• A question to all the nodelist clerks

    From Björn Felten@2:203/2 to All on Monday, May 27, 2019 18:07:13
    Now that we have "allow8bit 1" as an option in our modern MakeNl configs, how many of you chose not to use it? I.e. not to allow characters above the old ASCII seven bit restriction.

    It would be interesting to know where the eight bit option is held back, because obviously it still is, despite the developer's attempt to remove that archaic restriction.

    Zone wise, region wise, net wise, hub wise, whatever.

    RFC.



    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Michiel van der Vlist@2:280/5555 to Björn Felten on Monday, May 27, 2019 18:33:11
    Hello Bjrn,

    On Monday May 27 2019 18:07, you wrote to All:

    Now that we have "allow8bit 1" as an option in our modern MakeNl
    configs, how many of you chose not to use it? I.e. not to allow
    characters above the old ASCII seven bit restriction.

    I use it.

    It would be interesting to know where the eight bit option is held back,

    Among others: YOU!

    because obviously it still is, despite the developer's attempt to
    remove that archaic restriction.

    A project called the UTF nodelist has been running for about half a decade in Z2. You have been invited to participate several times, but you declined. You are in a good position being the RC for Scandinavia where sysop names and cities are spelled with non ASCII characters. But you won't play.

    So stop whining.

    Instead connect to the DAILYUTF file echo (available here) and see what happens.


    Cheers, Michiel

    --- GoldED+/W32-MSVC 1.1.5-b20170303
    * Origin: http://www.vlist.org (2:280/5555)
  • From Björn Felten@2:203/2 to Michiel van der Vlist on Monday, May 27, 2019 19:04:45
    MvdV> A project called the UTF nodelist has been running for about half a
    MvdV> decade in Z2.

    Do you seriously believe that people would start using your local UTF nodelist, not supported by any known nodelist to mailer compiler, just to get what we already can get with the present global setup?

    You really are in dire need of a big chunk of FTN reality check, Michiel.



    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Björn Felten@2:203/2 to Michiel van der Vlist on Monday, May 27, 2019 19:27:07
    MvdV> You are in a good position being the RC for Scandinavia where sysop
    MvdV> names and cities are spelled with non ASCII characters. But you won't play.

    I know, and as the responsible RC I try to be, I've been trying to play for many years now. And now, finally, all it takes is for Ward to add that elusive "allow8bit 1" to his MakeNl config -- but somehow he still refuses to do that.

    And I sense that you have something to do with that with your beloved UTC nodelist that nobody needs.

    But my question was most of all directed at all the *Cs not involved with the R28/R29 war going on.



    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Kees van Eeten@2:280/5003.4 to Bj÷rn Felten on Monday, May 27, 2019 19:02:50
    Hello Bjrn!

    27 May 19 18:07, you wrote to All:

    Now that we have "allow8bit 1" as an option in our modern MakeNl configs, how many of you chose not to use it? I.e. not to allow characters above the old ASCII seven bit restriction.

    It would be interesting to know where the eight bit option is held back, because obviously it still is, despite the developer's attempt to remove that archaic restriction.

    Zone wise, region wise, net wise, hub wise, whatever.

    That has to wait until we decide what codepage is to be used.

    May I propose that every Region uses the codepage, that best fits the
    characters used in the region. ;)

    The code page name is to be put in the systemname field of the Region host. ;)

    There is a Nodelist encoded in UTF-8. Only R28 and R56 have contributed.

    Kees

    --- GoldED+/LNX 1.1.5--b20180707
    * Origin: As for me, all I know is that, I know nothing. (2:280/5003.4)
  • From Michiel van der Vlist@2:280/5555 to Björn Felten on Monday, May 27, 2019 19:27:11
    Hello Bjrn,

    On Monday May 27 2019 19:04, you wrote to me:

    MvdV>> A project called the UTF nodelist has been running for about
    MvdV>> half a decade in Z2.

    Do you seriously believe that people would start using your local
    UTF nodelist, not supported by any known nodelist to mailer compiler,

    1) It is not a local project. Iy is a Z2 wide project supported by the ZC2,

    2) AFAIK any 8 bit transparent nodelist to mailer compiler just compiles it.


    Cheers, Michiel

    --- GoldED+/W32-MSVC 1.1.5-b20170303
    * Origin: http://www.vlist.org (2:280/5555)
  • From Kees van Eeten@2:280/5003.4 to Bj÷rn Felten on Monday, May 27, 2019 19:30:38
    Hello Bjrn!

    27 May 19 19:04, you wrote to Michiel van der Vlist:

    MvdV>> A project called the UTF nodelist has been running for about half a
    MvdV>> decade in Z2.

