntax conforms with
    > the RFC and that there's no MID collision with an article they already
    > have. I don't know of any that check that the FQDN is actually the
    > domain from which the article is coming, which may or may not be
    > required by the RFC. (It's a very old RFC, much more ambiguous than
    > ones written today.)


    Take a look at the MID's generated by your message and this one. Then
    look at one generated by the server. Yours and mine are similar in that
    they contain the date/time albeit in different formats (mine is hex),
    followed by a randomly generated number and domain. Then look at a
    server generated MID. There appears to be no standard other than being
    unique.

    Date.random@domain

    47D2FAA6.3040302@Senior.Envision ## Sample of my MID
    20080308171517.5775e1a1@bellgrove.remarqs.net ## »Q«
    WZGdnS8LzIrYk07anZ2dnUVZ_sqinZ2d@mozilla.org ## Server

    Rinaldi
    --
    Bubble Memory, n.:
    A derogatory term, usually referring to a person's
    intelligence. See also "vacuum tube".

  11. Re: Generation of message id

    "Rinaldi J. Montessi" wrote:

    > »Q« wrote:


    > > Yes. AFAIK, most servers can only check that the syntax conforms
    > > with the RFC and that there's no MID collision with an article they
    > > already have. I don't know of any that check that the FQDN is
    > > actually the domain from which the article is coming, which may or
    > > may not be required by the RFC. (It's a very old RFC, much more
    > > ambiguous than ones written today.)

    >
    > Take a look at the MID's generated by your message and this one. Then
    > look at one generated by the server. Yours and mine are similar in
    > that they contain the date/time albeit in different formats (mine is
    > hex), followed by a randomly generated number and domain. Then look
    > at a server generated MID. There appears to be no standard other
    > than being unique.
    >
    > Date.random@domain
    >
    > 47D2FAA6.3040302@Senior.Envision ## Sample of my MID
    > 20080308171517.5775e1a1@bellgrove.remarqs.net ## »Q«
    > WZGdnS8LzIrYk07anZ2dnUVZ_sqinZ2d@mozilla.org ## Server


    Yeah, the spec is pretty much just to guarantee uniqueness;
    timestamps are a common way of doing it. It also rules out some
    characters which must not be used. I really wish some format had been
    chosen that's not indistinguishable from e-mail addresses.

    I'm setting followup to m.general, since I think we're getting away from
    any SeaMonkey issues.


  12. Re: Generation of message id

    "Rinaldi J. Montessi" wrote:

    > »Q« wrote:


    > > Yes. AFAIK, most servers can only check that the syntax conforms
    > > with the RFC and that there's no MID collision with an article they
    > > already have. I don't know of any that check that the FQDN is
    > > actually the domain from which the article is coming, which may or
    > > may not be required by the RFC. (It's a very old RFC, much more
    > > ambiguous than ones written today.)

    >
    > Take a look at the MID's generated by your message and this one. Then
    > look at one generated by the server. Yours and mine are similar in
    > that they contain the date/time albeit in different formats (mine is
    > hex), followed by a randomly generated number and domain. Then look
    > at a server generated MID. There appears to be no standard other
    > than being unique.
    >
    > Date.random@domain
    >
    > 47D2FAA6.3040302@Senior.Envision ## Sample of my MID
    > 20080308171517.5775e1a1@bellgrove.remarqs.net ## »Q«
    > WZGdnS8LzIrYk07anZ2dnUVZ_sqinZ2d@mozilla.org ## Server


    Yeah, the spec is pretty much just to guarantee uniqueness;
    timestamps are a common way of doing it. It also rules out some
    characters which must not be used. I really wish some format had been
    chosen that's not indistinguishable from e-mail addresses.

    I'm setting followup to m.general, since I think we're getting away from
    any SeaMonkey issues.


  13. Re: Generation of message id

    »Q« wrote:
    ------------------------------------
    >> Take a look at the MID's generated by your message and this one. Then
    >> look at one generated by the server. Yours and mine are similar in
    >> that they contain the date/time albeit in different formats (mine is
    >> hex), followed by a randomly generated number and domain. Then look
    >> at a server generated MID. There appears to be no standard other
    >> than being unique.
    >>
    >> Date.random@domain
    >>
    >> 47D2FAA6.3040302@Senior.Envision ## Sample of my MID
    >> 20080308171517.5775e1a1@bellgrove.remarqs.net ## »Q«
    >> WZGdnS8LzIrYk07anZ2dnUVZ_sqinZ2d@mozilla.org ## Server

    >
    > Yeah, the spec is pretty much just to guarantee uniqueness;
    > timestamps are a common way of doing it. It also rules out some
    > characters which must not be used. I really wish some format had been
    > chosen that's not indistinguishable from e-mail addresses.


    FWIW Jamie Z wrote a pretty compelling argument against using server
    assigned MID's. Involved messages both posted to USENET and emailed;
    and the possible response scenarios thereto.

    > I'm setting followup to m.general, since I think we're getting away from
    > any SeaMonkey issues.


    Yeah. Heaven Forbid! (sigh)

    Rinaldi
    --
    Where humor is concerned there are no standards -- no one can say what
    is good or bad, although you can be sure that everyone will.
    -- John Kenneth Galbraith

  14. Re: Generation of message id

    Bernard Mercier vient de nous annoncer :
    > I have a little discussion about the generation of a valid (rfc compliant)
    > message id.
    > Some news reader permit non-compliant message id's some don't.
    > What is the status foe SeaMonkey mail/news and thunderbird.
    > Is there a possibility to force, f.e. in the user.js or otherwise, a non
    > compliant message id?

    Thank you all for the info given.
    I did put the question on purpose here as I wanted to know how
    SeaMonkey/Thunderbird handled the matter.
    It was not intended to have a general discussion about msg-id.

    --
    Belgische Ardennen http://users.kbc.skynet.be/fi001005 Ardennes Belge



+ Reply to Thread