KTurtle update
cies....@gmail.com
- the syntax (less logo'ish -- people where complaining)
- the featureset (focus on maintainability)
- the way to translate it (all by the translation framework)
- the way to add new commands (mostly from one ruby file that
generates c++ code)
- removed the need for highlighting schemes (no longer needed, syntax
highlighting is deduced from the interpreter's tokenizer itself)
- the editor is no longer based on kate (this made the previous point
possible)
- the whole thing is now migrated to the new Qt4 and KDE4 libs.
- testing modes (for the commandline) have been added
- URLs can be opened directly
- works on 64bit platforms
- ... (believe me i can go on)
All these changes are steps in the direction of a 1.0 release, that
will come (hopefully not too long) after the KDE4 release.
There is even an experiemental OSX port out there!
http://ranger.users.finkproject.org/kde/index.php/Home
(i had no time to try it, but if you can: please report your experience
here)
So why do i write all this?
Well.. i'd like to keep you all a bit up to date with what is happening
on the KTurtle front. And maybe there are people interested in helping
out in this stage. For instance:
- if you do kde4 develpment, can you try to build kturtle, run it,
play with it and share your experience? (don't expect everything to
work -- most things work though)
- you can have a look at the TODO file if you feel like C++ coding
- you can write some cool little app for it, or even cooler write an
app that crashes in a currently unknown way (see the BUGS file for know
ways to crash kturtle)
After KDE4 is released with KTurtle i will try to set up a little PR
campain for KTurtle. Im thinking:
- a little contest for KTurtle apps (the best get included in KTurtle
as examples)
- asking feedback from specialists on this field
- doing some interviews, pushing some articles, etc.
- getting others to talk, write, blog, etc. about kturtle
- push kturtle into some edutational institutes (or, as some prefer:
schools)
if you have any ideas on this little PR campain, please also make
yourself heard here...
thanks for reading,
cies breijs.
npettiaux
One point that I have found, and for with I have not found the time to
report - sorry - is that kturtle is not interactive as the "original"
dos LOGOWriter by LCSI that many primary teachers in Belgium have
learned, and that the syntax and, more importantly, the keywords are
not the same. Maybe the last piece could be arranged if I send you a
list of these keywords ? These facts have generated some reactions
around me when I have shown kturtle ... that I appreciare nonetheless.
Thanks,
Nicolas Pettiaux - nicolas.pettiaux (at) ael.be
cies
this is a common "request" for kturtle: 'please comply with the X
implementation of LOGO'.
i have decided never to comply with any logo, and even to make the gap
bigger to diffrentiate from LOGOs. since i dont want to endlessly try
to close the gap with all those many different LOGOs, there are too
many, too many versions, too many translations, too many features, too
many parameters.
so:
kturtle != logo
and will never try to be.
i'm sorry if that is a problem for you.
kturtle tries to be unique on its own, with it's own set of goals, and
it's own governing body (KDE).
regards,
_cies.
--
"Computer games don't affect kids; I mean if Pac-Man affected us as
kids, we'd all be running around in darkened rooms, munching magic
pills and listening to repetitive electronic music." -- Kristian
Wilson (Nintendo, Inc), 1989
Clem
there are 170 known implementations only about 10 are active and these
can be divided into a smaller number of closely related distros.
* Any user of MSWLogo or FMSLogo would be delighted if UCBLogo were
implemented (they being a superset)- as would a user of aUCB so that
takes in about 50% of the active distros.
* Others are offsprings of Comenius Logo.
* Another major player is Micromundo/Microworlds this is a
commercial product so can be discounted.
* Logographico is free and used in LatinAmerica- it differs from
UCB in the way it draws with a definable bitmap, but other than that
the syntax is identical, I believe that the drawing system could be
integrated into a vanilla UCB.
* X Logo (fr) is another contender but seems to be functionally
identical to UCB- it is internationalised into FR EN ES PT. The command
set (shortforms) are slightly different for example in French TG for
tournegauche, not GA but this is a matter of lookup table- and one can
live with either.