    Do you seriously believe that people would start using your local UTF nodelist, not supported by any known nodelist to mailer compiler, just to get what we already can get with the present global setup?

    You really are in dire need of a big chunk of FTN reality check, Michiel.

    Just like makenl-ng, the Nodelist compilers don't care.

    I have an experimental dailylist where all ae oe and ue in R24 are converted
    to the corresponding characters with diaresis. Also the ringel S is taken
    care of.

    Region 40 names and locations are converted to Hebrew.
    Region 42 names and locations are written properly
    Region 45, 46 and 50 names and locations are backported to Cyrillic, where
    sensible.

    It all works, problems are at the presentation layer unless the can handle
    UTF-8.

    Why have I tried all this, it was a challenge.

    It also failed to convince me, that in my environment more than 7 bit encoding
    is not worth the trouble. In most cases, when it cannot be presented in
    ASCII, I cannot read it anyway.

    Kees

    --- GoldED+/LNX 1.1.5--b20180707
    * Origin: As for me, all I know is that, I know nothing. (2:280/5003.4)
  • From Ward Dossche@2:292/854 to Björn Felten on Monday, May 27, 2019 21:43:12
    And now, finally, all it takes is for Ward to
    add that elusive "allow8bit 1" to his MakeNl config -- but somehow he
    still refuses to do that.

    For me to refuse that, you probably must've asked it.

    I do not recal having received such a request "ever"...

    But my question was most of all directed at all the *Cs not involved
    with the R28/R29 war going on.

    So there's a R28/R29 war going on?

    If so it must be the most boring war ever because I'm not aware, neither is RC28 nor RC29 ... Please enlighten us ...

    \%/@rd

    --- D'Bridge 3.99
    * Origin: Do not meddle in the affairs of wizards (2:292/854)
  • From Kees van Eeten@2:280/5003.4 to Bj÷rn Felten on Tuesday, May 28, 2019 12:55:50
    Hello Bjrn!

    28 May 19 12:36, you wrote to me:

    Once upon a time, up in the northernmost part of Europe, there was a constant struggle about what codepage to use. We had PC, Amiga and Mac owners who all pushed for their hardware installed code page.

    Eventually we got CP850, and it was the perfect compromise. It had all the characters normally used in Swedish, and most other west European languages, in place. And later on, when Windows came into play, their Amiga-ish encoding also fit perfectly well.

    So, what code page do you think would be best?

    Your solutions apparently fits the Scandinavian country, but that is not
    the majority of Fidonet. The encoding of the of the Nodelist is a matter of
    the whole of Fidonet.

    For the Whole of Fidonet the least common denominator is still 7bit ascii.
    The other side of the spectrum is UTF-8. But fidonet is not tuned to multi-
    byte characters.

    Kees

    --- GoldED+/LNX 1.1.5--b20180707
    * Origin: As for me, all I know is that, I know nothing. (2:280/5003.4)
  • From Björn Felten@2:203/2 to Kees van Eeten on Tuesday, May 28, 2019 12:36:02
    That has to wait until we decide what codepage is to be used.

    Once upon a time, up in the northernmost part of Europe, there was a constant struggle about what codepage to use. We had PC, Amiga and Mac owners who all pushed for their hardware installed code page.

    Eventually we got CP850, and it was the perfect compromise. It had all the characters normally used in Swedish, and most other west European languages, in
    place. And later on, when Windows came into play, their Amiga-ish encoding also fit perfectly well.

    So, what code page do you think would be best?


    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Björn Felten@2:203/2 to Ward Dossche on Tuesday, May 28, 2019 11:58:45
    So there's a R28/R29 war going on?

    If so it must be the most boring war ever because I'm not aware, neither is RC28 nor RC29 ... Please enlighten us ...

    In March 2002 R29 was merged into R28 in the ongoing effort in zone 2 to reduce the number of regions.

    Then, 15 years later, October 2017 (almost exactly 30 years after it's original creation), R29 was once again it's own region. From the outside it sure looks like a civil war, where some were seeking independence from the 15yo
    union. Why? Please enlighten us...

    Source:

    http://nodehist.fidonet.org.ua/?address=2%3A29%2F0

    Thank you Pavel Gulchouck @ 2:463/68 for this fantastic service!



    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Ward Dossche@2:292/854 to Björn Felten on Tuesday, May 28, 2019 13:08:04
    Feel free to answer this simple question:

    Why did you suddenly increase the number of regions, when so many of
    us (e.g. R20, R21, R22 and R23 merging into the single R20) did our best
    to reduce it?

    Because a request was received to do so.

