Can prolog take on big data? like apache sark n flare and openMPI

[polymorph self]

cna you supercompute with prolog on 1000s nodes?

https://flink.apache.org/introduction.html

http://cassandra.apache.org/

https://flaredata.github.io/

https://arxiv.org/pdf/1703.08219.pdf

http://spark.apache.org/

http://www.hypertable.org/

https://www.open-mpi.org/faq/?category=general

google bigtable

[kint...@gmail.com]

If anybody suggests to you that you should build and deploy your own distributed-data-store using one of these products or something similar than that person is an idiot or an empire-builder, probably both .

There are billion companies out there offering this as a nework solution , you as a client treat the whole thing as a simple webservice , for example . Data in , query , data out . The competition is so fierce your probably not even paying them cost they use venture capital for that .

Do you want to learn about these solutions , become an expert , and become responsible for building and maintaining an operational network that utilizes them ? Suddenly your all-powerful big-data solution is 90% of the cost of doing business and is the reason your company goes bankrupt and now you have to go work in an office being a dork again .

~~ kintalken ~~

[kint...@gmail.com]

You know , your name kind of reminds me of TeX .
TeX can "polymorph self" as well , it is called ``\let`` .
It is definitely not called ``\catcode`` .

~~ kintalken ~~

[kint...@gmail.com]

> https://arxiv.org/pdf/1703.08219.pdf

Any paper with the terminology "best-of-breed" in the first pragraph is best forgotten , IMO .

~~ kintalken ~~

[polymorph self]

http://mesos.apache.org/documentation/latest/architecture/

[polymorph self]

Not sure what you are talking about.
See tools I listed.

[Dhu on Gate]

He's telling you to not bother cooking your own (poor) imitation of the
listed tools (and many more like them), in Prolog or anything else.

Dhu


--
Je suis Canadien. Ce n'est pas Francais ou Anglaise.
C'est une esp`ece de sauvage: ne obliviscaris, vix ea nostra voco;-)

http://babayaga.neotext.ca/PublicKeys/Duncan_Patton_a_Campbell_pubkey.txt

[polymorph self]

why not?
why not dominate the indsutry!! using free software!!!!!!
prolog better!!
logic can maybe even be used to show the most profitable alternatives!!

[kint...@gmail.com]

> why not?
> why not dominate the indsutry!! using free software!!!!!!
> prolog better!!
> logic can maybe even be used to show the most profitable alternatives!!

Well , that's exactly it .
You want to "dominate" an industry .
You want some grand and significant achievement .
Since you can't make it yourself , you are a sucker for the pitch .
The pitch is provided to you by manufacturers of grand and signficant achievements . It is the same pitch that the manufacturers use to obtain venture captital .

If you want to be a good computer programmer , then you learn how to do test-driven development , you spend lots of time making foolish mistakes , and you learn to see beauty in your small achievements .

And guess what you have to spend A LOT of time doing ?
Making computer programs .
Got any Forth examples for me ?

If you want to be a bad computer programmer , then you read industry rags like published by the Gartner group , you make flowcharts and UML diagrams with fancy tools , you talk bizspeak and you seek power and acclaim by associating yourself with what you think are grand and important new solutions .

You can turn that into a good paycheque .
It works like this :
Go work at some mediocre but big company .
Alienate the real computer programmers with your ineptitude .
Associate yourself with management .
Become an advisor to management .
After 5 years you can become an "architect" .
Now you will be able to put all you have learned from reading about all those cluster-fuck big-data solutions you have been reading about to very good use .
Help management make decisions about what technology is used for the next big solution .
Make all of those decisions without consulting the real computer programmers .
The real computer programmers are just workers , stupid technicians , they do what they are told .
Drown them in a waterfall .

> prolog better!!
> logic can maybe even be used to show the most profitable alternatives!!

Why is everything a war , a competition ?
"Java sucks - prolog is better !!!"

It is ridiculous to say that Java sucks .
Java is almost certainly the most successful software development platform in the history of humanity .

But it is not ridiculous to say that Java sucks if that helps you feel like you don't suck .

I don't think you suck .
Just remember that you are awesome .

~~ kintalken ~~

[Dhu on Gate]

I prefer functional languages over procedural. They make for better
system-level logic that's easier to debug. If I was going to look for a
way to bring semantic cohesion to a composition of some tools mentioned
previously I might look for a way to use prolog over java.

[kint...@gmail.com]

> I prefer functional languages over procedural. They make for better
> system-level logic that's easier to debug. If I was going to look for a
> way to bring semantic cohesion to a composition of some tools mentioned
> previously I might look for a way to use prolog over java.

Hi Dhu ,

What are the distinguishing features of prolog such that it is functional ?


~~ kintalken ~~

Je suis aussi ca . Je ne parlez paw Francais .

[polymorph self]

c Id say is more successful.....my good man!
oracle is cancer
www.hypertable.org what do you think of this?

[polymorph self]

It seems modern data is liek 100 computer network......and pengines seems to be able to make 1 supercomputer out of 100 seperate cmputers.....for mega speed.....:)

[Dhu on Gate]

Well, not sure who my audience is, but a functional language operates as
a theorem prover, and while procedural languages do this to, what they
prove is very limited and most of the program is limited to "side
effects". Being able to prove things about the semantics of the subject
matter buys you a whole lot in debugging too.

[polymorph self]

hmmm
big picture?

[Riguzzi Fabrizio]

Please check
http://www.swi-prolog.org/pack/list?p=mpi

[polymorph self]

woa what can you build with that?

[Dhu on Gate]

Prolog is not the solution, just a solution.

[polymorph self]

Well jeesh if I was that good Id be on island with 10 blondes.

I mean can prolog given someone good or me when I get good outdo these systems?
I mean java is cancer like all oracle.

[polymorph self]

On Wednesday, May 24, 2017 at 3:41:57 AM UTC-4, kint...@gmail.com wrote:

nah u suck
you dunce
I want to use better tek
I am fast becoming a REAL computer programmer and already am a REAL systems guy u fucking fag.
your arrogance is laughable
trump2020
you pussy

[polymorph self]

I wonder why not more lisp haskell forth mutli node databases?

[polymorph self]

On Thursday, June 1, 2017 at 7:54:35 AM UTC-4, Riguzzi Fabrizio wrote:

Interesting...any word of anyone using a prolog openmpi cluster to solve huge problems? with great N node performance?

[Dhu on Gate]

Complexity.

[polymorph self]

but java is cancer and does it

ignite.apache.org
spark.apache.org