On Mon, 24 Sep 2007, Talha Yalta wrote: Hi Talha, Could you please save a gretl session containing such a graph and send me the file? Am I right in thinking that iso-8859-9 is the standard non-utf8 encoding for Turkish text? Allin.

On Tue, 25 Sep 2007, Talha Yalta wrote: Thanks, but I'm afraid you seem to be out of luck. Gnuplot doesn't support iso-8859-9, and gnuplot's postscript driver doesn't support UTF-8 (in fact, I think postscript + UTF-8 is problematic in general). I messed around with your file for a while, but I couldn't get eps output with the glyphs you wanted, nor could I get correct PDF. Allin.

On Wed, 26 Sep 2007, Talha Yalta wrote: Yes. Gnuplot includes a large set of drivers or "terminals" for different plot formats. They are trying to standardize these so far as possible, but they are still far from identical. And in fact they can't all be identical, because the inherent feature-sets differ to some degree across the various target formats. In this case, the key difference is that the PNG formatter accepts utf-8 text while the PostScript one does not. My thought was to convert the text encoding from utf-8 to latin-9 before passing the information to gnuplot, but that doesn't help, since gnuplot doesn't support latin-9 encoding. Presumably that could be fixed without too much difficulty, within gnuplot. I'll take a look at that issue. No, it's not GPL'd, but you can distribute modified versions. The license is a bit quirky. "Modifications are to be distributed as patches to the released version." qtiplot would drag in the Qt libraries as a dependency for gretl, and I don't want to do that. Although gnuplot has some problems, I still favour it because * It is so flexible: you can save the plot commands and do all sorts of things with them, with the help of the gnuplot manual. * It's under active development, is being improved all the time, and is probably the closest thing to a "standard" for scientific plotting in the open-source world. Allin.

Thanks very much. For Linux, we have to wait for them to release the new version right? Cheers, Talha On 9/26/07, Allin Cottrell <cottrell(a)wfu.edu&gt; wrote: -- "Remember not only to say the right thing in the right place, but far more difficult still, to leave unsaid the wrong thing at the tempting moment." - Benjamin Franklin (1706-1790) --

Only a question: ¿Is there any way to recover the std.errors and confidence intervals produced by the fcasterr command? -- Ignacio Diaz-Emparanza DEPARTAMENTO DE ECONOMÍA APLICADA III (ECONOMETRÍA Y ESTADÍSTICA) UPV/EHU Avda. Lehendakari Aguirre, 83 | 48015 BILBAO T.: +34 946013732 | F.: +34 946013754 www.et.bs.ehu.es