SBCS English will be sufficient as the primary language.
There are however considerations that should be discussed. The main consideration is that English (CCSID 37) and Polish (CCSID 870) do not share the same character set. CCSID 870 for instance supports (among several other characters) the characters Ł, Ř, Ş, and Ź which do not exist in CCSID 37 (along the same lines, there are an equal number of characters in CCSID 37 that do not exist in 870). These characters may or may not be used in Polish (I don’t know Polish but I do know CCSIDs) and therefore may or may not be used in the Polish translations. If they are used, and are important to your Polish users, then the display devices will need to be configured as using 870 in order to display and enter these “870 only” characters. The use of a different display device CCSID (actually CHRID on the *DEVD) can have signification implications if the Polish and English users are also using the same EBCDIC databases. If the databases are CCSID tagged 37 and the display devices are a combination of 37 and 870 then you are looking at some real headaches that are too large to go into in this type of question and answer approach.
The “quick fix” in this situation is to make sure the two sets of users are using separate databases. The “correct fix” in this situation is to use a CCSID such as Unicode which can support the character sets of multiple languages (and use Unicode enabled display devices).
You will also want to review the i5/OS Information Center. Language considerations are discussed under the topic Programming, subtopic i5/OS Globalization.
Bruce Vining Services, LLC.