Gretl will not get the visibility with Hansl. Sure Hansl is great. One
can even argue that it is better than Stata, R, etc. but those already
have a huge share of the users, many developers, thousands of packages
etc. So this is like complaining everyone is using Windows (on the
desktop) while Linux is so great. It is the network effect, stupid!
Gretl has always had and still has the comparative advantage in the
area of easy to use, menu driven software complementary to R etc. As
can be seen from Eviews, SPSS... there is need for such programs and,
unlike programming languages, there are really no good and free
alternatives for such GUI based programs.
Unfortunately, gretl has not been doing well in improving its GUI
facilities. Sure there has been some work, most notably the tabbed
model output. However, there are many many missing features,
inconsistencies, an increasing feel of loss of intuitiveness due to
the lack of a general design direction. It is also looking
increasingly dated and convoluted. Moreover, the decision to unsupport
localized input was a big, big, big mistake. If a spreadsheet program
such as Excel, Calc, Gnumeric or Google Spread said they won't support
decimal comma inputs because that is not very useful and/or hard to
do, everyone would laugh. However, gretl decided somehow along the way
that such way would be acceptable for most users. Hah.
Anyway, if you want gretl to become more visible and popular, then you
have to see and remember its strengths, and where it needs to fit and
fill an important need in the world of scientific software. Those
areas are (1) a modern GUI based approach (supported by good
scripting) for doing modern econometric analysis, (2) excellent
localization and internationalization support, (3) a focus on teaching
econometrics at the both basic and advanced level.
Unless the above becomes a priority once again, the year of gretl is
not coming. No my friend. 'Nuff said.
Talha
On Mon, Jan 7, 2013 at 8:38 AM, Riccardo (Jack) Lucchetti
<r.lucchetti(a)univpm.it> wrote:
Folks,
allow me to link a short and interesting read:
http://seanjtaylor.com/post/39573264781/the-statistics-software-signal
In short, this guy characterises software users of several math/stat
packages. I just have a few observations, on which I'd like to hear as many
opinions as possible:
1) We're not on the list. As I've been saying many times over the years,
gretl being listed as one of the mainstream alternatives must be one of our
priorites and we need to think hard on how to achieve greater and greater
visibility. We've come a long way since 2009
(
http://ideas.repec.org/h/ehu/ehucha/01-03.html), but in the average guy's
percetption, we still are a minor player compared (say) to Stata. For
example, one of the thigs we need most badly is, IMO, the Hansl manual.
2) If we were on the list, what kind of comments would you expect to see?
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
_______________________________________________
Gretl-devel mailing list
Gretl-devel(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel
--
“Blessed are the young for they shall inherit the national debt.” –
Herbert C. Hoover (1874-1964)
--