24 April, 2013

Which non-English Morse characters are the most important ones?

The Morse code for the 26 letters of the English language and the digits, you can find everywhere, e.g. here on Wikipedia. All one-, two-, and three-symbol combinations are in use.

In the international alphabet all but four of the four-symbol combinations are used. They are:

MorseGerman
++
Norwegian/
Danish
SpanishEsperantoPolish
16.3.2014
GreekRussianArabic
·-·-ÄÆ--Ą-Яع
---·ÖØ--Ó-Чز
··--Ü--Ŭ--Ю-
----Ch-ChĤ-ΧШش

The two or three first German letters are used in many other languages also, e.g. Swedish, Finnish, Turkish, Hungarian etc.

Note that the Ö/Ø Morse code is an O (---) followed by an E (·), usually written as OE. OE also happens to be how the letter is written if the proper symbol isn't available. That also shows the relationship with the French Œ, but that's a digression that has little to do with Morse code. Likewise, the Morse symbols for both the Ä/Æ and the Ü start with the non-accented letter and are AA and UT respectively.

All other characters use five symbols. These are the ones that I have been able to find:

MorseNorwegian/
Danish/
Swedish/Finnish
FrenchSpanishEsperantoPolish
16.3.2014
Icelandic
16.3.2014
·--·-ÅÀ----
·-··--È--Ł-
··-··-É--Ę-
-·-··-Ç-ĈĆ-
-··-·-Ê (also /)----
--·----Ñ-Ń-
···-·---Ŝ--
--⋅-⋅---Ĝ--
⋅---⋅---Ĵ--
--··-----Ż-
··--·-----Ð
·--··-----Þ

Here also there is a nice pattern for how many of these symbols are formed. Take Esperanto as an example, where the symbols are all formed by appending a short letter to the original one: Ĉ is CE, Ŝ is SN (which also happens to be the prosign for 'understood'), Ĝ is GN, and Ĵ is JE. The Polish  Ł and Ż follow the same convention and are LT and ZT respectively, and so does the Nordic Å which is AK.

There are also six-symbol combinations in Polish: Ź which is --⋅⋅-⋅ (ZN) and Ś which is ⋅⋅⋅-⋅⋅⋅ (SB). The German Wikipedia even lists a Morse code for the double S: ß, ⋅⋅⋅--⋅⋅ or SZ, but I believe two consecutive S's work well or even better.

I recently asked on the Elecraft list if the K3 Morse decoder could support some of these letters, and I asked specifically about Ä/Æ, Ö/Ø, Å, and Ü. These are the four which are shown in the first figure in the upper right-hand corner of this post.

But which ones do you think are the most important ones to include in Morse software?


Sources:


6 comments:

  1. Hi,

    I'm pretty sure that ---- is also used for CH in German.

    From what I hear on the bands, often, the special characters are often just sent either as the regular letters (e.g. sending clé as lce) or (like also in German writing) sending the regular letter followed by an e (e.g. sending schön as schoen).

    The only special character that I more sometimes are the four four-symbol characters.

    FWIW...

    73 de Frank PA4N

    ReplyDelete
    Replies
    1. I've updated the post with Polish and Icelandic and also added Ch to German as suggested.

      Delete
  2. Hi Frank,

    It is interesting that for the Ö which can be written as OE, the Morse signal is actually those two letters combined: --- + . = ---.

    I think the importance of accented letters probably varies from language to language. In Scandinavia we actually both learn and use regularly the Ä (=Æ), Ö (=Ø), and Å in Morse
    code, both in national QSOs and between countries.

    I know this is different in e.g. written French where it seems like one may actually chose to include the accents or not for capital letters and perhaps also in a headline. This may be so for Spanish as well. That would not work in any of the Scandinavian languages.

    I think that German is more like Scandinavian, but I am not quite sure.

    ReplyDelete
  3. Hi Sverre,

    I will try to post again. Last attempt was lost in space.

    I think the four German/Scandinavian characters are the most important. Accented character you usually can do without, such as in French where it is usually understandable even if you leave the accents out. However, there are exceptions such as the words sale (dirty) and salé (salty) to make one example. Many French also make errors when accenting and accent tréma or cédille is rarely used in private correspondance. I think the same goes for Spanish, but I think they are more strict with the use of their accented characters than the French.

    I think the norm is that capitals in French should not be accented, but it is not a rule and it is up to the writer to chose her style.

    All the four German/Scandinavian characters can be written as oe, ae, ue and aa. This is how you would write the German/Scandinavian languages on a non German/Scandinavian typewriter. I am old enough to have been trough that exercise many times. This would be the easiest method for adding these four characters to the KX3 display. The only drawback is that this method will use up more of the available display space, but as these characters are relatively infrequent it might be acceptable. We should actually look at the letter frequencies to see what overhead it might give.

    I expect the KX3 to be quite popular as a field-day rig and during field-days there is often non-ham spectators. It would then be very nice if the display would show properly readable German and Scandinavian instead of puzzling text with lots of asterisks. So I am very supportive of your idea.

    73 Frode, LA2RL

    ReplyDelete
  4. Hei Frode,

    Thanks for your thoughtful comment. Yes, the letters can be written as double combinations.

    But the most frequent of them, Å, is actually no 14 in frequency in the Norwegian language and is used 2% of the time according to http://no.wikipedia.org/wiki/Frekvensanalyse, so it isn't that rare really.

    ReplyDelete
  5. Hi again,

    I have just looked at the letter frequencies. For German it is:
    Ü = 0.65%, Ä = 0.54%, Ö = 0.30%

    Norwegian:
    Å = 1.48%, Ø = 0.89%, Æ = 0.26%

    Swedish:
    Ä = 2.10%, Å = 1.66%, Ö = 1.50%

    Danish:
    Å = 1.17%, Ø = 0.98%, Æ = 0.97%

    As we see the Scandinavian languages have higher frequencies than German with Swedish coming out higher than Danish and Norwegian.

    Finnish is a special case with
    Ö = 5.21%, Ä = 0.39%

    These figures are only rough indications as I expect radio amateur texts will have quite different frequency characteristic from normal texts in these languages due to all the special abbreviations, Q-codes etc.

    73 Frode, LA2RL

    ReplyDelete