    \%/@rd

    --- D'Bridge 3.99
    * Origin: Do not meddle in the affairs of wizards (2:292/854)
  • From Björn Felten@2:203/2 to Kees van Eeten on Tuesday, May 28, 2019 13:32:02
    So, what code page do you think would be best?

    Your solutions apparently fits the Scandinavian country,

    I can only repeat what I wrote in the FIDONEWS echo just a few days ago:

    "Ah well, CP 850 is the primary code page and default OEM code page in many countries, including various English-speaking locales (e.g. in the United Kingdom, Ireland, and Canada)."

    Are you saying that your country is not included?

    Or that it does not include almost all countries in the western world?

    Yes, I know that it can't handle Cyrillic, Arabic, Chinese, Korean, Japanese
    and a lot of other languages, but will you claim that it does not cover the vast majority of countries actively taking part of the international Fidonet?



    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Kees van Eeten@2:280/5003.4 to Bj÷rn Felten on Tuesday, May 28, 2019 13:43:16
    Hello Bjrn!

    28 May 19 13:32, you wrote to me:

    "Ah well, CP 850 is the primary code page and default OEM code page in many countries, including various English-speaking locales (e.g. in the United Kingdom, Ireland, and Canada)."

    Are you saying that your country is not included?

    I would not know, you are talking about what is customary in the MS world.
    I do not know what happens there. As I said, I have used Latin-1 for at least
    the last 18 to 20 years.

    Or that it does not include almost all countries in the western
    world?

    Yes, I know that it can't handle Cyrillic, Arabic, Chinese, Korean, Japanese and a lot of other languages, but will you claim that it does not cover the vast majority of countries actively taking part of the international Fidonet?

    If your favorite code page is so well fitting, why is it not mandatory in
    English speaking Fidonet? And if it is not there, why should it be for the
    Nodelist.

    Kees

    --- GoldED+/LNX 1.1.5--b20180707
    * Origin: As for me, all I know is that, I know nothing. (2:280/5003.4)
  • From Michiel van der Vlist@2:280/5555 to Kees van Eeten on Tuesday, May 28, 2019 13:33:37
    Hello Kees,

    On Tuesday May 28 2019 12:55, you wrote to Bjrn Felten:

    Your solutions apparently fits the Scandinavian country, but that is
    not the majority of Fidonet. The encoding of the of the Nodelist is a matter of the whole of Fidonet.

    And nit just that: CP850 does NOT fit the current majority of Fidonet.

    For the Whole of Fidonet the least common denominator is still 7bit ascii. The other side of the spectrum is UTF-8.

    So having two sets of nodelists one on each side of the spectrum is a reasonable compromise would you not say?

    But fidonet is not tuned to multi- byte characters.

    Fidonet is a chraracter encoding agnostic transport medium. It is the presentation layer of the user interface - strictly speaking not part of Fidonet - that has restrictions. The frgree of srestrictions depend on the implementation. Implementatiosn with less restrictions are slowy surfacing.


    Cheers, Michiel

    --- GoldED+/W32-MSVC 1.1.5-b20170303
    * Origin: http://www.vlist.org (2:280/5555)
  • From Ward Dossche@2:292/854 to Björn Felten on Tuesday, May 28, 2019 12:41:47
    From the outside
    it sure looks like a civil war, where some were seeking independence from the 15yo union. Why? Please enlighten us...

    "You" have stated for a fact there is war between R28 and R29. I don't notice it ... Kees and Wilfred are excellent counsels if I need some information, Michiel has assisted me in the IPv6 enabling of this node.

    So what are you talking about? What are your verifiable clues there is a state of war?

    If you don't have a verifiable clue then keeping quiet is a virtue.

    \%/@rd

    --- D'Bridge 3.99
    * Origin: Do not meddle in the affairs of wizards (2:292/854)
  • From Björn Felten@2:203/2 to Ward Dossche on Tuesday, May 28, 2019 12:56:24
    From the outside
    it sure looks like a civil war, where some were seeking independence from
    the 15yo union. Why? Please enlighten us...

    If you don't have a verifiable clue then keeping quiet is a virtue.

    On the other hand, when someone *have* a clue, keeping quiet is *not* a virtue.

    Feel free to answer this simple question:

    Why did you suddenly increase the number of regions, when so many of us (e.g. R20, R21, R22 and R23 merging into the single R20) did our best to reduce
    it?



    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Björn Felten@2:203/2 to Kees van Eeten on Tuesday, May 28, 2019 14:36:50
    If your favorite code page is so well fitting, why is it not mandatory
    in
    English speaking Fidonet?

