| PLEX86 | ||
|
Where should the type information be 137
Where should the type information be 138 It wasn't that simple. The problem was that the status quo was badly places. In particular... Trevor,
This is a public policy issue not yet wrung out in court or the legislature. Presently, the EULAs all contain language where you the user accept all responsibility for the inaccurate answers that the compilers routinely manufacture BY INTENT. New law would resolve this situation, where as of some date in the not-too-distant future such words would EXCLUDE intentional weaknesses, e.g. to achieve performance. Microsoft already has a program of certification for the Windows logo, but they have so trashed the reputation of Windows that a program would seem to be aberrant if it did NOT sometimes crash ;-) The problem is that society doesn't now treat these issues in the same way that it now treats mechanics who turn brake rotors below minimum, who put antifreeze into break reservoirs because it is cheaper than brake fluid, pharmacists who dilute liquid drugs to improve their profits, etc. Further, I doubt that ours, the best government that money can buy, could ever make this essential leap in public policy. To REALLY make you wretch, you should have seen what happened during the approval process of ANSI C. Two other people and I attempted to get 19 horrendous holes in the C standard fixed. First the committee lied about them and hid them, then when proof that they had them was produced and ANSI was demanding that they confirm to procedure, they still refused to fix the bugs. Then, when the standard was released, they Shredded the documents relating to the problems with it that they did not fix! People have since been end by smooth-reading programs that did unexpected things. In short, there really is NO realistic limit to what "people" will do to maintain the status quo. I see no prospect of this improving until we make a good example of some of them, like locking them up. Steve Richfie1d
|
||||||||