----boundary-LibPST-iamunique-502141743_-_-
Content-type: text/plain
AS Fyodor already mentioned, the goal is to "make it as simple as possible, but not simpler". That means, the simplest algorithm or tool _that_does_the_job_ should be used. This simplest algorithm or tool may in fact be very complex, and that's OK, as long as there is no algorith or tool that is simpler.
Let me give you a non-IT example. I'm a martial artist. If someone grabs my arm in a dangerous situation with bad intentions, in order to free my arm, I can do a Jiujitsu lever that takes 10 single movements, including steps, or I can do a lever that takes 5 single movements, or I can hit the attacker in the face with my free hand. All three options will free my arm. Obviously, the simplest is hitting the opponent. But there may be (legal) reasons that do not allow me to hit him in the face, so it might be better to do the 5 movement arm bar. It's not the simplest, but the simplest *for that situation*. Doing the 10 movement arm bar may be a total waste of time and effort in almost all circumstances.
Now to the other thing. You say the ifort Intel compiler is 50 times faster for loop tasks than BB. Are you talking about compile time or execution time?
Rene
Gesendet: Donnerstag, 22. August 2013 um 08:46 Uhr
Von: "Ivan Denisov"
[cleaned up version by Doug Danforth (native speaker of English)]
Thanks, Doug. It should have been my job, in theory.
> The BlackBox community can unify three things: 1. Style (maintain the
> minimalism of N.Wirth & A.Einstein). ....
> Minimalism works today but is becoming dangerous[??] and not every user
> of Blackbox enjoys this [who doesn't enjoy this?]. ....
"Excessive complexity is vulnerability" (the Kalashnikov Principle).
This is a fundamental law of Evolution.
The art of minimalism (= Wirth's "simplicity" or Guknecht's "purity") is not easy to master.
Ivan is not quite there yet.
The same is true about a vast number of "not everyones".
The OberonCore team, on the other hand, has been demonstrating competence amd reserve, and I have been relying on their bugfixing work for years.
And I think I can assume responsibility for fixing grammar in their English desctiprions of their bug fixes in future.
At least the bug fixing part should not be a concern from international perspective.
Details will be sorted out in the due course.
In my view, the Ominc announcement is mostly just words,
not much has changed in essence.
There is a number of big entities that could step in and leverage BB/Oberon to their advantage,
but they are not aware of the fact.
It is important to keep knocking on those various doors.
As an observer of Nature, I am looking forward to seeing how things develop :)
cheers
fyodor
Powered by the E-mail PIM - Info Select - www.miclog.com
----
To unsubscribe, send a message with body "SIGNOFF BLACKBOX" to LISTSERV{([at]})nowhere.xy
---- To unsubscribe, send a message with body "SIGNOFF BLACKBOX" to LISTSERV{([at]})nowhere.xy
---- To unsubscribe, send a message with body "SIGNOFF BLACKBOX" to LISTSERV{([at]})nowhere.xy
----boundary-LibPST-iamunique-502141743_-_-
Content-type: text/html
AS Fyodor already mentioned, the goal is to "make it as simple as possible, but not simpler". That means, the simplest algorithm or tool _that_does_the_job_ should be used. This simplest algorithm or tool may in fact be very complex, and that's OK, as long as there is no algorith or tool that is simpler.
Let me give you a non-IT example. I'm a martial artist. If someone grabs my arm in a dangerous situation with bad intentions, in order to free my arm, I can do a Jiujitsu lever that takes 10 single movements, including steps, or I can do a lever that takes 5 single movements, or I can hit the attacker in the face with my free hand. All three options will free my arm. Obviously, the simplest is hitting the opponent. But there may be (legal) reasons that do not allow me to hit him in the face, so it might be better to do the 5 movement arm bar. It's not the simplest, but the simplest *for that situation*. Doing the 10 movement arm bar may be a total waste of time and effort in almost all circumstances.
Now to the other thing. You say the ifort Intel compiler is 50 times faster for loop tasks than BB. Are you talking about compile time or execution time?
Rene
Gesendet: Donnerstag, 22. August 2013 um 08:46 Uhr
Von: "Ivan Denisov" <d.ivan.krsk{([at]})nowhere.xy An: BLACKBOX{([at]})nowhere.xy
Betreff: Re: [BLACKBOX] The official position of the Russian OberonCore team about BlackBox Support
Dear Fyodor and Douglas
Saying that "Minimalism works today but is becoming dangerous[??]", I mean that for many tasks we need to use not simple interfaces and difficult tools. I based in an interview of Wirth:
http://youtu.be/wrGytM2YTQY Even he must use very complicated and not well designed stuff in his work. And that is the only way in some cases. That I mean. The ifort Intel compiler is 50 times!!! faster than BlackBox in for loops tasks and 10 times faster than XDS Oberon Compiler with maximum optimisation. That can make sense then you make physics science.
Dear Fyodor
Could you be so kind to give a comment about BlackBox GUI Linux version sources (Draft but Working) and your plans on including them into the program of BlackBox development.