Sunday, March 22, 2015

Digest for comp.programming.threads@googlegroups.com - 24 updates in 8 topics

Ramine <ramine@1.1>: Mar 21 07:55PM -0700

Hello,
 
 
Here is finally what i want to say...
 
 
When you start playing a probability of 1/1000000000 , it is
as you are playing for the FIRST time, so this is why some of my
previous reasonning are not correct, so then there is still a problem,
as i have explained , the 1/16 probability of the level 2 nodes of
the skiplist has really a problem, because if you start each
time to play a probability of 1/16, since 1-1/6 is enough big
this can degenerate the skiplist, and this is not good for
realtime systems etc.
 
 
 
Thank you for your time.
 
 
 
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Mar 21 09:05PM -0700

On 3/21/2015 7:55 PM, Ramine wrote:
> as i have explained , the 1/16 probability of the level 2 nodes of
> the skiplist has really a problem, because if you start each
> time to play a probability of 1/16, since 1-1/6 is enough big
 
I mean: 1-1/16 is enough big and this can degenerate the skiplist
 
bleachbot <bleachbot@httrack.com>: Mar 21 09:06PM +0100

bleachbot <bleachbot@httrack.com>: Mar 21 10:09PM +0100

bleachbot <bleachbot@httrack.com>: Mar 21 10:24PM +0100

bleachbot <bleachbot@httrack.com>: Mar 21 10:25PM +0100

bleachbot <bleachbot@httrack.com>: Mar 21 11:07PM +0100

bleachbot <bleachbot@httrack.com>: Mar 21 11:08PM +0100

bleachbot <bleachbot@httrack.com>: Mar 21 11:38PM +0100

bleachbot <bleachbot@httrack.com>: Mar 22 12:08AM +0100

bleachbot <bleachbot@httrack.com>: Mar 22 12:09AM +0100

bleachbot <bleachbot@httrack.com>: Mar 22 12:23AM +0100

bleachbot <bleachbot@httrack.com>: Mar 22 12:49AM +0100

bleachbot <bleachbot@httrack.com>: Mar 22 01:59AM +0100

Ramine <ramine@1.1>: Mar 21 07:14PM -0700

Hello...
 
 
If you have read my previous post titled: "there is still a big problem"
 
I think i have understood what is the problem with my reasonning...
 
I think that even if my previous reasoning appear a good reasonning,
there is still a "contradiction", because a probability 1/1000000000
that an event appear, means that the event is really improbable to
appear, so we can not contradict this logical fact, so what's the
problem then with my reasonning, i think that a probability of
1/1000000000 that an event appear is like saying that the time that it
takes to the event to appear is like "infinite"(in relation with
infinity), it is like an axes of real numbers, so if we plug this fact
in my previous reasonning this will make the time infinite so if we
start to play the probability of 1/1000000000 it is like if the previous
time that we were playing is probabilitic and infinite, this will
elevate the contradiction with my previous reasonning of my previous
reasonning. So we are safe !
 
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Mar 21 07:29PM -0700

On 3/21/2015 7:14 PM, Ramine wrote:
> in my previous reasonning this will make the time infinite so if we
> start to play the probability of 1/1000000000 it is like if the previous
> time that we were playing is probabilitic and infinite, this will
 
I mean probabilistic, not probabilitic.
 
 
Ramine <ramine@1.1>: Mar 21 06:44PM -0700

Hello...
 
 
There is still a big problem...
 
I wrote:
 
"I mean when you start playing each time the probability of
1/1000000000, you can not say if you are playing for a long time or a
short time, it is like if the variable of "time" doesn't exist, this
is what i mean, but the "probability" that the time is inside the long
interval of time of the probability of 1/1000000000 is maintained, so it
is like we are saying that my affirmation: "because the event can
appear at ANY MOMENT" is also very improbable."
 
 
I will make the affirmation that there is still a problem in the
fondation of probability..
 
Because when you start to play the probability of 1/1000000000
the space of the events will be reduced , because if we divide the
interval of the long time of the probability 1/1000000000 by 8 the
higher part that is the longuer times will appear with a probability of
1/8, because when we start playing the probability of 1/1000000000, the
probability that we have started with a long time is reduced because if
we consider the long times being the above 1/8 part of the of the
probability 1/1000000000 that is a long time, this higher part of the
long times can appear with a probability 1/8 , so if the higher part of
the long times can appear with a probability of 1/8 this will place us
much nearer to the event of the probability 1/1000000000, that means
this will reduce the significantly probability that the event will not
appear of 1-(1/1000000000), so what i want to say that even with a
probability of 1/1000000000 we are not safe ! i mean that even a
probability of 1/1000000000 or smaller is dangerous.
 
 
This is what i want to say...
 
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Mar 21 07:16PM -0700

Hello..
 
 
If you have read my previous post titled: "there is still a big problem"
 
I think i have understood what is the problem with my reasonning...
 
I think that even if my previous reasoning appear a good reasonning,
there is still a "contradiction", because a probability 1/1000000000
that an event appear, means that the event is really improbable to
appear, so we can not contradict this logical fact, so what's the
problem then with my reasonning, i think that a probability of
1/1000000000 that an event appear is like saying that the time that it
takes to the event to appear is like "infinite"(in relation with
infinity), it is like an axes of real numbers, so if we plug this fact
in my previous reasonning this will make the time infinite so if we
start to play the probability of 1/1000000000 it is like if the previous
time that we were playing is probabilitic and infinite, this will
elevate the contradiction with my previous reasonning of my previous
reasonning. So we are safe !
 
