[Date Prev][Date Next][Thread Prev][Thread Next]
- Subject: Re: Lua's opportunity
- From: Jimmie Houchin <jhouchin@...>
- Date: Thu, 28 Dec 2006 09:48:36 -0600
Peter Sommerfeld wrote:
Jimmie Houchin wrote:
And since Lua does support OOP, there is no reason to not do it
consistently and well.
Different OOP schemes have different trade offs. And a broad
range of application means that there are different needs. The
most you can expect is a more or less consistent usage of
some schemes / idioms in specific projects or aereas of application.
Ok. I was somewhat interpreting comments in other threads. But I have
experienced very consistent OO in Smalltalk. And to my Smalltalk eyes
inconsistent OO in Python.
However, certain critical mass is nice.
That has been passed already in some areas (let's say games,
embedding, data description) and others seem to be on a
good way (webservices for instance).
Well, the arena I am speaking of in general is a grab and use tool for
people who understand their problem domain but aren't necessarily
trained programmers. People that Python enables. I like Lua so much more
than Python that I would dearly love it to out of the box handle many of
the same situations.
I agree with you that Lua is still week on the desktop probably
due to a missing common lightweight GUI. But that is a difficult topic.
Lua Cheia has tried something but seem to have failed.
What I would love to see, and I haven't explored fully, is a Lua binding
to something like Cairo. Where you could simply define your own set of
widgets and go on. Yes, I can hear the its not a native UI argument. Oh
well. How many non-native UIs do we use commonly on a daily basis. Many.
What's desired is a good UI. Something like this would give us a "crazy
portable" UI with the "crazy portable" language. :) And that is power.
Things like QT, wxWidgets, etc. are nice. But straight connection to
something like Cairo, and a higher level Lua based set of widgets is
powerful stuff. And that kind of thing would be available faster to any
new platform than QT or wxWidgets, et al. could ever be. Lua/Cairo
better than Tcl/Tk. Or some such combo.
Well, good documentation and installation on a bunch of
different platforms is a never ending story. My experiance
so far is that it takes some time to become familiar with
the "Lua way" and to manage its power but than it is easy ... :-)
I am studying hard to learn the Lua way. :)
I absolutely love the functional side of things, but yet its very practical.
I think part of the beauty of Lua (the language) is that it is small and
simple. I have read PIL2 several times. When I started, my eyes glazed
over on closures, iterators, coroutines, but I kept turning the pages
and reading. Now on my third time through, closures and iterators
clicked. Yeah! Concepts I don't think I would learn in my 1500+ page
Programming in Python book. :)
It would be nice though, to have a good road map to acquire an
understanding of the Lua way.
Lua best practices.
Lua naming conventions.
Lua coding conventions.
I look forward to Lua Gems being published. I hope that will help
learning the Lua way.