Closing an ANOVA window is now crashing gretl with the error message: *** glibc detected *** /opt/gretl/bin/gretl_x11: munmap_chunk(): invalid pointer: 0x08e9ee10 ***

I discovered a few more GUI bugs: 1)- When I enter, say c=1, the add new variable creates a scalar instead of a vector of 1s.

2)- When I choose "Define a matrix" X and build it numerically, gretl automatically opens up a new window showing this matrix. Now, if I choose "Define a matrix" again, nothing happens without first closing the previous window showing X. This only happens defining a new matrix just after having defined one.

3)- I think I might have reported this earlier: In the distribution grapghs, adding N(0,1) and then N(0, 0.8660254) results in clipping in the 2nd line.

4)- In the scalars window, I can keep left clicking the "Add" button and gretl keeps adding empty scalar lines.

>> I discovered a few more GUI bugs: >> 1)- When I enter, say c=1, the add new variable creates a scalar >> instead of a vector of 1s. > Not a bug: say "series c=1" if you want a series. Read the Help! I am aware of the series command for some time now. "Define new variable" should do what it is supposed to do: namely create variable items not scalar items. To add scalars, there is the "Add" icon. Check the Scalars window!

>> 4)- In the scalars window, I can keep left clicking the "Add" button >> and gretl keeps adding empty scalar lines. > Fine, don't keep clicking if you don't want empty scalar lines. Thanks for the workaround! Well, I will tell you about these things only if you want to hear about them.

All I was trying to say is that creating a constant series from the GUI is not currently possible due to this behavior.

The second one is a glitch that needs to be fixed if not immediately. The fix probably involves something relatively simple such as deactivating the add button when there is another adding in progress.

>> All I was trying to say is that creating a constant series from the >> GUI is not currently possible due to this behavior. > > Of course it is: type "series c = 1" into the box. Oh OK. Still, it is my humble opinion that it is only natural that entering c=1 in the "Define new variable" box should by default produce a variable item (a constant series that is) especially since typing "scalar c = 1" also produces a scalar (in addition to the easy add button in the scalars window). >> The second one is a glitch that needs to be fixed if not >> immediately. The fix probably involves something relatively >> simple such as deactivating the add button when there is another >> adding in progress. > > One might want to click "Add" (for scalars) 5 times, then put > values into the empty boxes. There's nothing wrong with this. I guess you are right, when you put it that way. Regards Talha -- “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) -- _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users

If I write c = 5 I would expect c to be defined by what is on the right hand side of the expression and this is a scalar.  5*const is a series and therefore c=5*constant defines c to be a (series).

> If I write c = 5 I would expect c to be defined by what is on the > right hand side of the expression and this is a scalar.  5*const is a > series and therefore c=5*constant defines c to be a (series). Thank you. This makes sense. But what also makes sense to me is that |Add|Define new variable| expects the user to enter a functional relation such as y=3+2x where x is a variable. So logic dictates (at least to me) that y=3 is a horizontal line, not a single point. I rest my case. Talha

Like Jack says this is an issue of perception. I don't know why I was thinking that way (maybe it is the gretl main window calling series variable while presenting scalars in a separate window) but now I am OK (really) with the idea that a variable is something that can vary such as a vector of values or a scalar (but not a matrix). I am sorry guys for keeping the list busy unnecessarily. I especially would like to apologize from Allin for my initial reaction. I didn't mean to annoy anybody. Regards Talha (Note: I am not being sarcastic.) On Tue, Dec 8, 2009 at 12:02 PM, Riccardo (Jack) Lucchetti <r.lucchetti(a)univpm.it&gt; wrote: > On Tue, 8 Dec 2009, Talha Yalta wrote: > >>> If I write c = 5 I would expect c to be defined by what is on the >>> right hand side of the expression and this is a scalar.  5*const is a >>> series and therefore c=5*constant defines c to be a (series). >> >> Thank you. This makes sense. But what also makes sense to me is that >> |Add|Define new variable| expects the user to enter a functional >> relation such as y=3+2x where x is a variable. So logic dictates (at >> least to me) that y=3 is a horizontal line, not a single point. >> I rest my case. >> Talha > > Hang on a second. As I said, I'm not particularly interested in arguing > intricate points on what is "best" in a GUI interface (it's simply something > I don't have a passion for, sorry). But this is way over the top. > > Talha, you say "|Add|Define new variable| expects the user to enter a > functional relation such as y=3+2x where x is a variable" as if this is > obvious. Well, it isn't. And don't try to convince me that it is, because it > isn't to me (hence it isn't, full stop). > > The point I'm trying to make here is that what we see as obvious is the > outcome of an infinite set of prior beliefs, preferences, a priori ideas > each of us has deeply ingrained in our minds. You should realise that maybe > other people have a different taste (so what is "clearly" better for you may > not be for someone else) and/or different priorities. On the present case: I > _personally_ think that the phrase "Define new variable" in the context of > an econometric package means "define a new container (short for 'a certain > subset of my RAM') and fill it with appropriate values". Maybe my mind works > this way because of my computer upbringing, maybe I'm just nuts, but I'd > never think of a "functional relation". Would you think of a "functional > relation" if you had encountered the phrase "Define new variable" in octave? > or in R? Why should gretl be different? Just because _you_ like to think > that the outcome of this operation is something that will eventually be > represented on the vertical axis of a Cartesian plane? Not enough to me, > sorry. > > (pause) > > Ok. > > (breathes) > > Sorry for the rant. I'm back in spectator mode. > > > Riccardo (Jack) Lucchetti > Dipartimento di Economia > Università Politecnica delle Marche > > r.lucchetti(a)univpm.it > http://www.econ.univpm.it/lucchetti > _______________________________________________ > Gretl-users mailing list > Gretl-users(a)lists.wfu.edu > http://lists.wfu.edu/mailman/listinfo/gretl-users > -- “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) -- _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users