RoboZZle Forums

http://scratch.mit.edu/projects/axorion/1095996

Interesting! I'm not surprised the Scratch community is ignoring me. You try so hard to rise above the din just to cast your pearls before swine. Who cares about that peanut gallery but this site is different. You are my people, my peers, my cronies. I though someone would at least make a crack.

This is the story of my life. I'm smart, hard working, talented and yet so socially inept that I couldn't sell a life persevere to drowning millionaire. I guess I deserve to be unemployed.

axorion,

I played around with your project a bit, and here is my feedback. Hopefully it helps.


1. User's perspective
------------------------------
You really have to look at your project from the user's perspective. Why would they think it is cool?

It sounds like you want this to be a sort of "RoboZZle video recorder". OK. How do you make people think it is cool? One thing you could do is to generate a bunch of "videos" yourself (say based on the "Random paintings" forum thread), and make them nicely browsable.

I know I would at least play around with something like that. Try to think in that direction - what can you do so that people find your project interesting / cool / compelling? Think of a narrow scenario, and do it well. Then, add further scenarios depending on user interest.


2. Appearance
------------------------------
User experience and appearance of the app is *hugely* important. Definitely focus in that direction if you spend more time on the project.

Designing controls for RoboZZle is a lot trickier than it looks. For example, for the Silverlight client, I went through multiple concepts before deciding on the one that seemed reasonably good. Same thing for the iPhone client - we (mostly Bridger) spent a lot of time thinking about the intended user experience.


3. Expectations
------------------------------
One thing to be realistic about: "hardcore RoboZZle players" is not a huge demographic. Unless your project has appeal outside of this demographic (and it is important to be realistic about this), you probably won't get a huge exposure.

One reason why the RoboZZle players are "ignoring" you is that relatively few people visit the "General" forum. Probably only a couple of people even saw your link.


Conclusion
------------------------------
If you want some project ideas and/or feedback, drop me an email. My email username is igoros and I use gmail.


PS.: One other thing - please don't use the title bar copied from the game. People will probably find that confusing.

Igor:

Thanks for the input Igor, although I think you completely missed the point of my work.

As I stated above the intended user was never the Scratch member but rather the RoboZZle member. The project should be viewed from the perspective of training new RoboZZle users. It would have been linked to this site the same as a wiki page. Will the wiki help pages also have to remove any screen shots containing the title bar?

I won't disagree that the appearance and interface are inferior to the current silverlight client. But I've looked at your first draft with the blue title bar and I don't think I did that bad. Compared to your current java version or almost anything on Scratch I think it looks damn good. The edit and execution modes are both fully functional and have been briefly tested to be stable in the Scratch editor window, the Scratch presentation mode and the Firefox java client. I will consider expanding the bare minimum interface if the training mode ends up requiring extensive user input.

As for my expectations. If I succeed in teaching 20 people how to manipulate program stacks I will be happy. Since that goal has nothing at all to do with RoboZZle I have decided to continue to develop this project. If at some point its quality and value exceed your standards I would love your endorsement. Until that time the RoboZZle title bar will remain absent from my work.

Long live RoboZZle!

Axorion,

I didn't mean any of my feedback negatively, and I apologize if I offended you.

It is awesome that you are developing a project that builds on the RoboZZle ideas.

What are your ideas for further development of this project? Are you planning to add some "guided" training experience? Or do you want to do something else?

Igor

Igor,

I'm not offended. I just hated to pull the "advertising" but it's not my call. I'll get over it.

Yes I am already working on a [?] Help button that pulls up a list.

Something like...
1) Introduction
2) Interface
3) Movement
4) Conditionals
5) Functions
6) Recursion
7) Loops
8) Conditional Functions
9) Counting
[Press a number key]

Then it will pull up a predefined toy a walk the user through with pop ups. I need to code a single step feature as well. Right now it's just Fast, Slow and Reset.

I know you're a busy man but have you downloaded it yet? I'm sure you would have some insights if you knew more about how it's structured.

I wish I could show the Stack list while it's running but it bogs down so much that and the Stack(last) skips. This is equivalent to a stack pointer error and the robot goes haywire. It's a moot point because I can't control the visibility of lists from the scripts anyway. Beside the addresses are just cryptic numbers.

I've hard coded all the sprite location to make them persistent. It's a pain but it makes the project very stable even in the editor window.
Also you have to be careful to not call lists out of range. The development environment will catch it but the java will crash.

Too much info?

Axorion,

The first impression I had from viewing the java version was that you've implemented a subset of robozzle features, without anything new to add to it, so there wasn't really anything to trigger a positive reaction. I'm guessing others had the same response.

I see you've since started on a tutorial, which is great and I'd encourage you to continue with that.

It seems that you are spending some effort overcoming limitations of scratch, and I think this shows in things such as the program editing controls. They work, but they require too many mouse clicks to use, and I gather you have designed it like this because creating more buttons is harder/slower in scratch. Now implementing robozzle in scratch may be an interesting challenge, and if that's all you want then great, but I think if you want things like a tutorial to be more widely used then you should consider working with Igor to add features to the main robozzle site. I don't know how hard it is to collaborate on the silverlight client, but even taking the javascript client, fixing the usability issues with it and adding a tutorial would be good, and programming in javascript is good real world experience too.

(In my opinion, an improved javascript client would help increase robozzle's userbase too. Personally I'm only tolerating the silverlight because I got hooked playing the javascript version first.)

I chose Scratch because of its large user base and at first it was very easy to code. Plus the fact that many game 'spoofs' are posted here. Because Scratch is open source it is like a wiki page where anyone can add and edit. Once the basic engine is agreed upon any RoboZZle member could add there own lessons to the tutorial section.

As it turns out the Scratch editor gets very sluggish as the scripts get longer. Plus the 'spoof' was a bust. I would not be averse to programming in java or something else. I haven't done allot of java programming but it is close enough to C and I tend to acclimate quickly. Such a choice would of coarse completely disconnect me from the huge Scratch audience so for now it's not an option. I want people to learn and posting advanced puzzles is going nowhere towards that end. If I fail to get to the front page of Scratch within a month or two I will look at other options.

The interface has been moved to top priority. The fixed and rather low screen resolution in Scratch led me to a fewer but bigger control philosophy. It needs to be readable when played in the browser. Plus changing costumes with a click almost codes itself. After taking some measurements I'm certain I can squeeze a full button set in without using smaller buttons. This is a major change and will take some time but I'm sure it will be worth it.