OpenEuphoria

1. The Eighth Passenger, Stoned goto, etc

• Posted by Shian_Lee Mar 18, 2015
• 1743 views

I guess, in general, that before implementing a "must have" feature, such as the The Eighth Passenger (eval), or Stoned goto (Try/Catch/Fall), or Time is money (Pass-By-Reference), or the III Millennium (Object Oriented Programming), or any other "must have", it worth considering the following:

1. Cost Vs Benefit. (i.e. time-effort-money-bugs-complications-broken_marriage Versus Is it that useful?).
2. Simple Alternatives. (i.e. who said that recursive-function is better then an efficient-simple-easy-to-debug while loop?, etc.)
3. How much need for it after all...? (i.e. so much work - but how many people and how many times will use it?)
4. The overall affect on the Euphoria language itself. (i.e. some features are almost offering a new language-dialect... a totally new concept... or simply throws to the garbage bin decades of hard work and effort to keep the language consistent, simple, fast and efficient).
5. Who, yes: Who, is going to deal with the extra bugs, documentation, tickets, cross-platform new release? (i.e. nobody likes to do these things, so: don't start something that nobody will finish. Not finished is equal to Not useful).
6. Is this "must have" feature is a "must have" feature? (no comment). (yes comment: the fact that nobody on this forum resist to this "must have" feature, does not mean that so many people actually need it or want it or will ever use it. Most users might need absolutely other "features", such as bug-fixes, more examples in documentation, tutorials or improving an already existing routines).
   
   

There is more...
One thing which is different between commercial / open project is Strategy and Obligation to it.
But one thing is obvious, both commercial and open project will benefit from being realistic and not greedy.

P.S. 1024/128/64-Bit support, UTF-8, and such, are more like Cross-Platform support, so these are obvious and I don't mention it.

3. Re: The Eighth Passenger, Stoned goto, etc

• Posted by petelomax Mar 18, 2015
• 1703 views

In general I agree. In my head (but no immediate plans to start work on it) there is now a clear, simple, elegant design for try/catch that should not trouble anyone who does not need it, whereas the discussion on string_eval() proved completely fruitless.

Pete

5. Re: The Eighth Passenger, Stoned goto, etc

• Posted by Shian_Lee Mar 18, 2015
• 1668 views

Thank you.

Just please remember Pete, if you throw an exception it might fly up above and reach the stratosphere, and if you throw it strong it might even reach the outer space where nobody can reach it.
i.e. throwing an exception is risky and backward.

Exception is not an exception until we define it as an exception.

About eval - soon or later it will be used incorrectly by someone careless:
the day after, some sadist writer from important web site will write in his/her front page:

That's what keeps people breathing. Sadism. (did I mention it before?...)

There are many ways to implement a macro language, or a limited set of commands which are more sane then eval, depends on what is the task. In most cases, I believe, it would be better for the programmer to create his/her own eval() specific routine - or in extreme cases: use the Euphoria interpreter.

7. Re: The Eighth Passenger, Stoned goto, etc

• Posted by petelomax Mar 18, 2015
• 1641 views

Fear not - that design in my head includes "if there is no catch, do exactly what it always used to do, that is display a message and create an ex.err".

Pete

9. Re: The Eighth Passenger, Stoned goto, etc

• Posted by Shian_Lee Mar 19, 2015
• 1640 views

All I can say is that - it's sad to see how all these talented people cannot truly cooperate, by defining a reasonable goals, and reasonable strategy, and reasonable duties.

I don't believe in anarchy (selfishness), or democracy (brutal-capitalism), or dictatorship (fake-god), or any other system.

Being reasonable and intuitive is enough.

But most people are too greedy and selfish to be reasonable and intuitive at the same time.

I believe that Open-Euphoria can reorganize the work and define more reasonable and stable development process, including refactoring/optimizing the current core and including stable/reasonable release cycle.

Like you, I wish that Euphoria will remain Euphoria. I hate to think that Euphoria will look like Python.

So I still have faith in Open Euphoria.