My point was that you can usually pronounce foreign words in your native language, if the alphabet is the same and your language has a straightforward mapping between spelling and pronunciation. And then you can derive the spelling from that fake pronunciation. For example, my native language is Finnish, and that approach works for most words in Germanic, Romance, and Baltic languages. Polish and Czech words are harder, as they often have too many consonants to have a Finnish pronunciation.
It goes beyond spelling - English doesn't even have all the phonemes to represent ÄLSKVÄRD if you wanted to (particularly the last half) so you just have to guess how that should sound anyways, especially since unfamiliaretters exist at the same time. If course you can do the approximation, it's not too hard to try, but you can do the same thing with spelling or any other attribute and end up with the same kind of problem - it just isn't the kind of word you'd deal with day to day.
I wouldn't mind if English had some standardized spelling though :D.