    Who would make it, and how? And why? The code page is all about presentation. Fidonet is a network. As such it does not care shit what data it transports. Net neutrality has always been important for Fidonet, the network.

    I know that Fidonet is usually confused with the mail that it transports. Exactly like Internet is confused with the WWW.

    Now it's too late to teach everyone the difference and how the OSI layer works, because nobody cares any more. All that they, save for a few remaining hardcore techies, care about is how what they receive is presented.



    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Kees van Eeten@2:280/5003.4 to Bj÷rn Felten on Tuesday, May 28, 2019 14:48:38
    Hello Bjrn!

    28 May 19 14:36, you wrote to me:

    If your favorite code page is so well fitting, why is it not
    mandatory in
    English speaking Fidonet?

    Who would make it, and how? And why? The code page is all about presentation. Fidonet is a network. As such it does not care shit what data it transports. Net neutrality has always been important for Fidonet, the network.

    I know that Fidonet is technology and does not care what it transports,
    but when using the same encodings for different presentatios, without
    indicating what encoding is used, it is useless. When I connect to R50
    echo and ignore the codepage, I get a lot of rubbish on my screen.
    There is help, there is information in the message, that tells me how
    to convert the message to readable glyphs. Not so in the Nodelist.

    I know that Fidonet is usually confused with the mail that it transports. Exactly like Internet is confused with the WWW.

    Now it's too late to teach everyone the difference and how the OSI layer works, because nobody cares any more. All that they, save for a few remaining hardcore techies, care about is how what they receive is presented.

    All great words, but it does not change the fact, that the presentation
    layer has to present the same characters for all users. You either use
    an encoding that encompasses all characters, or you have consensus on
    how a subset is encoded. If it is neccessary to define a codepage,
    existing or new to not turn Fidonet into a Tower of Babel, so be it,
    then that has to become a part of Fidonet ftn technology.

    If all that is required to satisfy the proper writing of your calling name,
    others may question if it is worth te effeort.

    In your case it is about Bjrn, Bjorn, Bjoern and Bj?rn. That is peanuts to
    what the sysops in the eastern countries have to enjure.


    Kees

    --- GoldED+/LNX 1.1.5--b20180707
    * Origin: As for me, all I know is that, I know nothing. (2:280/5003.4)
  • From Björn Felten@2:203/2 to Kees van Eeten on Tuesday, May 28, 2019 15:56:55
    That is peanuts to what the sysops in the eastern countries have to
    enjure.

    So you are too joining Michiel's quest for the Russian sysops?

    Do you really think they haven't already make substantial contributions to the Fidonet, using Latin characters? From my horizon I see a lot of that, without trying to get Cyrillic characters into the international Fidonet community. YMMV...



    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Michiel van der Vlist@2:280/5555 to Björn Felten on Tuesday, May 28, 2019 15:24:01
    Hello Bjrn,

    On Monday May 27 2019 19:27, you wrote to me:

    MvdV>> You are in a good position being the RC for Scandinavia where
    MvdV>> sysop names and cities are spelled with non ASCII characters.
    MvdV>> But you won't play.

    I know, and as the responsible RC I try to be, I've been trying to
    play for many years now.

    You have been trying to order a dish that is not on the menu.

    And now, finally, all it takes is for Ward to add that elusive
    "allow8bit 1" to his MakeNl config -- but somehow he still refuses to
    do that.

    Ward is running a MakeNl config with "allow8bit 1". Just not the way you want. Presently there are two flavours of nodelists distributed by the ZC2. There is the classic ASCII only flavour for those who can not or do not want to deal with non-ASCII in the nodelist. The other one is the UTF-8 version. It provides
    an equal opportunity for all Fidonet participants to have their data in their native language and native character set in the nodelist.

    A CP850 version for the Western World only is not on the menu.

    So you can either play along and join the UFT-8 nodelist project, or keep sulking and remain empty handed.

    Oh c'mon don't be so pig headed. Just join the UTF-8 project. In the unlikely case that you need any help, I am sure you can find my Fidonews articles where I documented it.

    And I sense that you have something to do with that with your
    beloved UTC nodelist that nobody needs.

    Your antenna is misaligned.

    But my question was most of all directed at all the *Cs not
    involved with the R28/R29 war going on.

    There is no link between the UTF-8 nodelist project and any presumed 28/R29 war.


    Cheers, Michiel

    --- GoldED+/W32-MSVC 1.1.5-b20170303
    * Origin: http://www.vlist.org (2:280/5555)
  • From Björn Felten@2:203/2 to Michiel van der Vlist on Tuesday, May 28, 2019 16:18:52
    MvdV> Ward is running a MakeNl config with "allow8bit 1".

