No worries man. I'd also be interested to hear your opinion on the grouping of ancient people by language rather than by ethnicity, political structure, culture, etc. It seems odd to me. You couldn't apply that sort of classification today unless you were very well acquainted with the differences in dialect within a language (such as Iberian Spanish and Mexican Spanish, 2 different peoples with a common language). Also, languages evolve while the group remains the same. Americans of 1800 would speak a different English than modern day Americans. Wouldn't this sort of classification system require historians to be expert linguists? Why did it become the classification of choice?
Sorry about that. A Semitic language is simply a language which belongs to that language family, more technically they genetically related to each other. Arabic, Hebrew, Aramaic and the extinct Akkadian all belong to this group and share common grammatical and phonological features. When one posits a genetic language tree the assumption is that they all descend from a 'Proto-Language' so without a doubt, at some point in time likely MANY millenia ago, all the Semitic languages were the same and they broke off from each other.
With regards to your comments here, there are two types of language classification; genetic and typological. Genetic classification, whilst relying on on common phenomena found amongst several languages will also pair languages up that have diverged greatly. Typological classification is just that, matching languages up by type, irrespective of genetic relations, so for example it would be possible in this form of classification to match up a native American language with similar characteristics to English but not on a genetic basis. That said, it is often muddier than that.
True, languages change. The technical point where one starts calling them different things CAN be reasonably simple. We have a good historical example of this in the history of Romance languages. We know (we have many documents) that at some point in time the languages of Italy, France and Spain were nearly identical, however, as time passed each region developed its own idiosyncracies and eventually diverged from each other to the point of mutual unintelligibility. At that point we can start calling them separate languages. Still this ignores the issue of politics. Swedish and Norwegian are so similar to each other as to be 80% mutually intelligible but are called different languages, but in China you have completely different languages relegated to the status of 'dialects' by dint of political influence and will.