Hello RJ,   
      
   On Wednesday May 22 2013 03:29, you wrote to me:   
      
    RC>>>   So, to be clear; you think that such functionality as that   
    RC>>> should be added and tested, before going to a version v3.4   
    RC>>> release?   
      
    MV>> Yes.   
      
    RC>    Thing is; of the three code patches since v3.3, two (v3.3.1 &   
    RC> v3.3.2) were changes that (thinking about it) really could or should   
    RC> have been minor version changes, rather than just patches against the   
    RC> v3.3.0 version, because they did involve functionality changes.   
      
   Aha. In that case I misunderstood your question. I read it as a call for what   
   needs to be added. If however the rule is that a change of functionality   
   should be signalled with a change in the minor version number, then that is   
   what should be done.   
      
    RC>    In other words; I think we should already be at least at v3.4, if   
    RC> not v3.5, because of the functionality changes that have been made but   
    RC> not yet represented by changes to the minor version number.   
      
   If you say so...   
      
    MV>> It is not a big change is it?   
      
    RC>    A good question, and I'm not sure yet if it would be or not...   
      
   What it involves is:   
      
   1) Define a boolean variable allow8bit.   
   2) Copy the code from for example "allowunpub" to couple changing   
      the variable to the the keyword, Allow8bit in this case.   
   3) Isolate the code that replaces characters with the highest bit   
      set with a question mark.   
   4) Skip it when the allow8bit variable is nonzero.   
      
   I'd say less than 15 minutes work for someone who has the make files   
   configured and who has familiarised himself with the source code. At least   
   that is what it would take me when I was still active as a programmer...   
      
   Contrary to fixing the "OS/2 problem". If that were easy to fix, it would have   
   been done already. But apparently, there never was a fully working OS/2   
   version. So it would take a programmer familiar with OS/2 to fix it and   
   considering that OS/2 is dead for all practical purposes, they are scarce..   
      
   I don't think we should hold the train for a very small minority who hangs on   
   to an OS that is on the way out. I loved OS/2, but the reality is that it is   
   doomed for extinction.   
      
    RC> I do think, though, that in any case it would involve a change to the   
    RC> minor version number when it gets implemented because it is a change   
    RC> in functionality.   
      
   Agree.   
      
      
   Cheers, Michiel   
      
   --- GoldED+/W32-MINGW 1.1.5-b20110320   
    * Origin: http://www.vlist.eu (2:280/5555)