1. Forum
  2. FidoNet
  3. FTSC PUBLIC

  • Re^2: Directly include binary data in messages

    From Tim Schattkowsky@2:240/1120.29 to Anna Christina Nass on Tuesday, February 22, 2022 21:40:16
    //Hello Anna,//

    on *22.02.22* at *11:22:00* You wrote in Area *FTSC_PUBLIC*
    to *Ward Dossche* about *"Re: Directly include binary data in messages"*.

    I thought that while the people who do write good software (like Rob or Tim) are currently talking about this topic, I could add an idea to that topic.

    Actually I appreciate and share your idea, but I am currently not certain about the technical implications. Needs a while to digest and get comfortable with / or find reasons why this wont work (I have some candidates).

    Regards,
    Tim

    --- WinPoint 400.2
    * Origin: Original WinPoint Origin! (2:240/1120.29)
  • From Tim Schattkowsky@2:240/1120.29 to Rob Swindell on Tuesday, February 22, 2022 21:55:14
    //Hello Rob,//

    on *22.02.22* at *19:22:28* You wrote in Area *FTSC_PUBLIC*
    to *Anna Christina Nass* about *"Re: Directly include binary data in messages"*.

    Pretty trivial. In fact, any echomail program that supports multiple packet formats (e.g. SBBSecho supports packet types 2.0, 2+, 2e, and 2.2) can be used as a gateway between old and even older technology systems. :-)

    IMHO this is lacking the total system perspective. There are open questions at that level. How is the sending system supposed to know what packet format a receiver expects?

    This could be done in a lot of ways:
    - explicit configuration
    - nodelist flags
    - dynamically during the mail session (i.e., the mailer eventually triggers the final packet format in some way depending on session handshake information)
    - ...

    The most important question for me is still, WHY is there any need for a new packet format? Other than the lack of zone (or even 5D) information, what are the points?

    Regards,
    Tim

    --- WinPoint 400.2
    * Origin: Original WinPoint Origin! (2:240/1120.29)
  • From Rob Swindell@1:103/705 to Tim Schattkowsky on Tuesday, February 22, 2022 14:40:36
    Re: Re^2: Directly include binary data in messages
    By: Tim Schattkowsky to Rob Swindell on Tue Feb 22 2022 09:55 pm

    //Hello Rob,//

    on *22.02.22* at *19:22:28* You wrote in Area *FTSC_PUBLIC*
    to *Anna Christina Nass* about *"Re: Directly include binary data in messages"*.

    Pretty trivial. In fact, any echomail program that supports multiple packet formats (e.g. SBBSecho supports packet types 2.0, 2+, 2e, and 2.2) can be used as a gateway between old and even older technology systems. :-)

    IMHO this is lacking the total system perspective. There are open questions at that level. How is the sending system supposed to know what packet format a receiver expects?

    In SBBSecho, each link can be configured with a preferred packet format. Of course, that only works for explicitly-configured links. Otherwise, it just defaults to stone-age type 2 packet generation.

    This could be done in a lot of ways:
    - explicit configuration
    - nodelist flags
    - dynamically during the mail session (i.e., the mailer eventually triggers the final packet format in some way depending on session handshake information)
    - ...

    The most important question for me is still, WHY is there any need for a new packet format? Other than the lack of zone (or even 5D) information, what are the points?

    <shrug> I'm not saying there is a need. I can theorize packet and packed-message header improvements all day long, but I don't know if that means there's a factual "need".
    --
    digital man (rob)

    Sling Blade quote #19:
    Doyle: I can't so much as drink a damn glass of water around a midget
    Norco, CA WX: 53.5°F, 66.0% humidity, 5 mph SSE wind, 0.00 inches rain/24hrs --- SBBSecho 3.14-Linux
    * Origin: Vertrauen - [vert/cvs/bbs].synchro.net (1:103/705)
  • From Maurice Kinal@1:153/7001 to Tim Schattkowsky on Wednesday, February 23, 2022 09:37:29
    Hey Tim!

    WHY is there any need for a new packet format?

    There isn't and wasn't.

    Other than the lack of zone (or even 5D) information, what are
    the points?

    It depends totally on the bossnode. I have successfully sent the type 2 pkts from a point to a suitable bossnode that knows what it's doing. If anything the pktHeader - no matter which type - acts more like drag on systems rather than providing any useable data worth the effort of extracting it. There are much better ways methinks.

    Life is good,
    Maurice

    ... Ne læt ðin ellen nu gyt gedreosan to dæge.
    Do not now allow your courage to fail on this day.
    --- GNU bash, version 5.1.16(1)-release (x86_64-moosile-linux-gnu)
    * Origin: Little Mikey's Brain - Ladysmith BC, Canada (1:153/7001)
  • From Carol Shenkenberger@1:275/100 to Tim Schattkowsky on Sunday, March 13, 2022 19:47:23
    Re: Re^2: Directly include binary data in messages
    By: Tim Schattkowsky to Anna Christina Nass on Tue Feb 22 2022 09:40 pm

    //Hello Anna,//

    on *22.02.22* at *11:22:00* You wrote in Area *FTSC_PUBLIC*
    to *Ward Dossche* about *"Re: Directly include binary data in messages"*.

    I thought that while the people who do write good software (like Rob o Tim) are currently talking about this topic, I could add an idea to th topic.

    Actually I appreciate and share your idea, but I am currently not certain ab the technical implications. Needs a while to digest and get comfortable with or find reasons why this wont work (I have some candidates).

    Regards,
    Tim


    Good thoughts Tim. Sorry for the late message. I'd sidelined this echo for a bit by accident, while checking fr messages for me and working on setups for some nodes in my net.

    xxcarol
    --- SBBSecho 2.11-Win32
    * Origin: SHENK'S EXPRESS telnet://shenks.synchro.net (1:275/100)
  • From Carol Shenkenberger@1:275/100 to Tim Schattkowsky on Sunday, March 13, 2022 19:55:15
    Re: Re^2: Directly include binary data in messages
    By: Tim Schattkowsky to Rob Swindell on Tue Feb 22 2022 09:55 pm

    //Hello Rob,//

    on *22.02.22* at *19:22:28* You wrote in Area *FTSC_PUBLIC*
    to *Anna Christina Nass* about *"Re: Directly include binary data in messages"*.

    Pretty trivial. In fact, any echomail program that supports multiple packet formats (e.g. SBBSecho supports packet types 2.0, 2+, 2e, and 2. can be used as a gateway between old and even older technology systems. :-)

    IMHO this is lacking the total system perspective. There are open questions that level. How is the sending system supposed to know what packet format a receiver expects?

    This could be done in a lot of ways:
    - explicit configuration
    - nodelist flags
    - dynamically during the mail session (i.e., the mailer eventually triggers final packet format in some way depending on session handshake information) - ...

    The most important question for me is still, WHY is there any need for a new packet format? Other than the lack of zone (or even 5D) information, what ar the points?

    Regards,
    Tim


    I can't answer as to why we need a new packet format, but of your 3 options above Explicit configuration, possibly triggered by nodelist flags.

    Say 2 softwares 'pop up' capable of both. One will autoconfigure from the nodelist, the other doesn't. Both can be manually setup though.

    I don't think option 3 is viable because it means at time of connection to repack *.pkt into another format. Once packed in one, you'd probably have to touch the actual message base to repack right? That or there wouldn't be much point in it that I can see? LOL, ok, I may be missing something!

    xxcarol
    --- SBBSecho 2.11-Win32
    * Origin: SHENK'S EXPRESS telnet://shenks.synchro.net (1:275/100)