The second consideration is the supporting publications
* UCB has Brian Harveys trilogy-excellent, which explains why it is
the definitive version for most Interpreter writers,
* MSW has the Jim Muller Book
* LogoGraphico has IgnacioLuppi's excellent book in Spanish.
http://ccgi.frindsbury.force9.co.uk/greatlogoatlas/?Manual_Repository
Now we come to the question of how are these Logos different- and at a
quick glance they are wildly different, but on looking closer they all
implement the old LCSI core- without the hardware specific
instructions, and the user interface instructions. Logo users will be
95% happy with any logo that implements 95% of the core.
* KTurtle is annoying in that it tries to reinvent the wheel
unneccessarily.
* KTurtles strengths are that it is KDE, and has a really nice UI
and is internationalised.
There are four types of of Logo Users who rarely understand each others
needs.
* -hard core purist- pushing for the perfect logo- using logo to
unravel and illustrate the frontiers of computer science, the universe
and the meaning of life.
* -the developer/student who writes a interpreter as a programming
exercise either at university or for fun
* -the specialist who is using a modified logo as a research tool
in mathematical modelling or what ever (NetLogo)etc.
* -the user/school teacher who see logo as an essential tool for
motivating difficult kids. They want a standard product on every
computer that they can just use- one that works now when they are 60 in
the same way as it did when they were taught to use it when they were
20.
So now that I have convinced you that KTurtle should aspire to become
KLogo, what should we do next to make all these groups happy.
1. If you are working in Java- download and study the XLogo code,
see that it is GPL and cut and paste. When finished test for
compatibility with the code snippits from the help menu of FMSLogo.
When it works rename KTurtle to KLogo and everyone will be happy.
2. If you are working in C, then download the UCB code and talk with
David Constanzo of FMS, and Andreas Micheler of aUCB. When finished
test for compatibility with XLogo.
3. If you are in England, give me a ring and we test out the
compatibility of a few beers.
Clem Rutter.
cies
but i have decided not to make KTurtle LOGO.
for the next version of KTurtle, all links with LOGO are cut.
the only statement is "loosly based on LOGO"
KTurtle is not LOGO, it will never be.
i dont want to satisfy anyone that _needs_ a "LOGO"
KTurtle is a programming teaching tool, it uses an interpreter written
in C++ (i lately rewrote it), that makes it possible to translate all
the commands straight from the KDE translation framework. the rules
out many other interpreter designs where the commands have to be
hardcoded at compile-time.
i have had too many angry LOGO-purist people telling me i was misuing
the word LOGO since Kturtle doesnt do X.
net thing users want their source code of some logo to run on kturtle.
right no i choose my own path. an educational tool for teaching (kids)
programming and a bit of math. and from a maintainance point of view
i'm right.
believe me clem, i have had this discussion many times. and im happy
we right no have it on the mailing list, so i can refer to it. at some
point i even thought of making kturtle compiant with some LOGO, but i
gladly did not...
> So now that I have convinced you that KTurtle should aspire to become
> KLogo, what should we do next to make all these groups happy.
i know free-software land: users are never "all happy" ;)
more bluntly: im not convinced, and probably wont be convinced. (or
you have to put a lot of programming efford in kturtle and help me do
it by just doing it youself, or you have to pay me a lot)
> 1. If you are working in Java- download and study the XLogo code,
> see that it is GPL and cut and paste. When finished test for
> compatibility with the code snippits from the help menu of FMSLogo.
> When it works rename KTurtle to KLogo and everyone will be happy.
i will look at that code. sounds nice!
> 2. If you are working in C, then download the UCB code and talk with
> David Constanzo of FMS, and Andreas Micheler of aUCB. When finished
> test for compatibility with XLogo.
i looked at that code. i cant read C.
> 3. If you are in England, give me a ring and we test out the
> compatibility of a few beers.
this is a very nice invitation. i live in holland (rotterdam to be
exact). i would like to meet you, sure, but my time/money is limited
(last year of my Ba's).
kturlte builds on the power of KDE. tranlatable commands is the core
feature. LOGO-like syntax is no longer a feature nor a wishlish item.
again, sorry to displease you.
chears,
_c.