program synthesis can generate code based on a user's intent

the ExCAPE team was able to develop a method that formalized and standardized the core computational problem in emerging synthesis tools. Called Syntax-Guided Synthesis

Microsoft started out by developing domain-specific synthesizers such as FlashFill and FlashExtract, each of which uses examples to generate custom code that improves efficiency. FlashFill, which was released as a feature of Microsoft's Excel 2013, allows data entered into one column of a worksheet table to be entered in a new table column using only a few keystrokes.

 

From the article:

Using a model of programming called automated program synthesis, however, computers can generate pieces of code based on a user's intent, expressed using various non-code-based forms, such as examples, demonstrations or natural language commands...

By removing the need for would-be programmers to learn esoteric programming languages, the method has the potential to significantly expand the number of people engaged in programming in a variety of disciplines, from personalized education to robotics.

Oh, goodie, just what the world needs: more complete, dribbling idiots churning out garbage software!

Perfect!!

 

program synthesis can generate code based on a user's intent

the ExCAPE team was able to develop a method that formalized and standardized the core computational problem in emerging synthesis tools. Called Syntax-Guided Synthesis

Microsoft started out by developing domain-specific synthesizers such as FlashFill and FlashExtract, each of which uses examples to generate custom code that improves efficiency. FlashFill, which was released as a feature of Microsoft's Excel 2013, allows data entered into one column of a worksheet table to be entered in a new table column using only a few keystrokes.
View attachment 110619

What am I missing? Can't we do that already? If cell A5 is set to some number and cell C7 is "=A5" won't cell C7 always be equal to A5? Does Flash Fill do a range of cells at one time?

 

There is only one reason for pursuing this line of research. That is to allow high priced managers to replace high priced programmers when it comes to writing bad application software. Pffft!

 

hmm?, . . . ai-ai-yaa -- it is some step towards artifical intelligence - the machines can learn / go nuts a lot of faster than on a slow mode of mind operating men - everybody familiar with generic- and OOP must have come to this conclusion e.g. the "program" can be set to learn / evolve itself on it's own - there are mostly RAM VRAM and time limitations (to efficiently process complex relational data structures)

 

I would posit that the difference between a programmer and a manager is not simply that the programmer knows some arcane language, but rather the programmer knows what he really wants the thing to do. That is, if his program is to be successful.

How many questions have we seem where it takes several, even numerous posts to decipher what the poster's real intent is? I have a hard time imagining that any program can convert such nebulous thoughts into a working program that would satisfy such a manager.

John

 

. . . to decipher what the poster's real intent is? I have a hard time imagining that any program can convert such nebulous thoughts . . .

. . . actually i just glanced the sciencedaily - thought - these news 're worth to notice - i didn't expected there'll be a discussion about

i don't have a hard time imagining a "program" that can convert fragmented intent map to mach it's probable whole v.-s . . . it is all about what you are used to know as executable for todays human -- being todays human - if you were a super computer system or some advanced intellect - in a sense of data proccessing and availability (this's the volume of data visible to you at any time) -- the human and it's intents were a quite a simple figure for you to see and locate what it sees and where it's going or imagining it goes (you'd see human lives as quite automated and limited process chains ... that are quite in signifficant on their own compared against the whole environment and it's ecology during a notable change in latter)

 

hmm?, . . . ai-ai-yaa -- it is some step towards artifical intelligence - the machines can learn / go nuts a lot of faster than on a slow mode of mind operating men - everybody familiar with generic- and OOP must have come to this conclusion e.g. the "program" can be set to learn / evolve itself on it's own - there are mostly RAM VRAM and time limitations (to efficiently process complex relational data structures)

Re: '... it is some step towards artifical intelligence ..."
But does it work as advertised? Or does it only work in the specific circumstances of the sales pitch?