    And you know this for a fact? How? To me he told me that he never heard about it before I brought it to his attention. Once again.

    Alas, we'll see in the next nodelist where I once again am trying it, after having tried it for three years with my 2:203/6 entry:

    http://nodehist.fidonet.org.ua/?address=2%3A203%2F6

    Once again many thanks to Pavel Gulchouck 2:463/68

    If my 2:203/208 is flagged ;E we know if you speak the truth.


    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Björn Felten@2:203/2 to Michiel van der Vlist on Tuesday, May 28, 2019 16:29:25
    If my 2:203/208 is flagged ;E we know if you speak the truth.

    Correction: if it has some question marks instead of the proper characters I
    submitted in it.




    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Ward Dossche@2:292/854 to Björn Felten on Tuesday, May 28, 2019 16:49:54
    http://nodehist.fidonet.org.ua/?address=2%3A203%2F6

    Once again many thanks to Pavel Gulchouck 2:463/68

    Do you have relatives from Dutch origin?

    \%/@rd

    --- D'Bridge 3.99
    * Origin: Do not meddle in the affairs of wizards (2:292/854)
  • From Michiel van der Vlist@2:280/5555 to Björn Felten on Tuesday, May 28, 2019 21:00:46
    Hello Bjrn,

    On Tuesday May 28 2019 13:32, you wrote to Kees van Eeten:

    Are you saying that your country is not included?

    Some characters are missing. The concatenated 'ij' that takes the place of the 'y' in the Dutch alfabet is missing in CP850. Pi is missing. The Euro sign is missing.

    Or that it does not include almost all countries in the western world?

    "Western wold" is a bit vague. Does that include Poland? I'd say yes, as it is part of the EU. The Polish accented 'c' is missing. Is Greece part of the western world. It is part of the EU. Many greek characters are missing in CP850.

    Yes, I know that it can't handle Cyrillic, Arabic, Chinese, Korean, Japanese and a lot of other languages,

    Arabic, Chinese, Korean an those lots of others are not much of a problem as there aren't any Fdionet participant with that as theior native language. Not any more anyway.

    but will you claim that it does not cover the vast majority of
    countries actively taking part of the international Fidonet?

    It is not the number of countries that matters, but rather the number of participant. That Cyrillic is not covered and claiming that is not a problem is
    an insult towards our fellow Fidoians in the East.


    Cheers, Michiel

    --- GoldED+/W32-MSVC 1.1.5-b20170303
    * Origin: http://www.vlist.org (2:280/5555)
  • From Björn Felten@2:203/2 to Michiel van der Vlist on Tuesday, May 28, 2019 22:39:59
    MvdV> That Cyrillic is not covered and claiming that is not a problem is an
    MvdV> insult towards our fellow Fidoians in the East.

    Here we go again. Who made you the spokesperson for the Russians?


    ..

    --- Mozilla/5.0 (Windows; U; Windows NT 5.1; sv-SE; rv:1.9.1.16) Gecko/20101125
    * Origin: news://eljaco.se (2:203/2)
  • From Michiel van der Vlist@2:280/5555 to Björn Felten on Tuesday, May 28, 2019 23:46:54
    Hello Bjrn,

    On Tuesday May 28 2019 16:18, you wrote to me:

    MvdV>> Ward is running a MakeNl config with "allow8bit 1".

    And you know this for a fact?

    Yes.

    How?

    Every day, around 00:14 my local time, I receive a nodelist from Ward that has bytes with the highest bit set.

    To me he told me that he never heard about it before I brought it to
    his attention. Once again.

    No my problem.

    Alas, we'll see in the next nodelist where I once again am trying
    it, after having tried it for three years with my 2:203/6 entry:

    Perhaps you are not following the proper procedure? IIRC I have explained the proper procedure in a Fidonews article. My memory is notoriously inreliable...


    Cheers, Michiel

    --- GoldED+/W32-MSVC 1.1.5-b20170303
    * Origin: http://www.vlist.org (2:280/5555)
  • From Michiel van der Vlist@2:280/5555 to Bjrn Felten on Tuesday, May 28, 2019 22:18:56
    Hello Bjrn Felten!

    28 May 19 22:39:59, Bjrn Felten wrote to Michiel van der Vlist:

    MvdV>> That Cyrillic is not covered and claiming that is not a problem is an
    MvdV>> insult towards our fellow Fidoians in the East.

    Here we go again. Who made you the spokesperson for the Russians?

    ⨭.


    Michiel


    --- goAtEd-windows/386 1.0.5-154-a1826692
    * Origin: (2:280/5555)