E
this post was submitted on 19 Feb 2024
0 points (NaN% liked)
DACH - jetzt auf feddit.org
8713 readers
1 users here now
Diese Community wird zum 01.07 auf read-only gestellt. Durch die anhäufenden IT-Probleme und der fehlende Support wechseln wir als Community auf www.feddit.org/c/dach - Ihr seid herzlich eingeladen auch dort weiter zu diskutieren!
Das Sammelbecken auf feddit für alle Deutschsprechenden aus Deutschland, Österreich, Schweiz, Liechtenstein, Luxemburg und die zwei Belgier. Außerdem natürlich alle anderen deutschprechenden Länderteile der Welt.
Für länderspezifische Themen könnt ihr euch in folgenden Communities austauschen:
Eine ausführliche Sidebar findet ihr hier: Infothread: Regeln, Feedback & sonstige Infos
Auch hier gelten die Serverregeln von https://feddit.de !
Banner: SirSamuelVimes
founded 1 year ago
MODERATORS
E
Kryptographiekurs flashbacks
But why is it not a German keyboard layout?
It is, I'm using it: https://www.neo-layout.org/
It is just not able to break the habit of typewriter compatibility.
They were asking why the heatmaps pictures in OP were not superimposed on a German keyboard layout.
They were not asking why no one has made a keyboard layout out of the heatmaps data :)
Just thought I'd clarify because you're basically having two different conversations at this point
Ups, I see my confusion. Thanks.
German layout is QWERTZ while English is QWERTY and your neo layout is neither
That one is fitting the pictures even less
Seen
Great project! I really like your design language! Though it would be nice if there was a dark mode and if it supported prefers-color-scheme.
Thanks. It should read prefers-color-scheme. I have dark mode by default, but it's also possible to set dark/light mode too.
A million words doesn't sound like a lot
Oh? Name all of them.
every word in one single picture:
It's incomplete, as you will only finde 95% of words used on ich_iel.
To clarify, it is not a matter of considering the total number of words but rather the number of unique words considered.
Have you considered a similarity search approach? They would handle your oddly specific synonym issue
That million words sounds like a lot.
A source: https://deykun.github.io/diffle-lang/de?p=about-language (It has tooltips displaying percentages for other letters)
Anzahl der Buchstaben in einem Wort
(Number of letters in a word)
For this metric, Wikipedia might not be a representative dataset. Wikipedia uses many technical terms and composite words which tend to be longer than words that are more common in an everyday dialect.