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Mar 21 05:32PM -0700

Hello...
 
I wrote:
> exist , because when you start playing it is also like you are
> playing for a long time or for a short time, so this can make this
> probability of 1/1000000000 to vanish at any moment and this can >
make the event to appear at any moment"
 
 
I think i have understood what is not correct in my reasoning...
 
 
The notion of time doesn't exist, but the "probability" that the time
is inside the long interval of time of the probability of 1/1000000000
is maintained, so it is like we are saying that my affirmation: "because
the event can appear at ANY MOMENT" is also very improbable.
 
 
So as you have noticed i am good also at reasonning and this makes the
fondation of probability safe !
 
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Mar 21 06:14PM -0700

Hello...
 
I want to be more precise:
 
I wrote:
>of 1/1000000000 is maintained, so it is like we are saying that my
>affirmation: "because the event can appear at ANY MOMENT" is also
> very improbable.
 
 
I mean when you start playing each time the probability of
1/1000000000, you can not say if you are playing for a long time or a
short time, it is like if the variable of "time" doesn't exist, this
is what i mean, but the "probability" that the time is inside the long
interval of time of the probability of 1/1000000000 is maintained, so it
is like we are saying that my affirmation: "because the event can
appear at ANY MOMENT" is also very improbable.
 
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Mar 21 06:12PM -0700

Hello,
 
I want to be more precise:
 
I wrote:
>of 1/1000000000 is maintained, so it is like we are saying that my
>affirmation: "because the event can appear at ANY MOMENT" is also
> very improbable.
 
 
I mean when you start playing each time the probability of
1/1000000000, you can not say if you are playing for a long time or a
short time, it is like if the variable of "time" doesn't exist, this
is what i mean, but the "probability" that the time is inside the long
interval of time of the probability of 1/1000000000 is maintained, so it
is like we are saying that my affirmation: "because the event can
appear at ANY MOMENT" is also very improbable.
 
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Mar 21 05:16PM -0700

Hello,
 
I have come to an interresting subject, i was thinking and thinking more..
 
As i said before , in mathematics of probability we say that if an event
have a probability 1/1000000000 to appear that means that the event is
very improbable to appear , but what i want to exaplin is that
this is not a proof that it will not appear, because the event
can appear at ANY MOMENT and this is what also i want to explain,
now if you respond to me that what i have just said: "because the event
can appear at ANY MOMENT" is also very improbable, you are making a
mistake, because you are including the variable of "time" inside
the equation, but when you start to play with the probability above
the notion of time is like it doesn't exist, because you will say
that a probability 1/1000000000 means that we have to wait a long
time for the event to appear, but what i am trying to explain is that
the notion of time do not exist , because when you start playing
it is also like you are playing for a long time or for a short time, so
this can make this probability of 1/1000000000 to vanish at any moment
and this can make the event to appear at any moment, so as you have
noticed that in this post i don't agree with the fondation that is the
interpretation of probability.
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Mar 21 05:30PM -0700

Hello,
 
I wrote:
> exist , because when you start playing it is also like you are
> playing for a long time or for a short time, so this can make this
> probability of 1/1000000000 to vanish at any moment and this can >
make the event to appear at any moment"
 
 
I think i have unerstood what is not correct in my reasoning...
 
 
The notion of time doesn't exist, but the "probability" that the time
is inside the long interval of time of the probability of 1/1000000000
is maintained, so it is like we are saying that my affirmation: "because
the event can appear at ANY MOMENT" is also very improbable.
 
 
So as you have noticed i am good also at reasonning and this makes the
fondation of probability safe !
 
 
 
 
Thank you,
Amine Moulay Ramdane.
 
 
 
On 3/21/2015 5:16 PM, Ramine wrote:
Ramine <ramine@1.1>: Mar 21 04:12PM -0700

Hello,
 
 
I was thinking more about Skiplists...
 
If you look at the the randomLevel functions:
 
randomLevel()
lvl := 1
-- random() that returns a random value in [0...1)
while random() < p and lvl < MaxLevel do
lvl := lvl + 1
return lvl
 
 
If p = 1 /4 that means that the nodes with 2 levels will have
a probability of (1/4)/ 4 = 1/16 to show and the the nodes with
3 levels will have a probability of (1/16)/4 = 1/64 to appear,
but i was thinking more about the the nodes with 2 levels that
have a probability of 1/16 to appear, a probability of 1/16
means that there is a probability of 15/16 to not appear, but
a probability of 15/16 is greater enough and this can make the system
to degenerates towards the "worst" case, so how can we guarranty
that the system will not degenarate towards the worst case with
the nodes with 2 levels or above ? this is not good for realtime systems
i think and this is not good if you are constructing your a bigger
in-memory database with your skiplist, your database will get too slow
and this is not good..
 
 
Other than that, in mathematics of probability we say that if an event
have a probability 1/1000000000 to appear that means that the event
is very improbable to appear , but what i want to say is that
this is not a proof that it will not appear, because the event
can appear at ANY MOMENT and this is what also i want to explain...
 
 
 
Thank you,
Amine Moulay Ramdane.
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to comp.programming.threads+unsubscribe@googlegroups.com.
Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)