To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.roboticsOpen lugnet.robotics in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / 25466
25465  |  25467
Subject: 
Re: Why [no single language] is good for Mindstorms
Newsgroups: 
lugnet.robotics
Date: 
Wed, 25 Jan 2006 18:27:56 GMT
Viewed: 
1779 times
  
On Tue, January 24, 2006 4:33 pm, John Hansen wrote:
In lugnet.robotics, Steve Hassenplug wrote:
On Mon, January 23, 2006 5:37 am, PeterBalch wrote:
Why couldn't the original RIS language interface have had a button to press
which converted your program to a textual form like a conventional
language? And this textual language had all the extra goodies that hackers
want?

The RIS v2.0 software actually DID save files in "Mindscript", which is a text
language.  I suspect they won't go that direction again, because it just wasn't
that popular.


TLG must provide software (firmware) that's "Bullet-Proof" for kids. (most of
their
users) • ...
I disagree.  The original software produced by LEGO was definitely far from
"bullet-proof".  Robolab is extremely far from "bullet-proof".

Sorry if I wasn't clear.  TLG must provide FIRMWARE that's "Bullet-Proof" for
kids.

It is a mind-set
and nothing more which says that kids need the simplicity and hand-holding that
a graphical programming environment provides.  First off, they don't.

I think you could ask anyone that uses a computer, and a VERY LARGE percentage
will tell you using the graphical environment of Windows or Mac is easier than a
text-based language like Linux.  Of course some will tell you they don't have an
icon for Linux?!?  And, yes, a very few will say they prefer the command prompt
to a mouse.

But, if you were going to make a single system for everyone, make it graphic.

And
secondly, neither RIS or Robolab is particularly simple, easy to learn, easy to
use, or very good at holding hands.

I'm not lonely enough that I look to software for hand-holding, which is good,
because I don't know of many pieces of software that are good at it.


Contrast learning to program using an unstable and buggy graphical environment
with learning to program using a full-featured text-based integrated development
environment with extensive learning aids, such as tutorials, sample programs,
context-sensitive help, code insight, syntax highlighting, code completion,
print preview, macro support, and many powerful debugging features.  There is
simply no comparison.

If you compare all the good points with one type of software to all the bad
points of another, you'll seldom get much of a comparison.  However, I don't
agree that either LEGO's RIS software or RoboLab is unstable.  In five years of
FLL coaching, I've never had any problems with them being unstable OR buggy.

In fact, over the years, with all the programs I've created, or watched other
people create, I've NEVER seen a typo in the graphic environments of RIS or
Robolab.

On the other hand, just slightly more (than never) is the number of times I've
seen a text-based program compile on the first attempt, with NO syntax errors.


The latter is far more beneficial to a child's learning experience than the
former.

I really want to agree here.  But I can't.  I'd have to ask exactly what this
child is trying to learn.

I think there is a HUGE percentage of people, both children AND adults, that buy
Mindstorms kits, who did NOT buy them because they want to learn to program.
THEY WANT TO BUILD ROBOTS.  And the software is just a tool to make the robot
work.

And with the former you get inefficient bloated programs while with the
latter you get highly-optimal programs that execute circles around the ones
written using the former.

Are you seriously suggesting an average 10 yr old would sit down with NQC and
write better code than the compiler would produce from Robolab?  I know many
adults that can't do that.  And, honestly, most of them don't care.

Assuming the logic is correct, their code could 100 lines long and takes 100ms
to execute, and few people will notice any difference between that and 50 lines
of code that took 50ms to execute.

The time the hardware takes is much more significant than the software.


And at the same time when you take the latter
approach you get a smooth, easy to follow migration path to even more
programming power via alternate firmwares and the various languages available
for them.

But many users are just not interested in following this path.  I've talked with
many people who are intimidated, or just not interested in learning a text-based
language.  They are happy with the graphical language.

I just CAN NOT agree with anyone who says ANY one language is good for everyone
and every application.

The NXT software will not be for everyone, any more than Windows is for
everyone.  However, I suspect a very large percentage of users will be fine.

Steve



Message has 1 Reply:
  Re: Why [no single language] is good for Mindstorms
 
(...) Linux isn't a language. It's an operating system. And nobody I know uses Linux without a graphical user interface. And nobody I know programs in text-based languages at a OS command prompt using VI. Do people write posts to Lugnet by dragging (...) (18 years ago, 25-Jan-06, to lugnet.robotics)

Message is in Reply To:
  Re: Why java is (not) bad for Mindstorms
 
(...) I disagree. The original software produced by LEGO was definitely far from "bullet-proof". Robolab is extremely far from "bullet-proof". It is a mind-set and nothing more which says that kids need the simplicity and hand-holding that a (...) (18 years ago, 24-Jan-06, to lugnet.robotics)

25 Messages in This Thread:










Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact

This Message and its Replies on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR