"because >99% of computer users are non-programmers, making the freedoms to study, modify & redistribute not directly relevant." This is wrong. It is very rel

"because >99% of computer users are non-programmers, making the freedoms to study, modify & redistribute not directly relevant." This is wrong. It is very relevant to users that their software be free, when they want to have it modified to fit their changing needs and when the corporation who sold it to them cannot and will not do it. (They will never do it, because either the corporation has folded since long, or it's too big to deal with small users and their petty specific needs). How many years are bugs lingering in widely used software: as long as they don't matter to a majority of users, they won't be dealt with. So, the situation where a would-be paying customer approaches a programmer to perform some correction or maintainance on some software, is not theorical at all, and in this situation the relevance of having freedom software is directly perceived by the customer and the programmer, since the later cannot work (at any acceptable cost, technically) on proprietary software, while it would be trivial on freedom software.