Re: FrameTicks or LogPlot bug?

*To*: mathgroup at smc.vnet.net*Subject*: [mg102432] Re: FrameTicks or LogPlot bug?*From*: annetts729 <davidannetts at aapt.net.au>*Date*: Sun, 9 Aug 2009 06:05:33 -0400 (EDT)*References*: <h5ebgf$1p1$1@smc.vnet.net> <h5gs8g$eok$1@smc.vnet.net>

Hi AES, <snippage> > Is this what's intended? Is there some logical reason for this > seemingly inconsistent behavior? Is it somehow consistent with other > related commands in Mathematica? Or did the failure to document the > first case perhaps lead to the inconsistency going unnoticed. That's something else entirely .... >From my point of view Mathematica has lost an amount of consistency in many aspects of its operation. Indeed, it loses more and more with each version. Strings as function options is one example. Random{Integer| Real|...} is another glaring one, and you've no doubt many, many more of your own. Like shooting fish in a barrel .... For what it's worth, I think it's a bug that was unnoticed in the rush to get 6.x out. Yes, it is inconsistent in some sense, but to fix it now would break things, even if it's just my code, or yours. And yes, it is annoying when I use FrameTicks, that if I want to do it that way, which is most of the time in my case, I've got to remember that the syntax for that form is {{left, right}, {bottom, top}} rather than {{bottom, top}, {left, right}} which seems more logical. And I'm quiet sure that the developer would argue passionately about why it _has_ to be that way and no other and why what seems to a simple user to be logical, is not. Taking an optimistic view, it may be an unavoidable consequence of a dynamic evolving language/system, that 'clangers' like this and many more, get dropped by the developers. Regards, Dave.